The Accounting Quest of Steve Keen

Economics stands before a summit whose existence it is largely oblivious to. It can’t see the mountain because of a self-imposed fog. It sneers in the general direction of the summit, without really understanding what it is sneering at.

One great exception to this is the book:

‘Monetary Economics’ by Wynne Godley and Marc Lavoie

These authors have scaled the summit in question. They have demonstrated skills that are too rare in economic thinking – first, they are very knowledgeable in the detailed logic of financial and macro level accounting, and second, they understand the fundamental importance of accounting to economics. The idea of accounting “closure” resonates throughout the book. It is at its core the simple idea that balance sheets connect to each other through double-entry bookkeeping over space/time, with the facility of consistent income and flow of funds accounting as the required linkages for change. It acknowledges and uses the powerful logic that accounting is not just a rear view measurement device – it is also a constraint on all forward looking projections of economic outcomes – meaning that it is an important condition in the substance and shape of good economic analysis. If economics were to become fully congruent with real world observation, such a book would be viewed as ground breaking for the precision with which it weeds out some notable mainstream economic fallacies. And that might help pave the way for what should be a more cohesive fusion of accounting, finance, and economics as a connected and interdependent set of disciplines.

There is no short cut to head-on recognition of the importance of accounting to economics. There is no rationale for departing from the correct logic and method of double-entry bookkeeping as it exists – no reason to “spice it up” with creative but anomalous departures from what it actually is. Double-entry bookkeeping is not “a work in progress”. It is what it is. The neo-classical concepts of exogenous money and the money multiplier and loanable funds and ISLM and supply/demand equilibrium are part of the fog within which mainstream has constructed some economic imagery that is in fundamental conflict with the facts of accounting logic and real world financial measurement. This is all documented pretty cogently in ‘Monetary Economics’.

Another economist, Steve Keen, is attempting to scale the same summit from a different angle, although he writes reluctantly about having until recently “avoided the dry and dusty topic of double-entry bookkeeping”. Steve has written a post on quantitative easing, with an admirable bent towards the importance of accounting in that context:

Steve Keen: Is QE Quantitatively Irrelevant?

He says, “I’m open to correction that I’ve wrongly characterised what banks do or how QE works here.”

I think there is definitely a range of mischaracterizations in Keen’s post about how the banking system responds to QE in both operations and accounting.  So I’m going to offer some comments, as invited. While his construction departs from the actual case, I do have respect for the general direction of his pursuit.

As backdrop, there is a systematic tendency in the blogosphere and elsewhere to misrepresent the impact of QE in a particular way in terms of the related macroeconomic flow of funds. Keen has made this sort of error. Most descriptions will erroneously treat the macro flow as if banks were the original portfolio source of the bonds that are being sold to the Fed, obtaining reserves in exchange. This is not the case. A cursory scan of Fed flow of funds statistics will confirm that commercial banks are relatively small holders of bonds in their portfolios, especially Treasury bonds. The vast proportion of bonds that are sold to the Fed in QE originate from non-bank portfolios. The functional role of banking in this context is to act in effect as a broker of bonds between these portfolios and the Fed. Many descriptions of QE instead erroneously suggest the strong presence of a bank principal function in which bonds from bank portfolios are simply exchanged for reserves. In fact, for the most part, while the banking system has received reserve credit for bonds sold to the Fed, it has also passed on credits to the accounts of non-bank customers who have sold their bonds to the banks. This is integral to the overall QE flow of bonds.

It is also the case that many writers when explaining bank reserve operations have a propensity to overlook the banking system’s critical role as the original distributor of Treasury bonds to non-banks. Most bonds issued end up in the portfolios of non-banks, not banks. The QE function that is the subject here may be viewed as a “reverse distribution” version of that flow, or a gathering up of bonds that are sold to the Fed (although not back to Treasury) mostly from non-bank end holders. In both cases, commentators often overlook the end points of the flows that define the true macroeconomic distribution effect of such Treasury bond flows.

Given this acknowledgement of the full span of the QE bond transfer, it must accordingly be the case that the Fed’s activity in outright purchases of bonds in QE is most often associated with banking system reserve creation AND bank deposit liability creation, since for the most part it is non-bank portfolios that are ultimately affected by these bonds transactions.

Thus, it is incorrect to suggest as Keen does that QE has no effect on the broad money supply. But the effect that it does have occurs through a mechanism that has nothing to do with the erroneous textbook “multiplier” explanation of deposit expansion. Rather, it reflects the more or less immediate bank deposit liability impact that occurs in conjunction with QE reserve expansion. Because most of the bond flow into the Fed does not originate ultimately from bank held bond portfolios, it must be the case that reserve expansion is offset elsewhere on bank balance sheets. The first instance of such an offset is on the liability side of banking, simply because ultimate sellers of bonds are receiving money credits for those bonds.

This dynamic may not be so obvious in the observation of macro statistics during the financial crisis and the overall period of QE. That is because the bank deposit liability effect of reserve expansion during QE has occurred against a backdrop of general deleveraging of bank balance sheets. Thus, cumulative QE has not necessarily shown up at all times in equal amounts of additionally robust net growth in bank deposit liabilities. Loan repayments in the process of deleveraging and other balance sheet processes such as bank recapitalizations, have had the effect of reducing some bank deposit balances in a way that might not have been the case in the absence of a financial crisis.

In fact, the effect of QE has been to provide the banking system with a considerable replenishment of deposit liabilities that would otherwise have been lost or at least grown more slowly in the counterfactual case without QE. The counterfactual case in the midst of the financial crisis would likely have involved an initial vicious deleveraging of the size of the banking system balance sheet, where absent the injection of QE reserves on the asset side, there might well have been a more damaging outright contraction of deposits on the liability side.

When depicting QE in his computer program ‘Minsky’, Keen introduces a strange form of repo accounting on bank balance sheets. This accounting is incorrect. It simply does not exist in the real world – not as a direct feature of QE transactions in the way he has described. The most detailed dissection of such a complex accounting distortion would be more complicated than is warranted here. But here are some of the basic points concerning what he seems to have constructed:

First, as noted, most of the QE bonds are sourced ultimately from non-banks, so repos appearing as bank liabilities would not be an issue even if the accounting were correct, which it is not.

Second, the Fed purchases bonds outright in QE. There is no repo activity as a direct feature of such purchases. There is no repo commitment on the part of the Fed to sell bonds back to those who provided them to the Fed, and such repos do not show up in either the micro or macro level accounting for QE bond purchases by the Fed. In fact, the Fed will decide if and when it wishes to sell back its QE bonds to the private sector as a function of a QE “exit” strategy. Moreover, many of the bonds may end up being matured on the Fed’s balance sheet instead of being sold back. So it will be the Fed’s option to sell in the case of both the occurrence and the timing of any such exit transactions. There is no legal or economic obligation, as is the case with actual repo transactions. Thus, the repo entries that are presumed in Keen’s presentation are quite erroneous as a reflection of reality. And he has compounded this accounting error with respect to the repo entry on commercial bank balance sheets by first creating a repo liability as the offset to an increase in QE created reserves (which is wrong), and then forgetting that under his own assumptions there is first a decline in bank holdings of bonds due to the QE sale of bonds from bank trading inventories into the Fed, which requires a single accounting entry as an asset reduction in its own right. Triple entry accounting doesn’t work.

In fact the accounting and operations that correspond in a correct way to what actually happens in QE are very straightforward, involving the transfer of outright ownership of bonds, the creation of bank reserve credits, and in most cases the creation of deposit liability credits, as described above. To insert a convoluted imaginary repo sequence in the midst of this quite normal bookkeeping is simply to distort the case of actual operations and accounting treatment for QE.

(It is worthwhile noting that the operations of the Fed in a normal, non-QE environment include the outright purchase of bonds in conjunction with the secular growth in banknote liabilities on the Fed’s balance sheet. Commercial bank customers who go to their banks to get banknotes pay for them with commercial bank debits to their deposit accounts. Banks pay for their purchase of those banknotes from the Fed with central bank debits to their reserve accounts. This typically creates a shortage of reserves in the system (i.e. in pre-crisis, pre-QE monetary mode). The Fed, in order to control interest rate levels, re-injects reserves into the banking system by purchasing bonds. The end result is that assets and liabilities on the Fed’s balance sheet expand together – newly acquired bonds along with newly issued banknotes. Given that the location of most of the end-sellers of bonds in the case of banknote expansion is the same as it is for QE (non-bank portfolios), it is also the case that the Fed’s purchase of bonds in the case of banknote liability expansion tends to replenish the deposit liabilities that were extinguished as a result of the original purchase of banknotes by the public.)

In reference to the issue of repo in general, we should emphasize that repo is obviously an important institutional element in the macro flow of funds. But it is not central to the Fed’s balance sheet when viewed in quantitative (stock) terms – in either regular or QE monetary environments. In fact, Fed system repos – which provide funds to the system and which would be the natural location of any Fed repo activity directly associated with QE asset expansion, currently track at roughly $ zero, due to the massive QE reserve liquidity position of the banking system, and the corresponding absence of any material requirement for Fed repo activity (in quantitative stock terms) for purposes of fine tuning system reserve levels. The fact that the Fed now pays interest on outsized excess reserve balances that have been created by QE means that such repo activity is no longer a critical adjustment mechanism for excess reserve management as is the case in normal non-QE environments. But we should hasten to add that even in non-QE environments, the outstanding balance sheet stock of Fed repos is not a material presence when compared to the stock of outright held bonds and corresponding banknotes issued, as described above.

Third, we turn to what must now be noted as an accounting error of extreme proportions – which is that it is certainly not the case that the Fed draws on its equity account when it acquires assets. The Fed creates reserve liabilities as a result of the payments process that is used in the acquisition of assets. The phrase “loans create deposits”, which has become popularized in the endogenous money view of commercial banks, applies equally to the Fed in its own case of asset acquisition and reserve creation. The equity account is not touched in such a transaction, just as it is not touched when a commercial bank makes a loan and credits a deposit to the borrower’s account.

This type of accounting error is symptomatic of a misunderstanding of how an equity account works in the more universal context of double entry bookkeeping. The balance sheet equity account is where income accounting intersects with balance sheet and flow of funds accounting. At the margin, revenues and expenses on the income statement have incremental and decremental impacts on equity, respectively. These effects are summed up in periodic income accounting, resulting in a cumulative credit or debit to equity, depending on whether the result is a profit or loss for the period.

That said, the vast bulk of financial flows captured in flow of funds accounting (the purpose of which is to account for changes in the composition of balance sheets) has no direct effect on the equity account. Again, for example, the mere act of commercial bank lending and deposit creation, or Fed bond acquisition and reserve/deposit creation, is captured (as a flow) only in flow of funds accounting. This type of transaction does not touch the income statement directly and therefore does not touch the equity account directly. The blogosphere (including some economists), in seeking to understand the nature of macro flows, shows an occasional propensity to confuse these two basic, different accounting modes – income accounting and non-income flow of funds accounting. This is too lengthy a topic to pursue in more detail here, but some of us have pointed to this sort of error on a fairly regular basis in the past.

By way of contrast, the Fed does spend from equity (at the margin, via the income statement) when it pays for its own types of accounting expenses – salaries, purchase of regular goods and services in running its daily operations, and even payment of interest on reserves. These items are recorded as expenses on the income statement of the Fed.

And we should remember more generally of course that interest revenue and expense on the banking assets and liabilities that are first recorded in flow of funds and balance sheet accounting are then recorded as subsequent effects through income and equity accounting.

Steve Keen also notes his objective to establish a new definition for aggregate demand by equating it to income plus the change in debt. This entails embedded accounting confusion. Notwithstanding the evidence of impressive historic correlations and causal connections between changes in debt and economic outcomes, it is nevertheless incorrect to add income and flow of funds (i.e. a change in debt in this case) in any expression deemed to be an equation or an identity. Such an expression at best can serve as a regression function, in this case relating current period income to prior period income plus the change in debt. It includes accounting variables that are simply incompatible in an additive sense for an actual equation to hold, either during a single period of time or at a single point in continuous time. Income and changes in debt are orthogonal accounting measures. Flow of funds accounting where the balance sheet equity account is not involved (e.g. increases in debt) cannot intertwine indiscriminately with income accounting in any fashion that could be considered stock/flow consistent. Moreover, there is the problem that changes in debt may well be a source of finance used, not for spending as captured in NIPA, but for asset acquisition as captured only in the flow of funds accounts. This component of the use of funds in the economy has nothing directly to do with income accounting. Finally, the notion that there can be such a strict relationship in terms of income and the change in debt overlooks the fact that aggregate demand can fluctuate at times due to changes in money velocity, without necessarily involving changes in debt. In either case, it is potential and actual spending rather than a change in financial stocks or the utilization of existing stocks (i.e. money) that is inherent in the idea of aggregate demand.

I should note again that ‘Monetary Economics” by Godley and Lavoie is crystal clear and comprehensive in its appreciation of the differences between these two accounting modes, and of the importance of that distinction to the understanding of stock/flow consistent projections or scenarios for future economic activity.

There are other errors in how Keen combines his interpretation of repo accounting with some odd suggestions about equity account effects. The repo margin he anticipates (incorrectly in this case) could not in any case become an ex ante entry in the equity account. As a general rule, neither the Fed nor commercial banks are permitted to capitalize net interest margins attributable to their balance sheet positions in assets and liabilities. Interest margins are accrued to the income statement over time, and thence to the equity account (after allowing for other expenses) as they are earned. This again is another dimension of standard micro financial and macro level NIPA accounting, in which it is important to distinguish between marked-to-market asset values and what is known as accrual accounting for income and expenditures. Repos even when accounted for correctly earn an interest margin on an accrual basis. And even if repos were applicable in this case, accrual of earnings to banking equity actually decreases the money supply. It does not increase it. This is true of all equity accounting in the banking system. Money effectively converts from deposit liabilities to bank equity as bank profits are generated and credited to the equity account. This is the result of interest being charged to borrowers (who pay the interest from their deposit accounts) being greater than interest being paid to depositors.

Next, whether or not banks increase the money supply by expanding their balance sheets through the acquisition of equity securities is a rather random take on bank balance sheet management as it may be affected by QE. We should understand the broader truth of the basic (post Keynesian) insight that banks don’t lend reserves. Neither does the banking system transform reserves in any technical way into holdings of equity securities when acquiring such securities from non-banks. Both lending and securities purchases are flow of funds decisions that in themselves do not require system reserves, but rather are driven by risk assessment combined with bank capital allocation commensurate with that risk. The reserves are for payment purposes, and remain in the banking system when used to make payments. So, to the extent that bank reserves are a non-issue for lending, they are also a non-issue for the purchase of equities, with or without QE. Either of these activities can expand bank balance sheets, including deposit expansion on the liability side, and both require capital management in order to justify them as risk taking activities.

Moreover, in fact the U.S. commercial banking system’s holding of equity securities is minimal – less than $ 100 billion – and there is no evidence to suggest that QE has played any meaningful role in a decision for banks to expand their holding of equities in a material way. This makes eminent sense, in that equity securities are the riskiest financial asset class. Therefore, the fact that bank reserves play no role in either the analysis of risk or the allocation of required capital in making asset mix decisions holds a fortiori when comparing the case of equity security acquisition to bank lending. And, similar to the case for bonds, nearly all of the QE effect on the distribution of equity security holdings (in this case an indirect effect) will have been reflected in the distribution as it is observed mostly in non-bank portfolios.

Also, a repo account is in no way some sort of quasi-reserve position for bank lending, as seems to be suggested in Keen’s construction. One of the puzzling things about Keen’s model is that he seems to have built in a capacity to view banking system lending or buying equities as an alternative to holding reserves. This is very odd, in that it conflicts notably with the normal post Keynesian attitude to the functional role of reserves. The functional requirement for any form of risk taking through balance sheet expansion is capital, not reserves.

There are a few other minor and mostly optical points of error. Double-entry bookkeeping doesn’t record liabilities and equity as negative amounts. That particular sign orientation is used in an “adding up constraint” under a Godley/Lavoie matrix approach. In effect, the balance sheet is represented as a column vector of positive and negative numbers in such an approach, rather than two columns of positive numbers adjacent to each other, which is what is done in actual financial accounting. The purpose of the Godley/Lavoie matrix approach is to capture the fundamental accounting constraint whereby balance sheets must balance, in a way that translates to a vector form that sums to zero in such an “adding up” visualization. That is a convenient mathematical transformation of an actual double-entry bookkeeping visualization. Double-entry bookkeeping incorporates positive signs for assets and liabilities, so that assets equal liabilities plus equity. In particular, increases in equity (such as retained earnings) do not show up as a negative number in double-entry bookkeeping.

Finally, the Fed’s equity account in reality does not reflect “the value of its charter”. It is a standard financial account. It is the cumulative result of original capital injections and any retained income effects (which will be small due to remittance of Fed profits to Treasury). In fact, the equity account is where losses would be absorbed, should they occur, just as is the case with a commercial bank. That said, this “charter value” idea is a small point that is arguably not so important in the context of using a simplified model for the Fed balance sheet. Nevertheless, the factual operation of the equity account in the context of balance sheet, income and flow of funds accounting is an important general point of understanding.

I’ve already noted where I’d look for the highest standard of coherent accounting in the context of full macroeconomic modeling (Godley/Lavoie). Their book ‘Monetary Economics’ presents an integrated accounting and economic framework, featuring a rich supply of income and balance sheet simulations covering various types of economic behavior, where the outcomes are in full accordance with the actual accounting facts of the monetary system. I notice that Keen’s presentation includes something he calls the “Godley Matrix”. I tend to doubt that Godley would have employed the type of accounting that Keen is attempting to develop, an inference I come to naturally from reading the Godley/Lavoie book.

Steve did invite input regarding his method, and I’ve provided some here. More power to him in his pursuit of a robust post Keynesian oriented accounting emphasis, in the interest of advancing facts that need to prevail over neo-classical misconceptions that only obscure a factual view of the monetary system at work. But I suggest in the strongest way that the raw material to get such an effort on a solid footing already exists in an accounting framework that is well established in practice at both micro and macro levels.

Disclosure:

A common reaction of mainstream economics to the suggestion that it has stumbled in the case of solid macro accounting requirements for some of its theory is for it to become rather snooty about the role of “the accountant” as a professional activity separate from economics. This is the sneering I referred to earlier. Such a purportedly “sophisticated”, self-elevating view avoids the substantive issue of how in various identified cases economic theory is simply not compatible with real world macro accounting closure requirements. Again, I refer the reader to Godley/Lavoie’s ‘Monetary Economics’ in order to get a sense of the full scale and meaning of this phenomenon.

Like Steve Keen, I am not, and never have been, an accountant.

Given a normal aversion to the excessive use of metaphors, I’ve refrained from expanding on the image of “the summit” with such bells and whistles as death zones, frozen bodies, reckless tourists, “into thin air”, and wise Sherpas. That would require an accounting extension of its own kind – which admittedly is quite tempting.

 

Comments

  1. Cullen Roche says:

    Great post, JKH!

  2. Excellent, and very much appreciate the commentary on bank vs non-bank asset sales as an important detail often glossed over in discussions of QE. It seems to me that it is much easier to make the case that QE doesn’t impact aggregate demand in any fundamental sense when banks are the original sellers of bonds, given what we know from the ‘loans create deposits’ view of the world. And this is often the assumption made when discussing the impact of QE. The issue seems much more complex when we consider non-banks as the sellers of bonds. Scott Fullwiler and Cullen have broached this issue in the past (with respect to propensity to spend out of wealth, behavioral ramifications of having deposits vs bonds, portfolio shifts, etc.). I think Ramanan did a recent post on this as well. I’d be interested to hear your take, if you have more to add. I also feel like this is an issue where empirical analysis / behavioral finance might be able to provide useful information rather than hopelessly confounded analysis.

    • Cullen Roche says:

      The way I usually describe the non-bank seller situation is that it doesn’t change aggregate demand because it doesn’t alter any necessary component of the private sector’s balance sheet. If spending is a function of incomes relative to desired saving then QE via non-banks would have to alter some element of that equation. I always emphasize that QE is not enabling govt spending because that implies there’s a lack of demand for t-bonds in a non-QE world. We know that’s wrong because QE2 ended and yields fell. So QE can’t increase incomes in any real meaningful way aside from some of the residual effects of the portfolio rebalancing and the interest rate impact. QE has a strong psychological impact that potentially drives up asset prices, but I don’t think there’s strong evidence supporting the sustainability of such moves as the underlying fundamentals of the assets whose prices are in question appear to have no real transmission mechanism from which to be altered.

      I’m not in the camp that says QE does nothing, but I do think it’s a lot less impactful than most others. I think fiscal policy has probably done most of the heavy lifting in recent years and QE has helped with varying degrees in the different ways its been implemented (for instance, QE1 was much more supportive of the economy than QE2 & QE3).

      I’d love to see the USA in a weak credit environment, with a balance budget, a current account deficit and massive QE. Then we’d see where the rubber meets the road. My guess is that the economy would perform very poorly in such a situation assuming QE was impacted as it currently is being implemented…..

      • “I’m not in the camp that says QE does nothing”

        Agree with that. One can take the “banks don’t lend reserves” meme too far in the case of QE – this is not necessarily the right question or the only question. Beyond that, I’ve probably spent more time pondering the nature of the right questions than considering the best answers at this stage.

        Regarding aggregate demand, there is the wealth effect to consider, in addition to the income effect.

        More general comment below.

    • wh10,

      “It seems to me that it is much easier to make the case that QE doesn’t impact aggregate demand in any fundamental sense when banks are the original sellers of bonds, given what we know from the ‘loans create deposits’ view of the world.”

      But that is true even when non-banks sell the bonds to the Fed. It doesn’t *directly* impact aggregate demand even though there is an effect on the money supply aggregates such as M1 etc. This is because monetary aggregates are not causes of anything – more of residuals.

      There is of course a portfolio balance effect as Bernanke himself argues. In my opinion it has a strong effect on equity prices.

      (Actually there is one more thing I ignored in my post on asset allocation – which is the reflux mechanism where economic agents extinguish their liabilities from the proceeds of the sale

      Here is from the Fed FAQ:

      “When the Fed buys an asset, the effect on the broad money supply depends on who sold the assets and what they do with the funds they receive. If the seller is a bank, reserves go up, but broad money only increases if the bank responds to the increase in its reserves by lending more to households and businesses. If the seller is an investor other than a bank, reserves go up, and broad money also goes up in the first instance as the seller’s bank puts a sum equal to the amount it receives from the Fed into the seller’s bank account. But if the seller uses the money to pay down debt, the broad money supply declines again by the amount of the debt repayment. As of early 2011, the behavior of the broad money supply, economic activity and inflation all suggested that recent money growth had not been excessive.”

      http://www.newyorkfed.org/education/lsap/
      )

      which is okay except that it doesn’t clarify and somewhat indirectly and incorrectly and unknowingly promotes the view of the causality from money supply to inflation.

      I am however interested in everyone’s views including JKH.

      • Ramanan, thank you for that NY Fed clip. The destruction of deposit money through debt repayment — principal, interest, and fees — is the story of our time. It supports my contention that deficit-funded stimulus is plutonomic, as the base money created is eternal as an asset of commercial banks, while the deposit money created is ephemeral as an asset of the public, subject to destruction on any pretext. Thus a $35 overdraft fee is a permanent subtraction from broad money.

        JKH, thank you for a classic post.

      • Agree – see comment further below.

    • I would suggest that any effects from QE cannot be seen by looking at sellers of bonds.

      They were selling anyway and therefore their behaviour has unlikely to have changed. Possibly there may be shifts in timing to gain advantage of the increased demand.

      I would suggest that the effects of QE would come from the potential buyers of bonds who were outbid. What did they do instead?

      QE increases the number of people bidding for Treasuries, because it adds a load of reserves to the primary banks – as offsets to the increase in liabilities to the secondary institutions. The primary banks will then try and swap them for Treasury bonds. You’ll notice that the balance sheets of the primary banks have more Treasuries in them than they did prior to QE. So you get even more buying pressure on Treasuries.

      In the UK we have a lot of the primary banks as the market makers at the Gilt Auctions. So it’s fairly easy for them to push their own book and bid up the Gilts.

      All the QE documentation seems to focus on sellers of bonds, and I don’t understand why.

      • “You’ll notice that the balance sheets of the primary banks have more Treasuries in them than they did prior to QE.”

        I did notice that and it’s an interesting point. It’s a gradual ramp up but it’s not huge as part of the balance sheet mix (much smaller than the reserve increase), and it’s leveled off.

        It underlines the point that the commercial banks haven’t been the net portfolio source of QE bonds – quite the contrary. My guess on that is that with the system on the mend in terms of internally generated capital, and with the Fed steadily telegraphing low rates, they became a little more adventurous in taking some positions along the yield curve. Those positions do require capital to support interest rate risk, but that’s less than for credit risk.

      • What are the accounting entries when banks purchase bonds from entities other than the Fed?

    • “And this is often the assumption made when discussing the impact of QE.”

      Right – see my comment to Cullen above and then further below.

  3. Dan Kervick says:

    Good post JKH. One comment about the issue wh10 highlights:

    Just to (over)simplify matters, let’s leave out the bank intermediary altogether and imagine that the Fed purchases a bond directly from the portfolio of company A, which happens to bank at Bank X; that the bond was issued by company B, which happens to bank at Bank Y; and that neither A nor B are themselves banks. So yes, the purchase results in both the creation of a new deposit balance held by A at Bank X, and a new reserve balance held by Bank X at the Fed.

    But the bond was an asset of A that A no longer has. Had it not been purchased, it would have resulted in a flow of money over time from B’s deposit account at Y to A’s deposit account at X (along with a corresponding settlement between Y and X). But now that the Fed owns the bond, as the required payments are made by B, a portion of B’s deposit balance at Y is just extinguished as payments go from B to the Fed, settled via Y’s reserve account. There is no corresponding addition to some other firms deposit balance.

    So while the QE purchase does result in the net creation of non-bank deposit balances in the immediate term, it also results in a net extinguishing of non-bank balances over some longer term. This could conceivably be useful, if the firm portfolios in question are in need of liquidity; but not so useful in the aggregate if non-bank firms are already in possession of much cash and other highly liquid assets, which I believe they are.

    • Dan,

      Yes, in fact the Fed has a huge amount of MBSs in its portfolio and as and when households pay their interest and principal, the MBSs amortize (typically monthly).

      The Fed has however over some times let this happen and in other QE versions reinvested the proceeds and purchased more MBSs.

      Plus yes over time, it will just let them amortize/expire.

    • Dan

      You seem to be including a private sector bond in the mix, which complicates it a bit.

      I think we have to be careful when relating interest payments of any sort to money supply.

      E.g.

      The payment of interest from one private sector non-bank agent to another does not increase the money supply.

      As noted in the post, the net interest margin accrual of banks actually reduces money supply.

      Given the Treasury practice of maintaining some degree of stability in the TGA account, the payment of interest by Treasury on bonds held by the private sector does not increase the money supply. At the margin, deficit financing pays for that interest by recycling the existing money supply.

      In the case of the Fed, if it is running at an overall positive interest margin and profit, that actually reduces bank reserves in itself, analogous to the commercial banking effect on deposits. That’s because Treasury is paying interest to the Fed from TGA and in effect sourcing the funds to do that via deficit financing, which draws down bank reserves. At the same time, the interest it credits back to the banks on reserves is less than what it is earning on assets, so that the net effect is a debit to bank reserves. The offset to that is an increase in Fed equity capital. But then that effect itself gets reversed back out when the Fed pays its profit over to Treasury, which again ends up recycling money supply back to the private sector. So it’s all a wash in terms of the net reserve effect, post Fed profit remittance.

      In theory, it takes a bank to expand the money supply by paying interest. But in the case of a commercial bank, there are accounting rules that prevent it from loaning too much in the way of interest due before the loan is declared non-performing. Similar rules would apply in the case of the Fed in terms of any risky assets it held (e.g. early crisis assets) in which interest was overdue.

      I’m always amused by those (other) macro Marx-type arguments about the capitalist monetary system being unable to pay interest at the end of the day. The simple explanation lies in the bank equity effect.

      E.g. assume a stationary state system with no growth. Banks earn profits and their equity accounts expand, which drains money supply.

      Answer:

      The system is stationary state with no growth, so banks don’t need to grow capital by retaining earnings. So they pay it all out as a dividend. That replenishes deposit levels. And all interest has been paid. And the growth case is easy from there.

      • Dan Kervick says:

        JKH, I introduced the private sector security because I thought that was implicit in the point you had raised, and wh10 followed up on, about the fact that QE purchases assets from non-bank portfolios.

        But my point is really that no matter who the issuer of the security is, the QE purchase removes from the private economy over time whatever principle and interest payments are attached to that security, at the same time as it injects some other amount of money in the immediate term. The asset represents the fact that someone in the private sector, somewhere, is financially bound to pay some defined amount of money to the holder of the asset, on some defined schedule. After the Fed purchases the security, that money gets paid to the Fed and hence extracted from the private sector.

        These asset purchases are conducted via auction, so I assume that the purchase price is not all that different from what would have been the purchase price if the security had been sold in private markets without central bank intervention. (We’re talking about QE, not TARP.) On the other hand, by entering the market as a big new customer, that should chase up the prices of the assets at least somewhat.

        I’m not all that interested in the money supply, since I think that’s a concept (or familiy of concepts) that is far less significant than much of the discussion of central bank policy seems to assume.

        • Cullen Roche says:

          QE doesn’t necessarily have to reduce the income to the private sector. If we describe taxation and govt spending as a sequence of flows or redistribution (as MR does, as opposed to “destruction” and “creation” as MMT does) then we can envision an environment in which QE could occur AND flows from taxation to spending actually remain the same. In other words, if the govt doesn’t reduce its deficit or actually increased it during QE then there is really no change in the amount of income that is being redistributed from one agent to another. Personally, I think it’s better to describe taxation as redistribution and not destruction as it keeps the flows of income in the economy in the proper perspective. I also think it avoids the unnecessary reserve accounting tricks that MMT tends to fall victim to.

  4. The neo-classical concepts of exogenous money and the money multiplier and loanable funds and ISLM and supply/demand equilibrium are part of the fog within which mainstream has constructed some economic imagery that is in fundamental conflict with the facts of accounting logic and real world financial measurement.

    Excellent post JKH. What a nice way (above quote) to summarize the profession. And the fact that the fog doesn’t work leads economists to add more and more fog to their models.

  5. here is no repo commitment on the part of the Fed to sell bonds back to those who provided them to the Fed, and such repos do not show up in either the micro or macro level accounting for QE bond purchases by the Fed

    Correct, but there is a ‘repo expectation’ which I think is what Steve was trying to get into the initial version of the model.

    Everybody talks about the QE ‘exit’ and therefore that must have an effect of the forming of expectations amongst those who see things differently than we do (which I suggest is that an exit is unlikely to occur other than by a run off of the bonds to Treasury on maturity, ie we can assume they will have no further effect a the currency issuer/currency user interface).

    In particular, increases in equity (such as retained earnings) do not show up as a negative number in double-entry bookkeeping.

    They do in many underlying computer models of accounting, but that is a simplification to make verification easy. A simple SUM across the database then makes sure the system is working properly.

    I don’t see a problem with seeing liabilities and equity as negative assets as long as you don’t start believing that there are less than zero of them around.

    A better analogy is probably matter and anti-matter.

    they understand the fundamental importance of accounting to economics

    It provides a rigour that is necessary to get the models to function and also so they can be independently checked against the actual outcome evidence.

    There have been some terrible models over the weekend that simply don’t follow the correct accounting, miss the difference between the primary banks and the secondary financial institutions or try and suggest that Central bank OMO bond issues are somehow more powerful than Treasury bonds despite them both causing the same reserve drain.

    This post has saved me a job refuting those. Thanks.

    • Neil,

      The point I tried to make on sign orientation is that there is a difference in the visualization of actual double-entry bookkeeping as it appears in standard financial statements and their macro equivalents, versus the visualization as it is transformed for purposes of analytical manipulation in such constructs as the Godley/Lavoie balance sheet and transaction matrices. In the case of the former, the balance sheet is in essence an (‘r rows’ x 2 columns) matrix of positive entries while the latter is a single column of entries that sum to zero, embedded within what is a larger cross-agent dimension matrix.

      This is important to the degree that Steve (and others) when getting into the area of providing guidance on the facts of double-entry bookkeeping may end up confusing those who aren’t yet far enough up the learning curve to understand those kinds of differences.

    • Cullen Roche says:

      I actually think it’s a pretty crucial error to describe this as a repo expectation. I am not sure what Keen was trying to do there, but there is absolutely no reason for the pvt sector to have to buy the bonds back at some point. If the Fed decides to hold these bonds on its balance sheet until they mature then that’s the grand “exit” strategy. The bonds will simply roll off slowly and the Fed’s balance sheet will naturally shrink.

      The majority of people out there seem to think the Fed has to “unwind” QE at some point and the fear mongering people on certain websites and representing certain political views like to portray this as some sort of world ending moment. It misunderstands the entire operation and really shouldn’t be described as a repo of any type. Even if you consolidate the Fed and Tsy (which really shouldn’t be done in the first place) you still have separate issuance of new t-bonds at time of auction (assuming that future issuance is somehow a repo, which it isn’t).

  6. Neil,

    “I would suggest that any effects from QE cannot be seen by looking at sellers of bonds.

    They were selling anyway and therefore their behaviour has unlikely to have changed. Possibly there may be shifts in timing to gain advantage of the increased demand.”

    It is hard to beat that logic but think about it: if the Fed starts buying equities directly, won’t equity prices rise? It may all end up badly – bubble and bust but the Fed purchasing equities will initially lead to rise in equity prices right?

    • The price of the bonds will rise. I don’t doubt that. What I doubt is whether that causes the rush for the exit in bonds that the authorities seem to believe will happen via ‘portfolio reconfiguration’.

      I would suggest that the holders of bonds are after the income and guaranteed capital repayment from the bonds and that if they switch out of bonds it will be to bank deposits – but only if their projected income needs can be fulfilled over the period of the bond. However bank deposits in the current market are riskier than bonds – as large deposits are not insured. Would you effectively invest in a bank in the current climate with ‘bail-in’ on the table all over the world?

      More likely investors will hold the bonds at their higher value providing more collateral for other activities.

      So I’m pushing the effect largely (but not exclusively) from the buying side – since they’re the ones that want the income and won’t be able to get it from bonds in aggregate to the extent that they were before. So where has the saved money gone instead?

      As ever in these cases this is pure speculation on my part based on belief, and it would be useful to get some good aggregate empirical data on what people have done throughout these QE periods.

      • Agree. I think you are saying that it is possible there is no “rebalance” after selling bonds to the Fed which is entirely possible.

  7. Art Patten says:

    Godley-Lavoie’s ME is a great book, but I think the authors would deny having ‘summited’. More like they’ve set up most of the caches and camps required to get there?

    And kudos to Keene for his self-critical candor. If only better-known economists (e.g., Krugman) were equally capable, rather than torturing IS-LM without regard for Hicks’ own alleged disavowal. (If anyone can link to documentation, thanks, I can’t seem to find it at the moment.) But that would mean having to rewrite their macro textbooks, of course.

    • http://www.jstor.org/stable/4537583

      IS-LM: An Explanation by John Hicks.

    • I think they planted the flag in terms of establishing the unambiguous importance of the accounting connection. That was the summit in question.

      No doubt there’s a higher mountain to resolve other issues. But first things first.

      (Spoils my metaphor situation a bit in the sense of identifying that higher mountain, but I can live with that for the time being.)

      • Art Patten says:

        Just append “range” to “mountain” and I think we’re all set. :)

      • beowulf says:

        If I remember the Book of Exodus correctly, its actually the valley visible from the summit which is the Promised Land.

  8. All:

    I would decompose QE into the following questions:

    a) What’s happening to the flow of funds
    b) What’s happening to interest rates and asset prices
    c) What’s happening to aggregate demand

    Regarding the first, the post attempts to highlight the topography with respect to the bank/non-bank distinction. The first order flow of funds effect is as described, with CB expansion of bonds and reserves, commercial bank expansion of reserves and deposits, and non-bank replacement of bonds with deposits.

    Before getting to second order flow of funds effects, I think you have to look at what’s happening to interest rates and asset prices.

    And I think the most impressive thing I’ve read on that subject is Woodford’s big paper from last year. If you ignore some of the finer details where there is some neo-classical noise in that paper, I think he got the interest rate transmission about right.

    I view it as follows. The Fed is signalling its determination to keep the path of short term interest rates low by buying bonds that typically incorporate an element of expectation for that short rate path in their long term yields. Thus, the Fed is signalling its own expectation for its own policy. Moreover, the fact that it is buying bonds means it has skin in the game. It is making a bet on interest rates – a bet that short rates will remain low because it will keep them low– a bet that it alone has the power to make come true – if it dares. That’s pretty big skin. The Fed in conventional banking terms is running an enormous “interest rate gap” – long duration assets and short duration liabilities – the kind of thing that took down the savings and loans several decades ago. (Although the Fed still has a $ 1 trillion partial interest margin cushion of banknotes that will remain zero interest cost when the policy rate rises.) Notwithstanding whatever one might say about the inherent economic ability of the Fed to absorb losses in a case where it ends up deciding to raise short rates sooner rather than later (i.e. the consolidated interpretation of the Fed and Treasury, etc.), they’re not going to want to be seriously wrong on this bet – i.e. they’re not going to wish for an outcome that requires dramatic interest rate tightening early on that will cause abrupt losses on their books. They’re going to want a gradual increase in interest rates over time, back to some semblance of normalcy, such that they can absorb the net interest margin cost effect smoothly with a combination of holding bonds to maturity and periodic, opportunistic selling into short term bond market rallies. The consolidated view of the Treasury/Fed books is fine for purposes of economic substance – but there will definitely be political pressure if things turn nasty in this other way.

    Regarding second order effects in terms of the flow of funds, I think the general interest rate effect is probably a lot more powerful than the particular actions of who is selling the bonds. The interest rate effect spreads over everybody – not just sellers of QE bonds. In the case of domestic institutional investors such as pension funds or mutual funds, it is likely that the interest rate expectation transmission may cause them to rethink their overall cost of capital and asset allocation models. In this way, the interest rate effect may ripple into repricing of risk and riskier opportunities – provided these institutions are comfortable with that risk and/or capital allocation. Some of this may well occur in the portfolios of those who sell the bonds, but there’s no reason why it wouldn’t also happen across the board including those who don’t sell bonds but are simply managing their risk and capital in any event. And the fact that Japan and China may well be the original source of some of the actual bonds that are sold into the Fed should illustrate the likelihood that the transmission doesn’t focus only those who sell the actual QE bonds, since those particular sales may have very little to do with the general repricing of risk by institutions operating in the domestic US financial markets.

    Aggregate demand effects? I’m open on that. It would involve the wealth effect of course – but that shouldn’t be totally discounted.

  9. Does the fed ever buy bonds directly from entities that are not member banks?

    • Cullen Roche says:

      No, and I think it matters more than some people think. If the Fed started buying plots of dirt from me then that would be the Fed doing fiscal. But the Fed doesn’t do that. They buy govt guaranteed assets that have already been issued and swap them out for reserves/deposits. The banks are the Fed’s “dealers” in these transactions and the Fed only deals in govt guaranteed assets. It’s important to understand in that the Fed could potentially alter its course as Scott Sumner has advocated and start buying anything and everything. What Sumner doesn’t understand is that this is the Fed doing fiscal so it’s kind of ironic that he frequently mocks fiscal when his big bazooka is actually the Fed doing fiscal….

      • Art Patten says:

        Not always the case: http://www.newyorkfed.org/markets/maidenlane.html

        But the question was whether the Fed transacts with non-member banks. It depends on what market they’re participating in, I belive. But once we assume a reasonable degree of arbitrage, how much does that matter, at least for open-market operations?

        • Cullen Roche says:

          I’d have to look up the details, but I am pretty sure Maiden Lane was funded with a bridge loan from the Fed. The Fed was authorized to do so using the entity as a lender of last resort. It’s different from QE as it’s presently being implemented, but does show the extraordinary powers the Fed has in circumventing the govt to prop up the banks directly.

          • Art Patten says:

            It was ‘financed’ (there was also a line of credit from JP Morgan), but that arrangement was more form (political) than substance, don’t you think? (Like a dirty repo that turned into a pretty nice investment runoff?)

            To the extent that anything was acquired by the LLC for above its current market value, it was a fiscal exercise, imo.

            • Cullen Roche says:

              Personally, I think its fair to describe the bailouts as mostly fiscal in that they secured capital gains and caused inflation that would have otherwise been losses and deflation.

      • beowulf says:

        I think the Fed could only if Tsy ordered it to, which would sort of make a hash of central bank independence.
        “…the Secretary or an agency [e.g. the Fed] designated by the Secretary, with the approval of the President, may deal in gold, foreign exchange, and other instruments of credit and securities the Secretary considers necessary. ”
        31 USC 5302

        Ponder how broad the words the word “securities” is. To quote the definition from the SEC Act:
        “The term “security” means any note, stock, treasury stock, security future, security-based swap, bond, debenture, evidence of indebtedness, certificate of interest or participation in any profit-sharing agreement, collateral-trust certificate, preorganization certificate or subscription, transferable share, investment contract, voting-trust certificate, certificate of deposit for a security, fractional undivided interest in oil, gas, or other mineral rights, any put, call, straddle, option, or privilege on any security, certificate of deposit, or group or index of securities (including any interest therein or based on the value thereof), or any put, call, straddle, option, or privilege entered into on a national securities exchange relating to foreign currency, or, in general, any interest or instrument commonly known as a “security”, or any certificate of interest or participation in, temporary or interim certificate for, receipt for, guarantee of, or warrant or right to subscribe to or purchase, any of the foregoing.” 15 USC 77b

      • “What Sumner doesn’t understand is that this is the Fed doing fiscal so it’s kind of ironic that he frequently mocks fiscal when his big bazooka is actually the Fed doing fiscal….”

        Yeah, he seems to want the Fed to get into the business of “picking winners…”

    • Let me try it this way.

      I take a gov’t bond to the fed and say give me some central bank reserves. It seems to me the fed will say no.

      I take a gov’t bond to the fed and say give me some currency. It seems to me the fed will say no.

      • Good point. It’s about who can force “redemption.”

        There are three types of assets that the Fed can issue and “retire” (absorb into, or emit from, their black hole): currency, bonds, and reserves.

        The Fed controls the bonds-reserves swaps (though only by selling/buying at attractive prices, which effectively “forces” the banks in aggregate to comply). Neither individual banks nor the banks in aggregate can force the Fed to make these exchanges.

        The banks and their customers control the reserve-currency swaps (the Fed is required to exchange these on demand with the banks, and the banks manage their currency holdings based on supplying customer needs).

        There are no bond-currency exchanges. Banks buy currency with reserve balances.

  10. Great post. Anyone who understands double-entry bookkeeping should read ‘Monetary Economics’. It should be the starting point i.e. the framework for all macroeconomic models from the accounting perspective. Now if I can oly get to grips with the steady state stuff, I might make some headway myself.

  11. “The neo-classical concepts of exogenous money and the money multiplier and loanable funds and ISLM and supply/demand equilibrium …”

    Does exogenous money and the money multiplier assume excess bank capital and an excess of creditworthy borrowers who want to borrow?

    Does loanable funds mean a 100% capital requirement?

    What if the ISLM curves intersect at a negative interest rate?

  12. Isn’t a big effect of QE a sort of fiscal transfer payment to bond traders by way of capital gains? QE generates price volatility for treasuries. That allows people to buy bonds when they are cheap and sell them a few months later when they are expensive. The Fed and treasury act as a patsy in the trade and so it is a way to provide hand outs to the finance industry. If 30year treasuries are swinging in price by 30% over the course of a year, that totally overshadows the 3% coupon payments.

    • Its an interesting type of handout. The Fed as a participant is making $ 100 billion a year from QE – which includes the running interest margin effect of its own un-booked capital gain – and goes to reduce the deficit.

      Anyway, apart from that, its the same thing that happens in the market in any down-cycle of interest rates. Think of it as a virtual Fed Funds rate that’s dropped below the zero bound and dragged the curve down a bit further as a result.

  13. Nick Edmonds says:

    I agree with most of this but I don’t understand what you’re saying in your critique of SK’s attempt to redefine aggregate demand. Surely we do add income and change in debt in the identity (for any agent, sector or entire economy) that says that expenditure equals income plus change in debt less change in financial assets (suitably defined, excluding revaluations, etc.)? Isn’t the problem really that you can’t just leave out one bit of the identity (acquisition of assets) and still expect it to hold?

    • For the economy as a whole, GDP = expenditure = income (basically).

      And this must hold for any accounting period.

      Aggregate demand is a measure of actual or potential expenditure, in some defined context.

      Now it may be the case that an increase in debt – somewhere or anywhere – can increase aggregate demand. Keynesian economics is an expression of that, mostly at the level of “autonomous” government expenditure, financed by a deficit.

      And that’s fine.

      But the fact is, the result of that (or the expectation of it) is still expressed in terms of that first equation. Both expenditure and income will expand, by tautology of consistent measurement over a given time period.

      So you don’t suddenly stick a change in debt factor into that equation in order to make it “balance”.

      It’s always in balance.

      The issue is how a change in debt can change the quantities in that equation – not how a change in debt suddenly gets inserted in that equation.

      This is what I meant by “orthogonal” measures in the post.

      Orthogonality can be seen immediately in another way.

      A bank makes a loan. It credits the borrower’s deposit.

      That is a change in debt.

      But nothing in that change represents a change in income.

      The borrower has to spend the money to change expenditure and income.

      And it’s the act of spending that provides the impetus to aggregate demand – and it doesn’t matter whether that spending comes from existing money balances (perhaps by a change the velocity of money), or from new money balances (perhaps by a change in debt).

      Viewed another way, the change in debt is a discrete financial event that precedes or is separate from the action by which the resulting money may be spent, which is a separate discrete financial event.

      The first change is captured in accounting flow of funds. The second change is captured in NIPA.

      Separable, although connected, through behavior and accounting.

      For example, connection is better captured in the notion of regression analysis.

      And such a regression is FAR from perfect – because debt can be used for purposes other than aggregate demand in the sense of the first equation – e.g. it can be used to lever up in the acquisition of equity securities, which has no direct bearing on aggregate demand, properly defined.

      But that first equation always remains the same for a given time period. It’s an accounting tautology.

      And contrary to Keen’s parallel mathematical quest, continuous time mathematics has nothing to do with it – the accounting does not change. The change in debt doesn’t suddenly become part of that equation simply because the time period changes from epsilon to zero.

      It’s just wrong, wrong, wrong logic, whether in discrete or continuous time.

      • Nick Edmonds says:

        This may just be a question of emphasis.

        I tend to start by thinking of in terms of individual agents, for whom the basic budget constraint is, as I said, that expenditure equals income plus change in debt less change in assets. Of course, this is true in discrete or continuous time.

        As you then aggregate, income and expenditure are added, but changes in debt and assets can be cancelled out where they involve flow of funds between the agents or sectors being aggregated.

        Once you have aggregated all the agents within an economy, you are then left with expenditure equals income less net acquisition of net foreign assets. At this point, the relevant income and expenditure of rest of world (i.e. exports, imports and net interest income from abroad) are added in to close it off.

        At this point, there should be no change in debt or assets left in the identity. So, maybe that’s the same thing and I just got there by cancelling stuff out and you got there by not putting it in in the first place.

        In any event, I completely agree that SK’s equation cannot be an identity and to try to explain it as being due to differences between discrete and continuous time or ex-ante and ex-post is unhelpful.

        However, I do think it useful to think in terms of the full identity including change in debt and assets, because it brings out some useful insights. A good example is the one you mention about use of debt to fund asset acquisition. If we consider the non-bank private sector as a whole, then we have to dig a bit deeper to understand the impact of increased debt. If an agent in the NBPS borrows from a bank, but acquires equity from within the NBPS (whether in the primary or secondary market), this has not given us enough to balance the identity. In order to get it to balance we either need someone else to reduce their debt or acquire assets from outside the NBPS, or we need an increase in expenditure (or conceivably a reduction in income). By assumption, whoever is doing this is not the person buying the equity, so we have to ask what things in the economy are going to change to make these other agents change their behaviour. Very often, the answer is going to involve, at least in part, an increase in aggregate income.

        Again, none of this justifies SK’s definition of aggregate demand, but it does help illustrate why increased debt may have an expansionary impact, including debt for asset purchase.

        • “maybe that’s the same thing”

          I think so

          Certainly agree that it always invites analysis for the effect on aggregate demand, and that there can be direct effects in the case of consumption or new investment demand, and even indirect effects in the case of any asset demand.

          But not an equation for aggregate demand!

          And here’s a nice illustration – one of millions – as to why it’s not an equation:

          http://money.msn.com/top-stocks/apple-borrows-to-fund-dividend-share-buyback

          I think Steve Keen should find a good MBA school, and sit in on a first year financial accounting course.

          He’d then be off to the races – and if he did so, should be thanking me for the rest of his career.

      • I thought part of Keen’s intent was to remove the distortion of categorizing investment vs. consumption, i.e. increasing debt for “investment” is often “mislabeled” and should be consumption. E.g. private residential. In this case, he is challenging the accounting categorization of investment vs. consumption, but perhaps this is still the same as challenging income vs. flow of funds accounting, but indirectly through the categorization.

  14. @JKH:

    It acknowledges and uses the powerful logic that accounting is not just a rear view measurement device – it is also a constraint on all forward looking projections of economic outcomes – meaning that it is an important condition in the substance and shape of good economic analysis.

    Just to reiterate: yes it’s a constraint on forward thinking. You can say:

    “Your thinking/projections would result — looking back at the end of the period — in an impossible accounting imbalance. So your thinking must be wrong.”

    Accounting identities can only — and only sometimes — show that projections must be wrong. They can never prove that they are right.

    The real problem arises when people use post-hoc certainties (i.e. Y=C+I) to project ex ante that if C declines, I certainly will, must increase. (Cue Scott Sumner.)

    • “They can never prove that they are right”

      You keep saying this Steve, and its right, but its a red herring in the context of the subject of this post.

      Nobody here is saying that random accounting compositions (including those that satisfy identities that are necessary but NOT NECESSARILY SUFFICIENT) can be used to predict the future – but it is a frequently used rhetorical attack used by those who don’t get the real point (cue Sumner, Rowe, Smith, et al …) – and I’ve anticipated that standard, commoditized neo-classical reaction in the first paragraph of this post.

      (The red herring on the part of those who have commoditized such an attack is that they reformulate an argument of necessity as being an argument of necessity and sufficiency. Such an attack is a misrepresentation and in essence a lie about the nature of the argument.)

      experiment – most of these guys believe in the money multiplier in some way.

      show me – using the actual, correct, detailed accounting entries, including timing, for central and commercial banking – how it can be the case that they believe in the money multiplier without being completely lost on the related accounting

      (i.e. proof by contradiction – you can’t both understand the accounting and still believe in the money multiplier – its impossible)

      and ditto for exogenous money, loanable funds, ISLM and supply/demand equilibrium models

      • I’ll try. I believe other people assume excess bank capital and an excess demand of creditworthy borrowers who want to borrow. They may not realize these assumptions. The bank does not want to borrow from the fed. If it creates a demand deposit and loan, it thinks the fed funds rate will rise.

        I now deposit $100,000 of currency (not demand deposits) in that bank. The fed funds rate starts falling. The fed does not react. The bank now makes $1,000,000 demand deposits and loans. The fed funds rate returns to where it was assuming a 10% reserve requirement.

  15. Haven’t read all the comments, yet, maybe covered already, but:

    The basic problem with Steve’s post is he starts with the banks having only reserves and loans as assets. But: they also have bonds.

    So his rather byzantine foray into bank liabilities is unnecessary. From the perspective of the banks’ balance sheet (assuming Fed buying bonds from banks), it’s all on the asset side: bonds for reserves.

    • I covered that in the post

      The banks have bonds in his description of the transactions – but he completely forgets about that in his accounting for the transactions – hence the magical “triple entry” accounting result that I alluded to

      As also described in the post, in the real world the original source of the sold bonds is not the banks, and as a result, following through the macro flow to the bank liability effect is absolutely essential to understanding what’s going on in QE – but that’s unrelated to whatever he’s illustrating, given that he goes off the accounting rails pretty early in the analysis (like out of the gate early)

  16. Why isn’t accounting the bedrock of all economists work?
    I wonder whether they don’t want bullshit to be constrained in the way that it would be if accounting were followed. Perhaps notions such as demand never constraining the economy would be shown to be false. They want a way of thinking that leaves things ambiguous so that preconceived ideas can be perpetuated????
    Noah Smith had a post about the problems with macroeconomics and in the comments seemed to be saying that accounting wasn’t an issue: http://noahpinionblog.blogspot.co.uk/2013/04/the-reason-macroeconomics-doesnt-work.html

    I’ve pasted those comments below:

    Stone:
    I thought that accounting was to macroeconomics what chemistry and physics are to geology. And yet apparently ideas such as loanable funds and the money multiplier still occupy macroeconomics and yet do not abide by the principles of accounting. Isn’t that like geologists getting tangled up in theories that contradict core principles of chemistry?
    Noah Smith4:37 PM
    No, I don’t think so. Accounting doesn’t give you any sort of theory about how the world works, it’s just a method of collecting, defining, and tabulating data.
    MMT people who think macroeconomics can be derived from accounting alone are basically full of it, if you ask me!
    stone5:03 PM
    I totally agree with you that thinking of deriving macroeconomics from accounting alone would be as inane as attempting a deriving geology from chemistry without looking at the earth. Equally though having macroeconomics theories that don’t conform to the principles of accounting is as inane as it would be to have geology theories that don’t conform to the laws of chemistry. Accounting doesn’t just tabulate data – it provides a framework into which any description of monetary reality has to fit. Just as chemistry and physics provide a framework into which any description of physical (eg geology, cosmology) reality has to fit.
    Noah Smith5:14 PM
    Could you show me, in math rather than words, an example of a macro model which does not “conform to the principles of accounting”? Remember, math, not words.

    • beowulf says:

      “I thought that accounting was to macroeconomics what chemistry and physics are to geology.”

      That’s a great analogy. The laws of geology cannot contradict the laws of chemistry or physics.

    • Smith’s type of mentally constipated view of the issue is exactly why I wrote the first paragraph of this post.

    • “an example of a macro model which does not “conform to the principles of accounting”

      in the post, I listed 5 in a single sentence

      • JKH, on that thread, other commentators waded in to diss the whole economics needs accounting idea:

        There is no model Noah.
        People who think macro can be derived from a couple accounting identities have probably never done actual accounting.

        PrometheeFeu5:56 PM
        “Accounting doesn’t just tabulate data – it provides a framework into which any description of monetary reality has to fit.”

        It’s impossible for any monetary theory to violate accounting principles because accounting is entirely agnostic as to what happens in the world. For instance, if money is disappearing into thin air, I can just have a “Money that has disappeared into thin air” account and whenever that happens, I put a credit in the cash account and a debit in the “Money that has disappeared into thin air” account. And just like that, physical impossibility has neatly fit into my balance sheet without the slight difficulty.

        Dohsan8:20 PM
        Right. Accounting isn’t to economics as chemistry is to geology (that is, a building block). Something that makes accounting sense does not have to make economic sense at all. Accounting is more like a language. It’s more like French to geology. You could write a piece about geology that is perfect French but violates multiple principles of geology (and chemistry and physics). People just decided on certain accounting rules so that information could be conveyed in an understandable (and hopefully honest) manner.

        stone2:57 AM
        If your macroeconomic theory requires a “money that has disappeared into thin air account” then you need to have that in the accounting and then it is up for observation whether operationally that is actually occurring. So for instance your theory might require a certain level of debt write downs so as to have that “money disappearing into thin air” BUT on the ground those debt write downs are not happening and yet the phenomenon you are intending to model is happening and so you know you have to go back and sort your theory out.
        I agree that something that makes accounting sense does not have to make economic sense at all (MMT fits that bill all too often) BUT although accounting is not sufficient it is NECESSARY. Homeopathy is biomedical science that has abandoned the constraint of conforming to chemistry. Much of macroeconomics looks to me (a lay person) like an analogous abandonment of the constraint of accounting.
        Do you dispute this stuff? :
        http://www.boeckler.de/pdf/p_imk_wp_100_2012.pdf

        • I’ve seen that paper before, but haven’t read it in detail yet

          Just noticed your necessary/sufficient distinction, which is exactly right – see my comments to Steve Roth above

          The real problem here is the flip side of my recommendation to Steve Keen – financial accounting and macro flow of funds accounting needs to be built into the basic economics curriculum

          (I’ve taken courses in both as part of a finance MBA, along with macroeconomics)

          In fact, a lot of economists simply don’t know this accounting – that’s very clear – and while some of them will admit it – when they admit it, they also dismiss it – return to paragraph 1 of the post

          It’s an inherently fear driven reaction in essence – hence the standard tactic of attempting to refute a necessity argument by charging that its not sufficient

          My observation is that it is very rare to see any discussion in economics that does not IMPLICITLY depend on a mutual understanding of accounting logic, at least, and quite often EXPLICITLY

          It’s a huge problem in the overall approach to the education element IMO

          One large piece of evidence in this regard is simply the degree of emphasis that post Keynesians put on accounting – they wouldn’t feel the need to do that if all was right with the world otherwise

          I’ll bet rings could be run around the neo camp of Noah Smith et al on actual accounting knowledge as it relates to how the financial system works – starting with the money multiplier – and I think Krugman was bluffing in his debate with Keen when he said he knew banking T accounts – and his debate with Steve Waldman a few months back also suggests something quite different

  17. Seeking feedback:

    I’ve been thinking of QE’s effect in a very simple way lately:

    It effectively decreases the net inflow of financial assets into the private sector (from Treasury and GSEs).

    Cullen showed this in the Citi chart here:

    http://pragcap.com/the-short-supply-of-assets

    That reduction in flow supply of financial assets increases the prices of all financial assets. While the reduction continues.

    (More detailed mechanism: reduced bond inflows increase bond prices and reduce yields. The market buys more equities to get yield, driving up equity prices.)

    The reserves created (directly when buying from banks, indirectly when buying from nonbanks with deposits) are immaterial until (if?) the Fed decides to sell.

    When buying from nonbanks, an additional financial asset is created in the private sector: deposits. But since those deposits could have been created by repoing the bonds, it’s not clear whether that has any macro effect.

    If the only thing that matters is the reduction in bond (financial asset) inflows to the private sector, resulting in higher bond (financial asset) prices, we would say that the only QE effect is the wealth effect.

    Thoughts?

    • Re QE effect – it’s not a reduction in financial assets

      (see post)

      It’s a reduction in securities

      And it’s a shortening of macro duration

      Which means duration becomes increasingly scarce

      Which means bonds get bid up (other things equal)

      Which means wealth effect

  18. Oh also I have a question on Cullen’s Citi chart: does it incorporate the natural retirement of bonds as they expire? If not, shouldn’t that be subtracted from the net flow? They are “departing the market,” after all…

    (I conceive the Fed purchases as effective retirements from the private sector, at least for the time being. It’s not clear when or if they will sell the bonds back to the private sector.)

    I’d love to see a net net net inflows time series/fever chart incorporating all that. JKH, Cullen, you got the chops to assemble that? I could do it but don’t know the FOAs quite well enough to know what’s included in various measures, would worry I’m getting it wrong.

  19. JKH said: “Aggregate demand effects? I’m open on that. It would involve the wealth effect of course – but that shouldn’t be totally discounted.”

    I end up selling $700 in gov’t bonds to the fed. At my bank C, my account A gets marked up by $700 in demand deposits, and bank C’s account at the fed gets marked up by $700 in central bank reserves. I now decide to stop saving and spend the $700 on a notebook at a retailer. The retailer has an account B at bank D.

    I write a check for $700 from my account A at bank C. The retailer deposits it in its account B at bank D. My checking account A gets marked down by $700, and bank C’s account at the fed gets marked down by $700. The retailer’s checking account B gets marked up by $700, and bank D’s account at the fed gets marked up by $700. I like to think of it this way. The demand deposits get transferred from my account A at bank C to the retailer’s account B at bank D, and the central bank reserves get transferred from bank C’s account at the fed to bank D’s account at the fed. Notice the central bank reserves are not in my account A at bank C or the retailer’s account B at bank D. The check/demand deposits settle the transaction.

    Assuming one commercial bank or assuming the check gets deposited in the same bank C allows that to be seen easier. With either of those assumptions, the central bank reserves don’t move at all from bank C’s account at the fed. The vault cash does not move either. The demand deposits/check move (mark down my account A then mark up the retailer’s account B) to settle the transaction and circulate in the real economy.

    I’m saying the check is payment. I basically bartered $700 in demand deposits for a notebook.

    Just to be clear:

    1) I don’t have an account at the fed for my central bank reserves that link to my checking account and the retailer does not have account at the fed for its central bank reserves that link to its checking account. Is that correct?

    2) Central bank reserves can’t be “deposited” into a checking account, correct?

  20. JKH,

    I came across this blog post which I hadn’t seen earlier. It is on Keen from last year:

    http://slackwire.blogspot.in/2012/04/case-of-keen.html + lots of comments.

    Generally okay but has some errors of its own especially on dimensional analysis.

    • very cool!

      “But until he writes in a language spoken by people other than himself, there’s no way to know.”

      exactly – including the known language and logic of accounting

      BTW, I’ve mentioned this numerous times, and it becomes more perilous to do so as the general piling on wrt Krugman continues apace – but Krugman was absolutely right on his central point in his opening salvo on Keen last year which WAS the form of the so called aggregate demand equation. Unfortunately, the K man got terribly tangled up in the barbed wire of peripheral endogenous money issues, as he often does. But his main point was THE point and it was right and its the same point as in the Slackwire post and as contained as a small item in my post. And Krugman specifically said that Keen should be called out on this.

      • Yeah and I hope he has learned something from the criticisms such as yours which is quite complete and doesn’t continue with his aggregate demand/change in debt stuff.

        We will find out in a video of a Harvard lecture he is going to present soon and “address” the issue (once again).

    • Actually, its interesting that when you read how Krugman writes about bank money, one of the errors he makes seems to be along the lines of the fallacy of composition – if you don’t know how endogenous money works at the macro level, you can really get screwed up when writing in a style that seems to conflate micro and macro views, which he tends to do

      • fallacy of composition??? When I read Krugman, I believe he says banks are only financial intermediaries. Is that the same thing?

    • Nick Edmonds says:

      I’d agree with that assessment, but SK certainly has difficulty with mixing dimensions. I made a related comment on one his recent posts and he seems to be aware of the problem.

      http://debunkingeconomics.com/2013/05/endogenous-money-and-effective-demand-2/#comment-1227

      • Your comment is yet another way of demonstrating that his AD equation cannot hold, IMO. Not only is the quantity relationship not 1:1, but neither is the timing.

        For one thing, even leaving aside the quantity relationship, the idea that the timing of the effect is uncertain just points to the fact that borrowing and spending are two discrete events.

        And invoking continuous time doesn’t change this. It appears to me that the original argument somehow suggested that invoking continuous time somehow implies that borrowing and spending magically become simultaneous. If that was part of the rationale for why continuous time delivered the knockout blow, it’s completely wrong. Continuous time has nothing to do with the discreteness of events.

        It’s just impossibly naïve as an equation IMO, and can only be used in a regression context, with appropriate timing subscripts to boot. Even there, its very rough. e.g. if I borrow to buy Apple stock, the direct or even indirect AD effect is almost completely unrelated to the quantity of my borrowing.

        • Nick Edmonds says:

          I don’t like SK’s aggregate demand definition any more than you, but I’m not sure you need to see borrowing and spending as discrete events to find fault with it.

          For a start it seems to me that some borrowing is inextricably linked to spending. Credit card spending is an example, or spending out of overdraft. Of course, you can always bifurcate these into a borrowing transaction and an expenditure transaction, but the actual book entries do not reflect such a split.

          However the important point is that, even if the only borrowing that took place was that arising in relation to an item of an expenditure, it would still not make SK’s definition correct. It could be the case that every item of borrowing correlates with an item of expenditure, but then the residual expenditure must be equal to income less saving. The only way you can add borrowing to income to get expenditure is if you also subtract out saving and in a complete system the borrowing and saving must cancel out.

          Whichever way you look at it, this equation cannot be an identity.

          • “the actual book entries do not reflect such a split”

            oh, but that’s exactly where you see the split

            Spending from credit card borrowing shows up in NIPA

            Credit card borrowing shows up in the flow of funds but not NIPA

            And they are different/split accounting entries

            And the same holds at the micro level if you were to construct a proper set of financial statements for households that was congruent with those for the rest of the economy, with a separate income statement and sources and uses of funds statement – and its only logical and orderly to conceive of macro and micro accounting as congruent in this way, and in fact necessary for coherent analysis overall

            But I’ve thought of the credit card case, and I take your point that it’s one where the transactions are nearly simultaneous (you can probably make that case for not exactly in detailed operational analysis, but that’s neither here nor there) and where the likelihood is that borrowing is for spending from income rather than for asset swaps from saving – so a regression component should work pretty well there

            And your final point seems about right although I haven’t been quite understanding the way you’re been phrasing it – my own version is that micro borrowing to consume amounts to micro dissaving at the margin, which must always be offset by corresponding micro saving at the margin, somewhere, to get macro closure. Maybe you’re saying that – I think you are – and I’m having a bit of trouble in reading it.

            • Nick Edmonds says:

              Actually what I meant by book entries was the actual journal entries. As I understand it, a credit card purchase would only normally entail the entries: debit expenditure, credit creditors, without any entries on the cash account or bank balance account. In contrast, if you took out a loan, then spent it, you would have debit cash, credit creditors followed by debit expenditure, credit cash. That said, I also am not an accountant, so I may be wrong on this.

              Anyway, I accept your point about needing to bifurcate to record the impact of the transaction in the national accounts.

              • I guarantee you that if a corporation uses a credit card to spend (which it might do for office supplies, etc.) , there must be entries for both the income statement showing the expense from spending and the balance sheet showing credit card balance due as a liability, if outstanding at the end of the accounting period, along with other required entries for net effects on cash, inventories, profit, etc. etc.. In accounting terms, the spending and the borrowing are separate accounting events, even though they may appear to occur simultaneously.

                (‘Accounting event’ is a term of the accounting trade; at least it was when I took my financial accounting course.)

                And as I said, household accounting if invoked would show comparable treatment in terms of spending, saving, and its balance sheet.

                • Nick Edmonds says:

                  Again, I’m talking about the actual journal entries, i.e. what goes into the ledger accounts, not how the final accounts are extracted from them. The individual ledgers in themselves say nothing about whether items are income statement or balance sheet items.

                  I would expect the journal entries for a credit card purchase to be what I indicated before: debit an expenditure ledger (say, office supplies), credit a creditor (here, credit card issuer). Production of the final accounts then requires production of a trial balance from all the ledgers, followed by a further double entry, specifically: credit office supplies, debit P&L. No such entry is made in relation to the creditor balance, so this ends up on the balance sheet.

                  Of course, if the credit card purchase is for a capital item, then you wouldn’t have that double entry to P&L, meaning that the item purchased would remain as a balance sheet item, but you would then have appropriate entries in the depreciation ledgers.

                  I have never actually looked at accounting for credit card purchases, so I can’t be sure this is correct, but this is how I was taught to do it for finance leases, where the same idea applies. The transaction is treated as one double entry (debit asset, credit lessor), rather than two (debit cash, credit lessor; debit asset, credit cash). Of course, once the financial statements are produced, this distinction is not visible.

                  I should say that my experience is with UK accounting rather than US and although, I am reasonably familiar with US GAAP, I don’t know for sure how ledger accounting works over there. However, I’d be surprised if it’s different.

                  • Thanks for the detail

                    I’m not a journal entry expert

                    So point taken, assuming you’re correct, which I can’t dispute and have no desire to

                    But the issue for the post is financial accounting (and NIPA and flow of funds accounting)

                    And that’s the ultimate reason for keeping a journal, I believe

                    just as a matter of interest and my own education:

                    Suppose you have a credit card balance due that you pay off with cash

                    By comparison, suppose you have a bond maturity due that you pay off with cash

                    Both those situations require some financial accounting, obviously, in order to complete both the income statement effect of the borrowing with interest cost and the balance sheet effect of paying off what’s due with cash

                    How is the point you are making regarding journal entries different in those two situations?

                    Thanks

                  • Like I said, I’m a dope when it comes to journal entries – in the context of what I do understand about financial accounting

                    But it looks to me from this following that journal entries apply to both the income statement and the balance sheet (which is what I’ve always assumed), and so must be done separately for each of those (which is what I’ve always assumed)

                    http://en.wikipedia.org/wiki/Journal_entry

                    A pretty amateur looking Wiki entry, but is it wrong?

                    • Nick Edmonds says:

                      Well, again, although I had to learn about double-entry book-keeping (many years ago), I’m really not an accountant, so I may be wrong on this.

                      That said, my understanding is that there would not be any difference between your two examples. Both would involve the entry: debit creditor, credit cash and since this transaction has no P&L impact, there would be no further entry and the end balances would simply show reduced cash and reduced creditors.

                      I don’t think that Wikipedia page is very helpful, because none of those tables are ledgers. The linked page on double-entry bookkeeping is better because at least it shows what the ledgers would look like. Unfortunately, the explanation is not very clear and nobody has got round to writing the text that explains how you extract the trial balance and P&L.

                      I can hardly remember how we even got into this, but I think it was actually in the context of me saying that I don’t think your position that SK’s definition is wrong (with which I completely agree) depends on the possibility of borrowing being separate from expenditure. My point was that even if all borrowing was inextricably linked to expenditure, it would still be wrong.

                      So interesting (?) as this stuff may be, I don’t want to distract from the fact that I am 99% in agreement with you.

                    • Well, I dragged out my old financial accounting text to figure out how I could be such a dope on this, and it turns out that I’m not (at least not on this).

                      Ledger entries must be made for both income statement items and balance sheet items (as I thought). In that regard, there is no difference for spending or investing or acquiring financial assets with cash from a bond issue than the case of spending from credit card borrowing. Both types require stock/flow accounting consistency, including the ledger. And the point of the previous bond/credit card comparison question was to draw out how you thought different.

                      The credit card issuance requires immediate balance sheet representation and the spending requires immediate separate income statement representation – as represented in immediate ledger entries, which will be used to construct the periodic financial statements.

                      So spending from debt and issuing debt ARE separate transactions, by definition and by accounting (which is what matters for this post) and by accounting events, including ledger keeping, and both require stock/flow consistent accounting to represent them.

                      The case where spending from debt and issuing debt are somehow viewed the “same” transaction doesn’t interest me, because they aren’t in any kind of reality that includes coherent accounting. The fact that entries seem to occur simultaneously doesn’t obviate this.

                      And ledger entries are merely the back office, green eyeshade building blocks for the construction of financial accounting statements. I’ve never been interested too much in the categorization of the construction process for accounting statements.

                      In fact, most sophisticated financial institutions are capable of preparing financial statements on a daily basis.

                      Don’t recall exactly how we entered this discussion, but it wasn’t because I was interested in ledger entries.

                    • Nick Edmonds says:

                      I’m really not sure that you’re saying anything different to me. I’m not suggesting that balance sheet items don’t go through the ledger and I’m certainly not suggesting anything that is not stock-flow consistent.

                      My point (and of course I was making the point in the context of saying that the point didn’t make any difference) was that certain forms of borrowing only involve one set of double entries, and others involve two. I don’t think you’re disagreeing with that, are you? Both types will appear exactly the same in the balance sheet and P&L, whether that’s the annual accounts or daily management accounts. And I think we agree on how they will appear.

                      I also agree that the distinction is rarely relevant to economics discussions. So maybe we can agree that it’s a matter of personal preference as to whether you view such things as one transaction or two?

                    • @JKH: “The case where spending from debt and issuing debt are somehow viewed the “same” transaction doesn’t interest me,”

                      Totally agree. While spending with a credit card seems like a single event, to properly represent what happens, and balance my books with the books of the credit card company and the retailer, my CC purchase has to be posted as two separate journal entries/accounting events in my books — 1. borrowing, and 2. spending. Then those entries are be posted to appropriate balance sheet “accounts” on my books. Those journal entries and balance-sheet postings balance against the retailer’s and the CC company’s journal entries and balance-sheet postings.

                      And this leads to a pet peeve of mine: the usage “spending out of income.”

                      It’s a handy vernacular shorthand for “spending less in a period than you receive in income.” But it’s also nonsensical. You can’t spend out of the instant of somebody handing you a dollar bill. You can only spend from the dollar bill in your hand or your pocket (on your balance sheet).

                      If you deduct spending from income before making the journal entry(ies) (telescoping the two events into one), you’re bypassing the multiple balance sheet posting, and hiding the true transaction flow.

                      This is the basic construct of flow-of-funds accounting as I understand it: instantaneous flows are represented by journal entries. Those journal entries are posted to balance sheet categories/accounts, changing stock measures.

                      “Spending out of income” conflates discrete accounting events just as when “spending from debt and issuing debt are somehow viewed the ‘same’ transaction.” You can telescope those events together, but you’re hiding an important reality.

                      Again, “spending out of income” is a handy vernacular shorthand. But it leads to accounting confusion by many, especially when you add debt and credit to the mix.

                    • I am disagreeing with that.

                      But I think we’ve discussed it enough so lets leave it for the time being.

                    • i.e. agree to disagree in an agreeable way

                      its not an enormous point

                    • ““Spending out of income” conflates … ”

                      Just means consumption function has income on the right hand side with a proportionality constant. No conflation.

                    • Steve Roth June 6, 2013 at 10:42 am

                      excellent point first half comment

                      quasi-excellent point second half, in terms of operational sequencing

                      I’d just say that spending out of income WITHIN A GIVEN ACCOUNTING PERIOD does make sense as a statement of cumulative effect, and it is stock flow consistent – i.e. a dollar flow of income in the first part of the period may be spend in the latter part of the period – and the income statement accounting will be tight

                      When it gets to spending with the proceeds of borrowing, you have to ensure flow of funds accounting exists and complements income statement accounting, and you have to ensure consistent periodicity of accounting results for the two together

                      But your point on the sequence of operations is a good one in general, I think

                    • “I’d just say that spending out of income WITHIN A GIVEN ACCOUNTING PERIOD does make sense as a statement of cumulative effect, and it is stock flow consistent – i.e. a dollar flow of income in the first part of the period may be spend in the latter part of the period – and the income statement accounting will be tight”

                      Right. I mainly object because if the entity borrowed in that period, was that spending “out of income” or “out of borrowing”? Fungibility, “what ‘funds’ what,” all that rot. Ditto problems if the entity paid off debt, lent, or received loan repayments. (Or issued/bought back stock.)

                      If it’s a company and they paid dividends but borrowed the same amount: was investment spending out of income/profits, or out of borrowing?

                      I guess I’m saying you can describe inflows (income and debt-related), and you can describe outflows (expenditures and debt-related), but assertions about their relationships can be quite arbitrary. Accountants try to manage that by looking at intentions — “we borrowed that money to buy drill presses” — but those are judgment calls — some more obvious than others.

                      (And of course those choices can be directed to virtuous/honest and less-than-virtuous ends.)

                    • ““Spending out of income” conflates … ”

                      Just means consumption function has income on the right hand side with a proportionality constant. No conflation.”

                      If you happen to be pondering a standard Keynesian consumption function, which does not include a wealth/net worth term and all spending decisions are a function of income, that is certainly true.

                      But many have pointed out that a consumption function absent a net worth term is not well founded.

                      I would quite agree, especially given my contention that one can only spend from a stock (though that stock may consist of your unused credit-card limit…), not from a flow.

                    • “If you happen to be pondering a standard Keynesian consumption function, which does not include a wealth/net worth term and all spending decisions are a function of income, that is certainly true.”

                      Not true. Godley/Lavoie have a consumption function which has both income and wealth.

              • Let me try. It will probably need cleaned up.

                Swipe credit card.

                Assets CC bank = $100 loan to John
                Liabilities CC bank = $100 in demand deposits in John’s account

                Assets John = $100 in demand deposits
                Liabilities John = $100 loan to John

                Immediately “move” demand deposits (mark down and mark up) to pay for real good/service. Central bank reserves may need to be “moved” too.

                Assets John = $0 in demand deposits
                Liabilities John = $100 loan to John

                Assets retailer = $100 addition to demand deposits of its account

                You just exchanged demand deposits (MOE and MOA) for a real good/service.

                • Nick Edmonds says:

                  What I’m saying is that I don’t think John’s deposit ever appears, i.e. no entries are made on any ledger for John’s bank account.

                  Also, those are not journal entries, they’re just balances.

                  • I believe that is the right way to think about it. Now whether a step gets skipped in the real world, I don’t know about.

                    Try this. Start with a $0 credit card balance. Now have a previous charge reversed so that the credit card account has a negative balance of $100. Now charge $150 on the credit card.

          • also, note that your credit card statement is in effect a specialized personal report that consolidates the standard separate financial accounting perspectives of spending (income/NIPA) and borrowing (sources and uses of funds/flow of funds)

        • I haven’t quite grasped the point that you are making about Apple borrowing to buy back Apple stock. Please explain more -sorry for me being slow on the uptake.
          I have wondered whether companies have a big and unfortunate effect on the economy when they borrow to buy back stock. The next step is to direct revenue towards paying down debt instead of towards investment in research or training or new machines or whatever. Debt allows a company to convert the prospect of a perpetual modest return into a short term larger return followed by death of the company. It is like cutting down the trees in an orchard to sell them off for firewood. You get big short term returns if you load a company down with debt and then manage to pay it off but over an extended time frame it doesn’t make any sense. It is slash and burn capitalism.

          • just that debt is not necessarily associated directly with spending

            when Apple borrows to buy back stock, its increasing the overall debt in the economy but there is zero direct impact on spending and GDP – it’s merely changing the capital structure of its balance sheet from equity to debt

            so that’s just one example where SK’s equation AD = income + change in debt just falls apart – because there’s absolutely zero direct connection between Apple’s debt issue and aggregate demand or income

            even any downstream derivative effects would only be attributable to very tenuous arguments IMO – e.g. any money “freed up” by the stock buyback is the same amount as the quantity of money that’s necessarily “used” as the source of funds for the debt issue, and in no way is the aggregate amount of the debt issue relatable to any equivalent aggregate demand or income effect

            • “when Apple borrows to buy back stock …”

              Is Apple borrowing from a bank or not?

              “so that’s just one example where SK’s equation AD = income + change in debt just falls apart”

              I think if someone looks far enough, Keen includes financial assets in his definition of AD. I think that is a poor definition.

              • And he includes supply of financial assets in aggregate supply which is equally hilarious. “Aggregate supply” is just a short form of aggregate supply of goods and services which he somehow doesn’t seem to know.

                • right

                  symmetrically hopeless in demand and supply

                • Right, maybe someone should point this out to him?

                  I believe this point is where he and Krugman got into a dispute?

                  • yeah guess so. We can debate whether Krugman’s language was good or not but he precisely pinpointed Keen’s errors.

              • possibly

                and yes, poor

            • JKH, I’m still not getting your idea that it is immaterial when debt increases in that way. When Apple issues debt to buy back stock; there is credit expansion. The change in capital structure is from a perpetual asset to a short term asset. Let’s imagine they sell $100Bn of bonds and a bank makes loans to buy those. The consequent $100Bn of bank deposits go to Apple and Apple buys stock back from shareholders. The share-holders then have $100Bn burning a hole in their pockets. There is less equity in Apple but bank deposits burn a hole in your pockets more than Apple shares do. If you just sit on equity, that equity (hopefully) will preserve value. Bank deposits on the other hand are looking for something to be spent on. Over the life of those bonds, the interest paid out to the bond holders will I guess have to find its way round to pay off the bank loans made to buy them. But that is over the course of the next few years. The equity to debt conversion gives an initial jolt of spending impetus followed by a long drawn out debt repayment hangover. It is especially grim because the shareholders are likely to spend the buyback money on bidding up the price of the remaining equity that will subsequently crash in price. Then Apple will have to cut down on its debt level and that deleveraging will suck money out of the economy.
              I guess change in the level of corporate debt in the economy is upstream of disposable income and GDP? Household income will be higher when corporations are leveraging up and household income will be lower when corporations are deleveraging. Increasing/decreasing household indebtedness then adds another layer before we get to disposable household income. People re-mortgaging their houses and then spending the money is much like the Apple buyback.
              I agree the whole phenomenon would make no difference to the economy IF people were spending the ready cash from credit expansion in exactly the same way as they spent the slow trickle of money that comes from holding stock or not having a mortgage to pay off. The reality isn’t like that though is it? Sorry if I’ve totally misunderstood everything.

              • Hi Stone,

                I’m not saying it’s absolutely immaterial – but arguably just about that – because it’s not necessarily material to the issue of aggregate demand effects, properly defined (i.e. in terms of income and GDP expenditure).

                I’d suggest two steps in thinking about the example here:

                Apple does a debt issue to repurchase stock.

                First, assume that those purchasing the debt are the same people that are selling their stock back to the company.

                Then, there is no net cash effect on the economy, so there’s no net cash to be spent. And there’s not much in the way of a liquidity effect otherwise. Bonds and equity securities are both longer term financial assets, and there’s no reason to think that one of them would have a materially greater direct effect on aggregate demand and income than the other.

                Second, now assume that the group purchasing the debt is different than the group selling their stock back to the company.

                The group purchasing the debt will use cash as a source of funds.

                The group selling the stock will gain cash as a use of funds.

                Perhaps the group selling the stock and gaining cash will have a propensity to spend. But it’s just as likely that they will engage in asset reallocation in their financial asset portfolios. People sell stock all the time. They don’t immediately go out and spend it all.

                The group buying the debt will be using up cash. Perhaps they will cut their spending plans. But it’s just as likely that they too are engaging in asset reallocation. People trade bonds all the time. They don’t immediately cut their spending plans to buy bonds. They may just be switching from some other asset in their financial portfolios.

                So in total, there’s absolutely no reason to identify the amount of Apple’s debt issue with an equivalent amount of incremental aggregate demand in the economy – in the sense of the SK equation. It’s not happening AT ALL at the level of Apple itself, because they are simply change the capital structure of the Apple balance sheet – without any net cash effect to the Apple treasure – insofar as the amount of the debt issue is concerned, which is the focus on my example. And it probably isn’t happening on balance at the level of Apple’s debt and equity holders – no more than what usually happens in the economy in the case of financial asset turnover in general. Aggregate demand doesn’t increase every time there’s a stock or bond trade.

                • JKH, thanks for the detailed clarification. You are meaning what I thought you were and it is based on the idea that bond purchases don’t use fresh short term credit (money created with a stroke of a pen) any more than does stock ownership. Am I so mistaken in thinking that that is far from true? I thought that credit default swaps meant that bonds could be bought by financial institutions with very very little capital. I thought if Apple converts from being $400Bn in stock to being $350Bn in bonds and $50Bn in stock, then in practice much of that $350Bn will be funded by expansion of the shadow banking system not a simple transfer between current stock owners having the same sized portfolios but with Apple shares replaced with Apple bonds.
                  In the 1980-2007 period, the UK had a very sharp increase in credit growth. Much of that was debt owed between financial institutions. I thought that that credit growth was what caused a phoney illusion that the real economy was growing and our current troubles come because that debt can’t be indefinitely expanded faster than the flow of revenue required to service it.
                  Steve Keen’s equation may be in a muddle but perhaps there is a real phenomenon that he is trying to get to grips with? Perhaps change in the total stock of wealth (equity and debt) is a more useful metric than just change in debt? I’m not saying that there is a neat one to one relationship between the total stock of debt and equity increasing by 10% and people spending more on goods and services but there is some (messy) relationship. The interaction between spending on assets and spending on goods and services is very messy but that doesn’t mean that it isn’t significant. Wasn’t all of the very real unemployment etc after 2008 all due to the messy spillover from the asset trading part of the economy into the real economy? Same thing with the bubble fueled good times. It seems like prior to 2007, credit expansion pushed up asset price inflation. Then deleveraging would have reversed all of that asset price inflation except the deleveraging has been offset by expansion of government debt.

                  • No question that debt isn’t an issue, as per SK

                    I’ve made that point enough times now

                    The point in the post is about accounting, understanding, communication, and by implication education – not about the importance of understanding how debt affects the economy – its a given that its important, as far as I’m concerned

                    Regarding credit default swaps:

                    Credit default swaps are derivatives, not bonds

                    It’s important to get the cash for the bond so the stock can be bought back

                    Can’t be done with a credit derivative alone

                    • I mean – debt IS an issue – roughly along the lines of SK

                      VERY roughly

                    • JKH, thanks for the clarification. So you agree that SK is right to worry about debt but concerned that his equation is getting his worthwhile quest tangled up.

                      What I was meaning about credit default swaps is that credit default swaps enable bonds to be held safely in a very highly leveraged portfolio. If say Goldman Sachs wants to use $1B of their capital to hold lots of Apple bonds, then they will be able to own say $30B worth of bonds, borrowing $29B (from short term shadow bank lending or whatever). If they wanted to use $1B of their capital to hold lots of Apple stock, they wouldn’t be able to leverage up as much (the numbers might be off but I hope the general point is OK).

                    • Stone,

                      That’s a good way of describing my agreement/concern mix. thx

                      And I see your point on the default swaps – although Goldman may not want to swap out of all the Apple risk (they’d be buying protection) – sort of calls into question why they’d want exposure to Apple bond risk in the first place, unless its a pure arbitrage between the bond yield and the cost of the CDS, etc.

                  • Nick Edmonds says:

                    Although not really relevant to the point being discussed, you are correct that credit derivatives enabled financial institutions to carry credit risks with much less capital and that this played an important part in the crisis.

                    The development of a substantial credit derivative market facilitated a massive migration of credit risk within banks from the banking book to the trading book, where it carried a significantly lower capital requirement. The majority of losses realised in the crisis were on credit positions in the trading book.

  21. Sean Fernyhough says:

    Wonderful opening paragraphs about a book I am currently reading and wish I had time to study. Anybody out there studying economics should commit to becoming familar with Godley and Lavoie’s book.

    • thanks, and glad you are reading it

      its a book that requires considerable study – and I mean considerable

      its the best thing I’ve seen so far that brings together all of the issues you see on the blogs regarding post Keynesian points of importance as to how the system works and what its inherent logical construction is

      and then there are the simulations of economic behavior as well

  22. I hope the general idea here is correct. The details may need cleaned up.

    Let’s say I save $100,000 in currency. Someone else wants to start a new bank. They sell me a $100,000 bank stock (bank capital). The reserve requirement is 0%, and the total capital requirement is 10%. I believe that means the capital requirement is 5% for mortgages and is 10% for ordinary loans. This example will be all mortgages.

    Assets = Liabilities plus Equity

    Under early 2013 conditions (IOR = .25% and 2 year treasuries and maturities less than that yield under .25%)

    Assets new bank = $100,000 in currency
    Liabilities new bank = $0
    Equity new bank = $100,000 of bank stock

    Next, the bank will exchange the $100,000 in currency yielding 0% for central bank reserves yielding .25%.

    Assets new bank = $100,000 in central bank reserves
    Liabilities new bank = $0
    Equity new bank = $100,000 of bank stock

    The bank now makes 20 mortgages for $100,000 each. The 20 people use the demand deposits to buy 20 homes for $100,000 each from 1 home builder.

    Assets new bank = $100,000 in central bank reserves plus $2,000,000 in loans
    Liabilities new bank = $2,000,000 in demand deposits
    Equity new bank = $100,000 of bank stock

    $100,000 / ($2,000,000 * .5) = .10

    The home builder sets up a checking account at the new bank. So 20 checking accounts at the new bank were marked up and then marked down by $100,000 each, while the home builder’s checking account was marked up by the $2,000,000.

    The home builder allocates as follows:

    $1,500,000 in a savings account and $500,000 in a 7 year CD. Bank is funded for now, and the reserve requirement for savings accounts and CD’s is zero so that takes care of a positive reserve requirement.

    Assume the interest payments for mortgages are paid out as a dividend so there are no retained earnings that could be used as tier 1 capital.

    I save $50,000 more in currency. I think the bank can now sell a subordinated term debt instrument (bond) of $50,000 as tier 2 capital. It could issue up to $50,000 as tier 2 capital.

    Assets new bank = $100,000 in central bank reserves plus $2,000,000 in loans plus $50,000 in central bank reserves
    Liabilities new bank = $2,000,000 in savings account and CD’s
    Capital new bank = $100,000 of bank stock and $50,000 of bank bond(s)

    The bank now makes 10 mortgages for $100,000 each.

    Assets new bank = $100,000 in central bank reserves plus $2,000,000 in loans plus $50,000 in central bank reserves plus $1,000,000 in loans
    Liabilities new bank = $2,000,000 in savings account and CD’s plus $1,000,000 in demand deposits
    Capital new bank = $100,000 of bank stock plus $50,000 of bank bond(s)

    $150,000 / ($3,000,000 * .5) = .10

    The home builder has a checking account at the new bank. So 10 checking accounts at the new bank were marked up and then marked down by $100,000 each, while the home builder’s checking account was marked up by $1,000,000.

    I believe the $3,000,000 of the home builder is considered by some people to be lending to the bank and borrowing by the bank.

    The home builder allocates as follows:

    $1,500,000 in a savings account and $500,000 in a 7 year CD (the same) plus an additional $1,000,000 to the savings account. Bank is funded for now, and the reserve requirement for savings accounts and CD’s is zero so that takes care of a positive reserve requirement.

    Assume the interest payments for mortgages are paid out as a dividend so there are no retained earnings that could be used as tier 1 capital.

    Notice the .5 times 10% equals the capital requirement for mortgages, 5%.

    Let’s look at what happened with the $50,000 tier 2 bank capital bond that was issued.

    I saved $50,000 in currency. I lent $50,000 to the bank, and the bank borrowed $50,000 from me. The bank lent $100,000 each to 10 other people, and the 10 other people borrowed a total of $1,000,000 from the bank. Some people would say that the home builder lent $1,000,000 to the bank thru the checking account and the bank borrowed $1,000,000 from the home builder.

    The bank borrowed only $50,000 from me and lent $1,000,000 for the mortgages at one step without the bank needing to borrow from some other entity.

    Overall, lending is $50,000 plus $1,000,000 plus $1,000,000.
    **** borrowing is $50,000 plus $1,000,000 plus $1,000,000. $ borrowed = $ lent

    Next, take the capital requirement to 100% for all loans including the mortgages.

    Assets new bank = $100,000 in central bank reserves plus $100,000 in loans plus $50,000 in central bank reserves plus $50,000 in loans
    Liabilities new bank = $100,000 in savings account and CD’s plus $50,000 in savings account and CD’s
    Capital new bank = $100,000 of bank stock plus $50,000 of bank bond(s)

    The bank borrowed only $50,000 from me and only lent $50,000 for the mortgages at one step without the bank needing to borrow from some other entity.

    If the capital requirement is less than 100%, then someone can’t say monetary policy does not work by increasing actual borrowing.

  23. JKH
    “They can never prove that they are right”
    You keep saying this Steve, and its right, but its a red herring in the context of the subject of this post.

    We’ve agreed on this sufficient/necessary stuff forever with different expositions, so I hesitate to take your time, but:

    We’re both objecting to common errors we see out there, conflicts with 1. accounting logic, and 2. fundamental logic.

    1. Arguments/projections that can’t be true because their logic would result in impossible accounting imbalances.

    2. Arguments that are claimed to be true because they conform to accounting logic. Misunderstanding necessary vs. sufficient.

    #2 errors are especially egregious when the assertion of accounting correctness is also wrong (#1 error). This is Sumner et. al. (money multiplier, exogenous money, loanable funds, ISLM and supply/demand equilibrium models). They get the accounting wrong and also claim that that (wrong) accounting is sufficient proof of their correctness.

    So: “Such an attack is *both* a misrepresentation and … a lie about the nature of the argument.” (I really don’t think they’re capable of knowing that they’re misrepresenting and even lying, though. Nick in particular is an earnest and honest fellow, and very curious and self-questioning within limits, and Scott champions quite a lot of salutary and liberal policies — notably allowing more inflation.)

    PKs are more likely to only make the second error. (For example the implicit or explicit claim that govt deficits *cause* private saving, proved because the two measures are linked via accounting identities.) This is why we tend to like their thinking better, but still frequently object.

    Keen understands the necessary argument, but sometimes fails by asserting sufficiency. And he’s a great advocate for avoiding #1 errors (kudos!) in theory. Too bad he fails a lot in practice — doesn’t know when his accounting logic is wrong. His heart’s very much in the right place, but he needs to sharpen his steel-tipped pen.

    • Steve,

      Regarding your # 1, # 2 classification of errors, I’d view it a little differently:

      I’m not sure there are that many # 1’s in the sense of impossible – more in the sense of improbable. I’d put Godley’s 1990’s warnings on the projected unsustainability of the US budget surplus and private sector deficit in this category.

      And I don’t think there are that many # 2’s.

      I’d construct the logical structure of “the problem” somewhat differently. I haven’t done this before, so let me try, and perhaps tweak it later as necessary:

      Like your construction, it has two core parts.

      Let me call them A and B.

      The A error is the strategy of orthodox to attack heterodox using your # 2 as a red herring in effect – claiming like Sumner does OR at least strongly implying that heterodox “reasons from” accounting identities. That strategy is a lie – it is an attack on an argument of necessity by accusing the proponent of that argument of presenting an argument of sufficiency.

      The B error is the policy of orthodox to present theory that is stock/flow inconsistent. I think this is different again from either your # 1 or # 2. It is the presentation of an argument that is demonstrably incomplete from an accounting perspective, and that therefore can’t map to reality, a reality which CAN be represented by correct accounting. For example, the accounting balances of the multiplier (including the right time sequence) might be possible, I suppose. But they aren’t possible in the system we have – simply because the system doesn’t work that way (and by the way probably never did, even under the gold standard – which is my personal view of what I believe is a Mosler MMT misconception- but that’s veering off course here). And it’s very easy to verify that the current system doesn’t work that way. It can be verified by facts. It is not theory. The money multiplier is a theory that in fact does not exist.

      My use of the term “lie” may be misleading. I don’t mean an intentional lie in the moral sense. I mean something somewhere in the intersection of inadvertent, stubbornly obtuse, and willfully ignorant. And I use those last two harsh terms in the sense of a resistance to not look at the facts. Sumner’s famous refusal to consider banking seriously as part of his monetary theory would be representative of this. It’s the result of an intellectual stance; not a moral tendency – and it requires an intellectual judgment on the part of someone who would make such a charge (e.g. me). I don’t ‘dis’ your impressions of those people at all in that sense. But – the money multiplier is a flat out “lie” in this sense. It can be proven by the facts on the ground to be such.

      Again, I do not believe PKers in general make your # 2 error all that often. E.g. I don’t think they reason from sector financial balances too often in a blatant, UNQUALIFIED way in the sense of “this sector goes down therefore this goes up” – UNLESS it’s a two sector government/non-government model, in which case it’s a tautology. And don’t forget that MMT in particular often runs general lines of thought along that two sector approach.

      Now, we have the interesting case (I think) of Steve Keen, in which I’m going to create a new category which I’ll call error Z.

      Error Z is the error made by a proponent of the necessity argument who gets the accounting wrong. My post is about what I view as SK’s error Z.

      (Another very mild error Z is Dean Baker’s misunderstanding of the definition of some national accounts / sector balance relationships, which I noted in the comments here, and with which Ramanan agrees with and has previously blogged on.)

      And there’s an interesting “ships passing in the night” compound error happening at times here. It’s rather amusing to consider.

      Recalling the Keen/Krugman debate, Krugman engaged in a fascinating composition of truth and error:

      He called out Keen on what is a variation of error Z – the erroneous approach to accounting was the catalyst for SK’s rather unusual AD equation. I think PK was quite correct in doing this.

      But then Krugman got caught up in his own error B after moving on to more detailed discussion of endogenous money in that debate – where SK (and Fullwiler among others) basically had it right and PK was out to lunch.

      And Krugman got caught up again with error B in his debate with Steve Waldman more recently.

      Oh, it’s a tangled web we weave!

    • “PKs are more likely to only make the second error. (For example the implicit or explicit claim that govt deficits *cause* private saving, proved because the two measures are linked via accounting identities.) This is why we tend to like their thinking better, but still frequently object.”

      Steve,

      That is not true. You are reading and/or misreading some Neochartalist blogs and extrapolating it to PKs.

  24. JKH
    Re QE effect – it’s not a reduction in financial assets
    It’s a reduction in securities

    Could you be more precise for me, explain the diff between the two terms in your usage?

    “And it’s a shortening of macro duration”

    But reserves have infinite duration (rather like consols or cash…). So it’s a reduction in the inflow of “durated” securities.

    “Which means duration becomes increasingly scarce”

    ?? Can there really be a supply (stock or flow) of “duration”? Can the price of duration go up?

    “Which means bonds get bid up (other things equal)”

    Right: I say simply because of reduced flow supply. (Aside: Somebody really needs to theorize flow vs. stock “supply” properly. Have seen nothing beyond Clower’s early stab at it in a single paper.)

    All this leading to a post I’ve been meaning write:

    “Reserves and T-Bills are not eqivalents.”

    T bills have expiration dates, variable prices and interest rates. Reserves have none of these.

    I’m pretty sure reserves can’t be used as collateral in repo etc. What would you repo them for? More reserves? So removing bonds removes (?) lengthy collateral chains. Love to be corrected on this.

    “Which means wealth effect”

    And once again we agree. ;-)

    • “But reserves have infinite duration (rather like consols or cash…).”

      I’m thinking central bank reserves have 1 day duration and are constantly being renewed.

      • So what happens if the fed sold enough of the bonds it holds to reduce excess central bank reserves to zero?

    • Financial securities are a subset of financial assets.

      Regarding duration reserve – can’t recall exactly which comment had duration in it, but in the context of the dual creation of reserves and deposit liabilities in QE, consider the following for either reserves or deposit liabilities:

      Duration has to be calculated using either a contractual or an estimated term for the “term to liquidation” for the final cash flow of the interest bearing or non-interest bearing asset.

      Arguably, required reserves are longer duration, but we’re talking about QE here, which means excess reserves.

      And duration was very short term for pre-2008 excess reserves, due to the incredibly high velocity of reserves. That’s considering excess reserves as commercial bank assets. And as central bank liabilities, the aggregate size of the excess was also very volatile, sometimes even negative. So duration was short.

      And its longer term now with QE excess, but arguably shorter than the bonds that are being bought in QE.

      And reserves always trade at book value, so the level of interest rates isn’t an issue for duration of reserves.

      As far as deposits are concerned, demand deposits created by QE are reasonably high velocity as well, due to ordinary turnover in transaction use.

      Perhaps some of them are converted into bank fixed deposits with longer duration, but the fixed term will normally be less than bonds, so there’s still a duration pick up.

      So there’s definitely duration shortening due to QE on both sides of commercial bank balance sheets – and also the in the portfolios of the non-bank sellers of bonds into QE.

      BTW, please refrain from invoking the hack version of “reserves as equity” or “currency as equity” that’s common place among academic economists who don’t look at the facts of bank asset liability management. That’s hogwash, right up there with the other neo-classical billboard of errors. It’s not sound accounting or finance.

      Yes, we always agree … gulp

      :)

      • Steve Roth’s post: “JKH
        Re QE effect – it’s not a reduction in financial assets
        It’s a reduction in securities”

        and JKH’s post: “Financial securities are a subset of financial assets.”

        I’m confused. Doesn’t that mean there is a reduction in both financial securities and financial assets?

        • QE swaps reserves/deposits for bonds

          looking at the private sector consolidated, there’s no reduction in financial assets – reserve holdings of banks replace bond holdings of mostly non-banks (perhaps some banks also) in that consolidated view

          but there is a reduction in securities – bonds

          and an increase in ‘non-securities’ – reserves

          both reserves and bonds are financial assets

          bonds are a subset of all financial assets

          as are reserves

      • “Financial securities are a subset of financial assets.”

        K I’ll take a stab. Securities are financial assets which are:

        1. Commonly traded in financial markets.

        2. Subject to nominal value change, determined by market forces (see #1).

        3. Packaged in large-dollar units. (You can’t buy $1 in bonds.)

        So bank/MM deposits and fed reserve balances aren’t securities, but they are financial assets.

        I think this a useful if slightly loose term/category description when we want to refer to a class of assets without using all the words above.

        So the Fed is reducing the net inflow quantity of securities to the financial markets, raising the price of securities. Is QE really that simple?

        • that looks OK as definition

          don’t know whether or not its “that simple” but seems OK as a short form depiction

    • “But reserves have infinite duration (rather like consols or cash…). So it’s a reduction in the inflow of “durated” securities.”

      In standard understanding reserves and cash have zero duration.

      Consols do not have infinite duration. Just because they are perpetual doesn’t mean their duration is infinite.

  25. JKH
    Oh, it’s a tangled web we weave!

    I’m not totally happy with either my or your taxonomy of errors, but like them both. Rather like obscenity: you know it when you see it!

  26. JKH and Nick Edmonds
    also, note that your credit card statement is in effect a specialized personal report that consolidates the standard separate financial accounting perspectives of spending (income/NIPA) and borrowing (sources and uses of funds/flow of funds)

    Great explanation. I’ve been thinking very hard about this credit-card business vis a vis Singh on collateral chains, “velocity of pledged collateral,” etc. (Spending up your credit-card balance is effectively pledging your future work/earnings as collateral.)

    In particular: if M in MV=PT (or Q, or Y…) is money available to spend, do our collective unused credit-card limits count as part of M? Just further pointing out the incoherence of monetarism…

    • yes

      and along with the incoherence of monetarism, let me join in and point out that the ocean is wet

      :)

    • it is a very good example though of the fragility/delicacy of the M and V of M ideas – as they relate to credit card transactions

    • “unused credit-card limits count as part of M?”

      I’m going to say no.

      The real problem is I’m pretty sure monetarism means M = monetary base = currency plus central bank reserves. I think M = currency plus demand deposits. That is the point of my notebook example above.

    • beowulf says:

      “do our collective unused credit-card limits count as part of M?”

      The Fed could order banks to cut all unused credit card limits in half. On the other hand, it’d be a violation of the Takings Clause to order banks to seize half of all demand deposits.

      I believe it was JKH’s nemesis RSJ who made the observation that MZM (M2 less the time deposits, plus all money market funds) is usually in sync with publicly held debt and when the two get out of whack, something is amiss.
      http://tinyurl.com/lglkm8k

      • mild but recurring irritant (by citation) more than nemesis, I think

        or is there a legal definition of nemesis I’m not familiar with?

        doesn’t matter; I’m building up immunity to the frequency

        :)

        • beowulf says:

          I had originally put arch-nemesis but figured that was too strong.
          The link is to a FRED chart showing MZM and debt held by public (includes Fed-owned debt, excludes govt trust fund-owned) are presently the same size and trendline.

          • “MZM and debt held by public … are presently the same size and trendline.”

            That’s quite fascinating. Are you suggesting an MMT explanation — that the stock of money = cumulative govt debt? Or…?

      • “The Fed could order banks to cut all unused credit card limits in half. On the other hand, it’d be a violation of the Takings Clause to order banks to seize half of all demand deposits.”

        Aren’t the demand deposits issued when the credit card is used?

      • Sorry what is this discussion of the MZM with public debt?

        Interesting graph but there isn’t really anything more to it.

        If that is incorrect – what is claim?

        • Long story about left winger RSJ who Beowulf thinks is my “nemesis” for some reason. It’s the old idea about carving out bank deposit margins and handing that part of the banking business over to the government – as per the Chicago Plan – so the government can grab “the seigniorage”.

          The reason I had some impatience with the RSJ model is not because it is a figment of the normal left wing frivolity toward bank operating costs that it refuses to acknowledge as existing, but because it is completely unaware of the fact that banks routinely do the type of interest margin partitioning that is implicit in such proposals as a matter of risk transfer and pricing – everything RSJ says is already done by the banks themselves – there nothing new there.

          • beowulf says:

            The link isn’t to an RSJ article, its a FRED chart showing relationship between MZM and publicly held debt (ex intergovernmental) over several decades. I cited RSJ because he’s the one who first pointed this out. I called him JKH’s nemesis only because of the abreaction the name induces.

            If I was arguing for RSJ’s position, I’d have started by making an argument. Instead my goal was more modest, to link to a chart I thought was interesting.

            • I cited RSJ because you cited him – regarding the chart – not because you were making an argument.

              In an attempt to assist with full context for the answer to “anything more to it” – as in the original application of the chart to an argument – which I summarized.

    • Nick Edmonds says:

      A good point and one that I have been making to colleagues for some time now. Many payments in a modern economy (e.g. credit card purchases, payments received into an overdrawn account) are not actually transfers of money, since the money is created or destroyed in the process. Some do not involve money at all (such as an overdraft to overdraft payment). In some transactions, there is no way of telling whether they involve money or not. In my view, the quantity theory of money cannot cope with this.

      • Right. I really think we need to stop using the word “money” in technical discussions, instead always using specific technical terms of art like “bank deposits,” “reserve balances,” “bonds,” “equity securities,” etc. “Money” just has too many possible definitions. Using it just obfuscates.

        Using category words like “securities” is sensible, as long as we’re quite sure that we all know what those categories comprise (at least fairly specifically).

        Saying “you need to define what you mean by money” in every discussion is not a useful answer. It makes discussion impossibly unwieldy, at least with disparate dicusssants who aren’t on the same page, “money”-wise.

        • Nick Edmonds says:

          Totally agree.

          I’ve made exactly the same point to others on several occasions.

  27. JKH, an FYI that might interest you:

    Marc Lavoie is at the 2013 IGLP Pro-Seminar 5 at Harvard this week with Stephanie Kelton. http://www.harvardiglp.org/iglp-the-workshop/pro-seminars/

    Lavoie told her and she tweeted: “Reading Warren Mosler’s Soft Currency Economics, I first thought he was crazy. Then I realized he was right.” ~Marc Lavoie
    https://twitter.com/deficitowl/status/342352501200277505

    • Soft Currency Economics seems to have been written pre-IOR, still assumes that the Fed sets the funds rate by managing reserve levels, as opposed to the current corridor system where the Fed funds rate is managed much more directly and explicitly via IOR and discount-window rates. Fed balance-sheet moves have entered a whole new realm/sphere of influence. (Can you say “wealth effect”?)

    • BTW, I catapulted to Mosler’s site to have another quick look at Soft Currency Economics, and it no longer seems to be on-line. Looks like you have to get the kindle or the paperback now.

      Has the guy suddenly become a capitalist again?

      Looks like a seigniorage play to me.

      :)

      Anybody know if I can get it for 0 cents anywhere?

  28. Nick,

    Yes Keen seems to be mixing dimensions and I prematurely said he isn’t so. I think the Slackwire blog caught him at an incorrect point but the blogger’s claim seems right but explanation wrong.

    Now models in PKE such as G&L use discrete time and things are easy. Keen arrives on the scene and asserts continuous time is somehow superior and discrete time inferior.

    But he doesn’t do continuous time modeling right. In continuous time formulation you don’t worry about discrete events and model everything smoothly. That is I am consuming $x/day – as if I am consuming an apple in the middle of my sleep :-). Which isn’t wrong because it is a model and such things shouldn’t matter much.

    But Keen seems to be using *both* smoothing and discrete events. Either he should use income and debt both as discrete events and integrate them mathematically over a time period for it to make it look like an official statistical release. But he has income smooth (except jumps) i.e., piecewise continuous in mathematical language but debt events discrete.

    Strange.

    The bad thing about Keen’s approach is that he blames his confusions as others’ problems.

    • “Either” .. the “Or” for the either is … Or he should use everything smooth.

    • Nick Edmonds says:

      I think that’s very well put.

      I find it very odd that he seems to think he can derive results in continuous time that are qualitatively different to what you’d get doing discrete analysis.

    • Ramanan,

      I would have thought you can integrate either discrete events or continuous events (or both) in continuous time – with piecewise continuous results in the case of discrete events.

      e.g. you can integrate the accrual of interest as a continuous event – until its paid out as a cash flow, which is a discrete event – all in continuous time

      SK’s error is that due to definitional and accounting errors, he’s attempting to integrate something as a function of itself – i.e. calculating aggregate demand with income in the integrand – while carelessly adding on something which is just not a component of either aggregate demand or income (change in debt) when viewed in the context of stock/flow consistent accounting recognition and basic definitions of income and debt

      If you’re integrating something, the stuff you’re integrating has to be a component part of the result – and it can’t be that if its incompatible on purely definitional (i.e. accounting) grounds

      And just because monetary/real/accounting events are simultaneous or perceived as simultaneous – e.g. issuing debt and spending (including credit cards BTW) – that doesn’t make them the same event

      which is to say he’s doing regression at best

      • and its a poor regression from a purely logical perspective, given the propensity for ‘leakages’ from the issuance of debt into asset acquisition rather than spending

      • Yeah!

        Accrual leads to some flow being piecewise continuous instead of being spikey.

        So interest is accrued and is not spikey. Even salaries can be thought of like that.

        But people do impulse purchases and stuff which are spikey.

    • btw, you saw that some sort of presentation of SK at Harvard is up at MNE?

      kind of a weird thing of a video though:

      http://mikenormaneconomics.blogspot.ca/2013/06/steve-keen-keen-harvard-seminar.html

      • Yeah. I think he’s just spread too far, doesn’t have time to do anything as well as he could. (He created the minsky model that he demoed there during the session?) Losing his U department/job could be part of that. Irregular as he is, I’d like to see some rich patron give him a solid place to stand, no need to earn, a comfortable office and home somewhere, and see what he comes up with over the next decade or two…

        Do you guys think that Minsky could be as significant as I do?

        Oh on that, JKH, your dissing of his “Godley table” usage is I think misplaced. Just cause his Godley table is mal- or ill-formed does not mean that the software construct is badly named.

        • “your dissing of his “Godley table” usage is I think misplaced”

          Fair enough.

          I shouldn’t presume to know what Godley might have thought of it or whatever in relation to the naming of it.

      • Yeah seems to end abruptly.

        By the way his point about Obama and banks – I am not sure. Did you ever comment on his blog on this? Sort of recall but I may be wrong.

        • sorry, can you remind me of the Obama/bank point?

        • Keen points to some interview of Obama in which he is asked “why bailout for banks, where is my bailout” and according to Keen he replied saying each dollar in the bank leads to higher lending.

          http://www.debtdeflation.com/blogs/2009/09/19/it%E2%80%99s-hard-being-a-bear-part-five-rescued/

          “the truth is that a dol­lar of cap­i­tal in a bank can actu­ally result in eight or ten dol­lars of loans to fam­i­lies and busi­nesses, a mul­ti­plier effect that can ulti­mately lead to a faster pace of eco­nomic growth. (page 3 of the speech)”

          Old post but he says that in the recent video as well.

          Keen then goes on to say Obama believes in the money multiplier or his expert advisors told him the story … or something like that.

          Now this is funny IMO. Banks were not well capitalized during the crisis and the US Treasury capitalized them, so that they could lend more. Without the capitalization, they may have reduced lending and would have caused more pains which has nothing to do with the money multiplier which Keen confuses.

          • You’re exactly right

            (I think I’ve made a similar point somewhere)

            Keen is misidentifying an error

            The fact is that Treasury was injecting capital, and any associated reserve effect was a byproduct

            The objective was stated and executed clearly as capital – not reserves

            TGA management was pretty messy at that point – might have been a net reserve injection of about the same size at the same time – although I tend to doubt it

            More to the point, its quite conceivable that Summers explained the desired effect to Obama as a capital multiplier, in effect – which is a perfectly legitimate way of viewing the mechanics of capital deployment – and not as a reserve multiplier

            And its difficult to see how it would not have been explained that way, given that the request was for capital and not for reserves

            In this case, largely to prevent a contraction due to previous capital losses, in addition to praying for some resumption of more normal normal bank lending based on replenished capital adequacy

            • :)

            • I’d like to confirm my understanding of capital/reserves, and perhaps thereby explain Obama’s confusion.

              A bank has a big pot of liquid assets: bonds, reserves, currency/coins. (I’ll exclude collaterizable/securitizable loans for the moment.)

              You can point to any chunk of those holdings and say, “that’s the bank’s capital” or “that’s the bank’s required reserves.” But the statement is arbitrary. Those holdings/assets are essentially fungible in an accounting sense (and in a real sense; the bank could trade all its excess fed reserves with other banks in return for bonds).

              You can’t really say, “that vault cash is the bank’s capital” (or reserves or whatever).

              Now in the course of accounting some of those specific assets may get “locked down” into particular (capital, reserve, etc.) accounts via the posting of journal entries to particular balance-sheet accounts. But viewed from on high, there’s just a big pot of assets, some proportion of which is reserves, some capital, etc.

              So when Treasury/Fed transferred reserves to the banks’ accounts, it’s all a matter of where those journal entries get posted. In this case they were posted to the banks’ capital accounts (with the associated liabilities somehow hidden from those capital accounts?)

              The banks’ reserve balances necessarily increased, but not the banks’ “reserves.” (This is a very widespread source of confusion.)

              That right?

              • no

                capital is an entry on the right hand side of the balance sheet

                period

                its purpose is to absorb unexpected losses

                which happens when there is an unexpected negative number coming out of the income statement

                that number gets charged (debited) to capital, and capital contracts

                within banks, capital is allocated as a source of internal funds, to various businesses, based on the amount of capital that must be attributed to those businesses according to risk weightings, so that a portion of the bank’s total capital is available to absorb losses, business by business

                and I feel quite certain that there are a whole lot more people confused about the role of capital than Obama was

              • Steve,

                No disrespect, and it may be just the bank context, but have you taken a formal course in financial accounting?

                If you have, then it may well be the bank context that’s the issue.

                But if you haven’t, then I would URGE you to find a good MBA program and sit in on the first year financial accounting course.

                It would be perfect for you, as you are ripe for absorbing it, based on the kinds of questions you’re looking for answers to.

                Same advice as I would give to Steve Keen, who I’m certain hasn’t taken such a course.

                If the prof is any good, he/she would be able to help expand it to banks and to macro NIPA and flow of funds.

                • Good suggestion, but for now: I don’t understand how what you said conflicts with what I said. It seems like they’re different explanations, but not contradictory. ??

                  • I’m wondering the same thing. These numbers probably won’t be exact, but the point should stand.

                    The capital requirement is 5% for mortgages, and the reserve requirement is 10%.

                    Assets of new bank = $1,800,000 in mortgage loans and $300,000 in gov’t bonds

                    Liabilities of new bank = $2,000,000 in demand deposits

                    The capital requirement and reserve requirement are met, right?

                  • They’re totally contradictory – as in – you’re wrong.

                    “You can point to any chunk of those holdings and say, “that’s the bank’s capital” or “that’s the bank’s required reserves.” But the statement is arbitrary.”

                    Totally incorrect.

                    I just said – capital is on the right hand side of the balance sheet.

                    And its not arbitrary – there are definitions of what it is – according to Basle, local government regulations, etc. And common equity capital is always the core, whatever the refinements are into preferred equity, etc.

                    Risk assets are on the left hand side of the balance sheet.

                    Capital on the right.

                    Distinct – assets from capital, left from right.

                    It’s accounting – plus add on knowledge – with correct accounting as the required foundation – as per the theme of the post.

                    Do that accounting course. You probably have some good academic contacts. Use them. Contact them. Go for it. Really. Don’t procrastinate. It has to be a full year, 1st year MBA course. Nothing short of that. A full year. Discipline. Work. No substitute. You’ll thank me. Believe it.

                    • Isn’t bank capital “stored” as an asset somewhere?

                      Example, I start a bank in Spain.

                      Sell 100,000 stock and buy Spanish bonds of the gov’t.

                      Assets = 100,000 in Spanish bonds
                      Liabilities = 0
                      Equity = 100,000 bank stock

                      The next day the gov’t is forced to give 10% “haircuts” on its bonds.

                      Is the balance sheet now?:

                      Assets = 90,000 in Spanish bonds
                      Liabilities = 0
                      Equity = 90,000 bank stock

                    • “Isn’t bank capital “stored” as an asset somewhere?”

                      No that’s the whole point that I’ve known forever, but that I’ve just had drummed into me yet again, some more, quite rigorously:

                      Shareholder’s equity, aka capital, is just a residual: assets minus liabilities. (Basically net assets.)

                      You can’t point to any particular assets that are “capital” (or “required reserves” for that matter).

                      If assets increase or decrease (due to income or market-price changes of the assets), capital goes up or down.

                      If liabilities increase or decrease, capital goes up or down.

                      If those asset/liability changes cause capital to decline to the point that it’s not sufficient under regulations, or it’s gone completely, that’s spelled T-R-O-U-B-L-E.

                      Even if there’s still a big pile of assets.

              • “I’d like to confirm my understanding of capital/reserves, and perhaps thereby explain Obama’s confusion.”

                Steve,

                It is not Obama who is confused but Keen. Obama may be confused in other places but that’s not under discussion.

                When a bank gets capitalized by the Treasury, it sells new shares to the Treasury and receives reserves in exchange but as JKH mentioned this is the byproduct of the transaction. You have to look at the right hand side of the balance sheet to understand the improved capital position.

            • I’m so doubtful about the capital multiplier effect supporting the real economy. I wonder whether the real economy wouldn’t have been much better served if instead of recapitalizing the banks, that money had been transferred directly to bail out customers of banks, pension funds and insurance companies that had instead been left to go bust. Over 80% of bank lending is for purchasing pre-existing housing stock. Much of the rest is for purchasing other types of pre-existing assets. Recapitalizing banks looks to me more about preserving the bloated size of the financial sector and inflated asset prices -not about the real economy.
              I had a go posting about that http://directeconomicdemocracy.wordpress.com/2013/04/28/bail-out-the-customers-not-the-banks/

              • Stone,

                I also don’t believe in the capital multiplier story but the situation in 2008 was really bad and banks would have tried to severely contract lending leading to a more fall in activity and a worsening of their own capital and a negative feedback loop like that. They needed to be capitalized. So with the credit crunch scenario there is a kind of multiplier effect due to the help from the Treasury. Not true in general.

                • Ramanan, I just wonder whether even if bank capital had dropped by say 90%, then the lending that mattered to the real economy might not have been constrained. Let’s guess that say 5% of bank lending was to small non-financial businesses (large businesses are not reliant on bank lending). That lending could have still happened even if banks were much smaller. That lending didn’t happen even after the banks were re-capitalized because the banks are wary of lending because of lack of aggregate demand. In the UK if RBS and HBOS had been left to entirely evaporate, entities such as say Tesco’s Bank etc were in place to take over.
                  After WWII Germany had an economic boom and that was after massive debt write downs. Perhaps if 2008 and been left to unfold with massive massive debt defaults, the stage would have been set for a global renaissance????

                  • Stone,

                    “In the UK if RBS and HBOS had been left to entirely evaporate, entities such as say Tesco’s Bank etc were in place to take over.”

                    I am really not an endorser of idea of letting banks fail in 2008. It is true that the doctrine of too big to let fail is bad but the situation was so bad at the time that some action was needed.

                    What you are saying is like saying that the fire in the neighbour’s house won’t reach my house.

              • The point there was about preventing a severe contraction and an even worse stock market meltdown if bank capital losses had not been replenished. Banks need capital or the market will take the stocks to zero – as with Lehman, near-zero with Bear Stearns, etc. etc. Its a technical fact regarding market response to an actual balance sheet condition, not an ideological statement about how much bank lending can goose the economy at a particularly bad point in time.

                Referring to the “capital multiplier” is undesirable form. It’s merely a mathematical relationship between the amount of capital and the amount of risk assets that capital can support. The only reason to use the term multiplier, and only in a loose descriptive way, is to contrast a correct ex ante relationship – you have to have capital to lend – with an incorrect ex ante relationship – the false idea that more reserves allows more risk lending.

                • I really think it is important not to dismiss out of hand the idea that the economy would now be much better off if most of banks had gone the way of Lehmans in 2008. After WWII, Germany had an economic boom. They had had the ledger wiped clean. Overall the post WWII economic golden age across the world looks to me to have been largely a result of all the waste and destruction having rebooted the system. I don’t see why it is considered so unthinkable that we need the ledger wiped clean today. It is a tragedy that we seem to need a world war to force it rather than just letting the financial system slough off excess debt by way of defaults and closure of excess financial institutions.
                  Ramanan, I don’t think burning down a house is a good analogy. We are talking about ledger entries not life and death or even shelter. The government could have bailed out citizens directly rather than attempting to do so via the banks. In fact as it stands there are people unemployed today due to economic depression. Perhaps if money had not gone to the banks but instead had gone directly to bank customers, then those people would now have jobs???

                  • Stone,

                    Good point. Two different debates. One whether banks should have been rescued and second about pure monetary economics.

                    Obama avoids the question by saying capitalizing the bank is better or so. I agree the argument is incorrect. Obama should have also used fiscal methods to boost demand.

                    But if Keen wanted to criticize he should be criticising this point rather than saying Obama thinks in terms of the money multiplier. And Keen ends up confusing capital and reserves. Disaster!

                    • right

                      although you might argue that Obama had much greater operational leverage to convey the urgency of a decision in the case of bank capital than he would have had and still does not have in the case of fiscal

                      fiscal by the nature of the political and economic process is a slow thing by comparison to actual monetary decisions

                      bank recapitalization by the state is pretty much a binary decision compared to endless debate on potential fiscal variations

                      the stock market is a much more immediate reactor than the response of the economy to fiscal

                    • i.e. note that the capital injections were basically monetary – not fiscal – asset swaps of bank capital for risk free assets

                      And the technical outcome at the margin of the capital injection has actually been a positive “margin” in the sense of a positive return on capital – very analogous to the way in which the Fed has increased return on its own assets by extending duration through QE

                      Both have resulted in returns from greater risk taken by way of asset swaps

                  • By “money”, do you mean debt?

              • “I’m so doubtful about the capital multiplier effect supporting the real economy.”

                It seems to me the capital multiplier is how more demand deposits (medium of account [MOA] and medium of exchange [MOE]) are created.

          • beowulf says:

            “Now this is funny IMO. Banks were not well capitalized during the crisis and the US Treasury capitalized them, so that they could lend more.”

            “Well capitalized” is a relative term. Instead of injecting capital into banks to re-capitalize them, the govt could have just redefined “well capitalized” without even changing the ratios. The govt can create regulatory capital almost as easily it can write a check.

            Of course this would almost certainly trip over the Prompt Corrective Action Act, but you’d think Tsy capital injections & creative forbearance (e.g. “stress tests”) would as well. William Black has been going bonkers on this issue for years.
            http://dailybail.com/home/bill-black-the-banks-are-still-insolvent-and-obama-is-not-on.html

            • That’s another one of Mosler’s gimmicks and its nonsense.

              The private sector wants to see identified capital there to absorb losses.

              That’s the only way the private sector itself can be encouraged to join in later on to participate in subsequent capital issues.

              Mosler’s gimmick is just nationalization – abandon all norms of capital allocation, and just go with the random government backstop.

              It’s the usual MMT inference of nationalization, and its just lazy thinking about resolving the problem on a permanent basis.

            • Beowulf,

              That doesn’t work and looks illegal as well. The government also has to abide by the laws.

              Even if it is somehow allowed, one can perhaps create scenarios – many banks with different numbers and show something incorrect happens or can happen.

              Let us say one bank Citi has to be capitalized and others are fine. The government simply gives a forbearance? That gives other banks an excuse “eff you I am taking risks and will look for forbearance”.

              The think about capitalization may have to do with the fact that the Treasury makes money on the stocks – gets higher dividends etc. This is a disincentive to other banks to look for forbearance.

              I am reasonably sure if one looks for more loopholes one can find at least 10.

              Your point – which I guess is also Mosler’s point is simple handwaving. He seems to think “oh this is so easy for the Treasury, just a few steps which takes a few seconds” type of thing.

              • beowulf says:

                I’m not disagreeing with you guys. I even mentioned it would violate the Prompt Corrective Action Act. I’m simply agreeing with Bill Black that the actual steps the Fed & Tsy did take weren’t all that more legit.

                As for me, I’m with Sheila Bair (and the commentator from Airplane) on this.
                http://www.youtube.com/watch?v=Pn0WdJx-Wkw

          • “the truth is that a dol­lar of cap­i­tal in a bank can actu­ally result in eight or ten dol­lars of loans to fam­i­lies and busi­nesses, a mul­ti­plier effect that can ulti­mately lead to a faster pace of eco­nomic growth. (page 3 of the speech)”

            The problem with that is that the solution to too much debt is not more debt.

  29. All of this returns us to Keen’s demand/debt bizness.

    His fundamental point is rather obvious and uncontreversial: more borrowing by the real sector from the financial sector can/will increase demand for real goods (though not 1:1 because of securities purchases/sales).

    I think the problem might lie in the notion of “demand.” And I think Nick Rowe may be right here (I think he said this somewhere), that it’s kind of dumb to talk about “increased demand”; it’s kind of meaningless. (Especially, I would add, as nobody seems to agree — except among particular groups and “schools” — on what “effective” and “aggregate” demand mean.)

    One way to say something that is substantive is to instead talk about the demand curve changing/moving.

    Viz: increased borrowing by the real sector (an identifiable and quantifiable accounting “event”) will move the demand curve for real goods up and/or right, increasing prices and/or production. And it might also change the slope of the demand curve.

    “Economics” (as opposed to accounting) is all about determining which of those happens, and what interrelated chains of cause and effect are at play in that happening.

    But the key point is important, even if sloppily stated by Keen in English, economic, or accounting terms: increased borrowing will result in “increased demand.” (With liabilities generated, which retain power to affect future demand.)

  30. If the real sector borrows (with the intention of spending) at 11:59pm on Dec. 31, 2013, demand in 2014 will be higher than it would have been absent that borrowing. (Well, unless the real sector changes its mind and instead pays off the debt in 2014.)

    • That’s right.

      Now please inform SK.

      Because that’s not what his equation says.

  31. Luis Enrique says:

    I need to properly read Godley …. and the idea that a detailed picture of the accounts of appropriately chosen agents/sectors of economic activity is a useful constraint on our thinking, seems uncontroversial to me. As my PhD supervisor put it, narrowing down the range of plausible hypothesis is v useful. Of course all manner of behaviour is still compatible with accounting constraints, and it’s what Godley et al do about behavioural assumptions that I’d be most interested to learn.

    with that preamble. isn’t there a parallel with national accounts, which economists certainly do make ample use of. Example:
    http://stumblingandmumbling.typepad.com/stumbling_and_mumbling/2013/03/on-financial-balances.html
    plus all those economists who work on international macro and worry about unsustainable trade deficits etc. they look to me to be doing the same thing Godley does (“this flow can’t go on forever”). So perhaps that approach isn’t as alien to economists as you think.

    of course plenty of mainstream econ models are stock flow consistent, just very simplified. I’d be interested to see examples of mainstream economic models that predict the impossible because if you drilled down they’d violate some accounting constraint. NB I’m not interested in banking examples, I’ve had enough of that.

    • It should be uncontroversial – unless it can be demonstrated that its been violated.

      You should really have a close look at Godley/Lavoie. What they demonstrate is that the list of models I provided, money multiplier, loanable funds, ISLM, supply/demand equilibrium … are incompatible with accounting closure in the real world.

      Mundell-Fleming is another big one that violates accounting closure, and very interesting. The fixed/floating bifurcation that is the anchor for Moslerian MMT therefore becomes quite questionable.

      • Luis Enrique says:

        OK. well there’s only one way to find out, but what you write doesn’t fill me with optimism. The money multiplier I am familiar with is derived by rearranging accounting identities. The idea that supply/demand equilibria somehow violates accounting constraints looks like nonsense to me.

        • perhaps (supply/demand) – have a look though and see what you think – its in the book

        • if you get the chance, would be interested to know how you think he goes wrong on that, or if I’ve misinterpreted what he’s saying in effect

        • Nick Edmonds says:

          I’m not sure that the money multiplier or models that involve supply/demand equilibria are necessarily inconsistent with accounting constraints. It’s true that they might be used in a way that is inconsistent, but equally I can’t see any reason why they can’t be constructed in a way that is consistent. Even models like IS/LM, I’d tend to think of as ignoring the stock-flow dynamics, rather than actually being inconsistent.

          • The money multiplier is inconsistent in the sense that the sequence of accounting entries that is implied by the normally interpreted money multiplier story is wrong. The sequence is wrong. Therefore the accounting logic is wrong, even though the final appearance of the accounting may look the same. The accounting process is wrong. That’s what it means in context.

  32. Nick Edmonds says:

    Steve Roth
    The banks’ reserve balances necessarily increased, but not the banks’ “reserves.” (This is a very widespread source of confusion.)

    I’m not familiar with how reserve requirements work in the US. Is there are difference between reserve balances and reserves?

    • I think he is talking of “capital reserves” when he mentioned “reserves” as opposed to “reserve balances”. But I still like to know why he said that comment.

      • Nick Edmonds says:

        By “capital reserves” do you mean regulatory capital?

        • Nick,

          I am not sure if the phrase “capital reserves” has an unambigious definition. I should have probably said Steve Roth thinks the phrase reserves should always be in the context of bank capital and reserve balances in the context of deposits at the central bank.

          IMO, it is okay to use the word reserves in the context of banks’ deposits at the central bank because this is common usage. Perhaps the word is misleading because a bank can lose reserves when customers shift funds to another bank and it can easily obtain reserves from the central bank.

          Steve Roth should have instead said “ban the phrase reserves” rather than saying reserve balances are different from reserves because the latter is used commonly even in official language.

    • “So when Treasury/Fed transferred reserves to the banks’ accounts, it’s all a matter of where those journal entries get posted. In this case they were posted to the banks’ capital accounts (with the associated liabilities somehow hidden from those capital accounts?) The banks’ reserve balances necessarily increased, but not the banks’ “reserves.” (This is a very widespread source of confusion.)”

      Wrong.

      Reserves don’t get “posted to capital accounts”. That’s meaningless from an accounting logic standpoint in general.

      “associated liabilities hidden from capital accounts” is also meaningless from an accounting logic standpoint in general.

      Reserves left; capital right.

      Capital right is allocated to risk assets left according to bank internal flow of funds arrangements. That’s the purpose of capital. That’s how its managed.

      In fact, reserves as a risk free asset require zero capital.

      And none of it is “posting”.

      It’s an internal funds transfer system – not financial account posting.

      If the overall capital pool is short because of excessive losses, it has to be made up in order for the required internal capital allocation to be matched with actual capital in aggregate. The short position in required capital has to be covered by external capital raising.

      Confusion is widespread indeed.

      • Nick Edmonds says:

        Is it normal for banks to have an actual internal flow of funds for capital allocation? I’ve seen banks make a specially identified loan to foreign branches representing capital allocation, in order to help identify income attribution for tax purposes, but otherwise I didn’t think that generally happened.

        • its the way banks work

          • Nick Edmonds says:

            Well definitely not all banks, because I’ve worked for some and that’s not how it was done. What I’ve seen is capital allocations operating like credit or position limits, not involving any funds transfer, with net asset positions being 100% funded by loans from central treasury operations or the money market desk. And I’m talking about global departments, not just single desks.

            It’s possible someone in a back office was making some adjustments somewhere to shift capital around and that I was unaware of it, but I think that’s unlikely.

            However, I’ve no idea how it is managed at other banks, which is why I asked the question.

            • There’s definitely a funds transfer of capital on the internal books. One way of doing it is for capital in respect of each business line to fund a risk free asset position in the amount of the capital attribution, with the loans being matched funded for interest rate risk, but not for capital assignment. You probably saw only the latter piece. The combined margin effect then determines the ROE for the business. But the net effect is that there is a return on capital for each business just as if that business were a stand alone bank with its own capital. All large banks do it that way. They have to in order to determine ROE for each business line.

              (You can sort of relate that mechanism to the Sharpe model, or whatever that thing is called)

              • Nick Edmonds says:

                Interesting.

                I’m familiar with the calculation you describe, but I always assumed that this was made as a notional adjustment (after all, it’s a fairly straightforward adjustment) rather than anyone actually moving money (apart from the foreign branch example).

                • Since you’re interested, let me try a highly simplified example that starts to decompose the pieces:

                  One way to illustrate it is to imagine a bank with one asset business, but with the same generic internal funds transfer system.

                  The bank has loans L, deposits D, and equity capital E.

                  Assume all the capital is assigned to the risk lending function.

                  (In real life, capital is assigned for all sorts of purposes, including capital assignment for deposit businesses, but lets assume that away.)

                  L would be funded 100 per cent on a matched basis from Treasury, for purposes of interest margin determination and interest rate risk management by Treasury. That’s what you’ve referred to.

                  Suppose it earns Li and pays Fi for a margin of Mi.

                  But its funding in fact is composed of capital funding and non-capital funding. It has to be – because it’s the only asset business and therefore it can’t be funded entirely by non-capital funding. That would be incompatible with the bank’s actual funding structure.

                  On the non-capital funding, that’s the end of the story – it earns Mi on that piece.

                  On the capital funding, that’s not the end of the story.

                  The business is allocated capital of E.

                  So in addition to Mi, it earns Fi on its equity E.

                  (In fact, treasury will swap the interest rate risk with respect to the delivered term structure of Fi into some other profile that is risk free, not only in terms of credit risk, but in terms of interest rate risk. That’s a detail of interest rate risk management, which is managed centrally by the Treasury function. So I’ll assume that away and assumed that the delivered risk of the rate Fi is the rate that Treasury actually desires in respect of equity funding for the bank as a whole.)

                  So the business is funded by E capital funding + (L – E) non-capital funding.

                  (Note (L – E) in this simplified example is the actual deposit funding of the bank D)

                  The total return for the lending business is Mi + (Fi times E).

                  And it calculates its ROE accordingly.

                  (And it goes through a similar process of interest margin determination for the deposit business, centralizing any interest rate risk management in a similar way. Although I’ve excluded capital allocation for the deposit business to simplify.)

                  • Nick Edmonds says:

                    I am familiar with how ROE is calculated; but I was just thinking a bit more about the distinction between internal funds transfer and notional adjustments, but this is probably down to some confusion on my part.

                    I’m a little bit wary about saying anything on specifics, but I wondered if you could expand on one point for me, if it’s not too much trouble. I’m interested in what equity measure most banks use in the management accounts for the calculation you describe. In particular, do they use the actual equity of the bank, in which case, how do they allocate that between different business areas? Is that pro-rata to actual usage or some other method? Alternatively, do they use the actual regulatory capital usage of the business area, in which case how do they deal with the difference between regulatory capital used and actual equity?

                • P.S.

                  “moving money” in an internal funds transfer system is much broader than the legal booking of funds flows – legal booking is for required legal reasons – internal funding is for risk management purposes – the two are compatible in that internal funds transfer is a comprehensive, unified risk management overlay that “looks through” the peculiarity of legal booking points

                • @Nick: “I always assumed that this was made as a notional adjustment (after all, it’s a fairly straightforward adjustment) rather than anyone actually moving money (apart from the foreign branch example).”

                  Can’t resist going meta here:

                  Reallocating funds between accounts is “moving money” in a very real — perhaps the only real — sense (dollar bills being mere physical tokens of entries on those actual or sometimes only implicit tally sheets). A transfer to a foreign branch is a re-allocation between accounts, with one account happening to be external or semi-external. A bank can move money internally without any transfers between external deposit accounts.

                  That devil “money” word again…

                  • I view it as a rather glorious word

                    Too sublime to be defined

                    :)

                  • Sorry – only just noticed this comment, but (at the risk of opening up a can of worms again) I thought I’d just explain what I meant.

                    I know how banks can carry out internal accounting by allocating amounts between different profit / cost centres, and that this does not normally involve any movement through an account with an external party. It’s just internal record keeping and although (as you say) it does not involve “money” in the sense of conventionally defined broad money, it is as real as any monetary transfer can be in a modern economy.

                    However, the thing I was trying to distinguish is this. If you want to know the return on capital of a business, particularly if it’s based on actual usage, you don’t actually need to record transfer of capital to do it. You just divide recorded income by the product of risk-weighted assets and the required capital ratio, and then add the designated risk-free rate. That’s what I meant by making a notional adjustment.

                    The point about foreign branch capital is that it’s not there to calculate ROE (and in fact ROE may be calculated on a different basis); it’s there so the bank can point to a specific amount of interest to be disallowed in computing the taxable profits of the foreign branch. It doesn’t require the foreign branch and parent to operate different external accounts (although they generally do).

                    • Excellent comment – agree with all that

                      In particular:

                      a) “it is as real as any monetary transfer can be in a modern economy” – i.e., it as if the bank is composed of a set of mini-banks, each defined as a coherent business unit with specified capital allocation and funding sources (or funding uses in the case of liability businesses and capital)

                      b) “you don’t actually need to record transfer of capital to do it” – right, although the internal books are set up to partition capital allocation among businesses, to transfer price all required internal funding (or available funds for use in the case of liability businesses and capital), to include the appropriate interest rate risk adjustment/transformation on the rate of interest earned on capital as funding for each business so that the business units will end up with a uniform “neutral interest rate profile and return” on their core capital allocation. In fact, the net effect of all of this is that allocated capital becomes a source of funding at the business unit level, and this is fully reflected in the internal flow of funds and transfer pricing.

                      c) “the point about foreign branch capital is that it’s not there to calculate ROE” – important and well stated, as the legal allocation of capital for legal reasons is transformed though appropriate internal funds transfer to something that is consistent with internal allocation of capital on an economic risk adjusted basis, as per b)

      • @JKH:

        Reserves don’t get “posted to capital accounts”. That’s meaningless from an accounting logic standpoint in general.

        “associated liabilities hidden from capital accounts” is also meaningless from an accounting logic standpoint in general.

        Right. I phrased that totally wrong. More below.

        Ramanan: “When a bank gets capitalized by the Treasury, it sells new shares to the Treasury and receives reserves in exchange”

        Thanks. I was under the impression that these transactions were significantly more complex, multilayered, and opaque because of hyperventilation about “nationalization” and government ownership. But even if they were, this is basically what happened.

        Banks’ journal entry:

        Sell shares to Fed (for reserve balances)

        Post:

        1. Reserves received to Reserve Balances asset account. (LHS)

        2. A. Equal amount to “Capital” (Shareholder’s Equity) liability account. Or: B. that measure just increases as a residual of assets-liabilities=shareholder’s equity. I’m not sure which would be the preferred/accepted accounting language/methodology to describe this in words, but the result is the same. IOW, I get it. RHS.

        What I should have said:

        You can’t point to any chunk of those holdings [assets] and say, “that’s the bank’s capital” or “that’s the bank’s required reserves.”

        Which I think JKH will endorse wholeheartedly.

        (I said “can,” but said the pointing would be arbitrary, saying basically the same thing. I can see why the “can” would set you off, though, cause it’s not true.)

        “Can’t” because:

        The first (capital) refers to a liability account.

        The second (reserves) refers to a regulatory (pro)portion.

        IOW I was saying the same thing but saying it wrong.

        @Ramanan: “receives reserves in exchange but as JKH mentioned this is the byproduct of the transaction”

        I would suggest that that is a somewhat arbitrary statement.

        Does the bank’s capital/shareholder equity (assets minus liabilities) increase because A. it issued shares, or B. it received reserve balances?

        My answer would be “Yes.”

        @JKH: I did, decades ago, take a fairly rigorous one-quarter financial accounting class at NYU MBA school, previously referred to. (And I was kind of a teacher’s pet cause I was this guy with a degree in Theory of Literature who was acing the class.) But much has slipped away. I really apologize for imposing on you to (re)educate me — you’re incredibly generous — but I’ll also say that I receive immense value from it, without going through a great deal of work and thousands of dollars in tuition, much (but of course not all) of it to re-iterate things I already understand. (And I’ll never earn anything with what I learn.) Also: I’m an addictive autodidact, get very impatient, drumming my fingers because classes just don’t move fast enough. Character flaw. But still, I might take your advice.

        • And yes I understand that capital (“capital right”) is allocated internally. I’m discussing the top-level balance-sheet.

        • “You can’t point to any chunk of those holdings [assets] and say”

          I would put ‘any’ in italics and leave out “chunk”

          Then I believe we’re thinking and saying the same thought in respect of this piece?

          • i.e. in respect of capital only

            I’ll leave the reserve piece for Ramanan – out of sensible division of labor

            :)

          • @JKH: “I would put ‘any’ in italics and leave out “chunk”

            Then I believe we’re thinking and saying the same thought in respect of this piece?”

            Yes. Though I think my statement might make the point we’re making even more strongly, make it even clearer to those who might misunderstand. But whatever.

        • “Does the bank’s capital/shareholder equity (assets minus liabilities) increase because A. it issued shares, or B. it received reserve balances?”

          Simultaneous events but it could have received promissory notes from the U.S. Treasury.

          But Keen thinks that Obama thinks that since it has received reserves it will start lending via the money multiplier process. Which Obama didn’t say. Obama said banks would be in a better capitalized position to continue lending.

          • @Ramanan: “But Keen thinks that Obama thinks that since it has received reserves it will start lending via the money multiplier process. Which Obama didn’t say. Obama said banks would be in a better capitalized position to continue lending.”

            Love this. But: I wouldn’t be at all surprised if Obama was confused on both points. So many are… Meaning Keen would be right about the thinking errors, but wrong about the thinking error expressed in that particular statement.

            • Steve,

              Yes Keen is right about the thinking errors but he is wrong on three things.

              As you say “wrong about the thinking error expressed in that particular statement.”.

              But there are two more confusions.

              One: he is wrong about thinking that the originators of the bank capitalization plan (not Obama but his advisors) thought that the aim is to increase reserves which isn’t so. The aim was the increase in capital. (For example, in Ireland the banks were capitalized by issuing promissory notes to the banks instead of increase in reserves in exchange for purchase of banks’ stocks.)

              Second he presumes implicitly the plan has no effect.

              In in all three errors.

  33. Mainly for Nick but also for Ramanan:

    Reserves

    1. “Reserve balances.” These are banks’ deposits at the Fed. Liability of the Fed, asset of the banks. Financial assets/liabilities that appear and are identified on the balance sheets.

    2. “Required reserves.” A regulatory amount (percentage of deposits) that banks are required to hold in specified “safe” assets — significant examples being treasuries, vault cash, gold (in their vaults or the Feds’), and…reserve balances. “Required reserves” does not appear on banks’ balance sheets.

    3. “Excess reserve( balance)s.” I add “balances” because this explicitly refers to that particular type of holdings. A bank could (in theory) have sufficient required reserves held in treasuries and cash, so all of its reserve balances could be “excess reserves.” (Depending on which of the bank’s assets you want to point to and call “required reserves.” That thing is a regulatory (pro)portion, not a financial asset or balance-sheet entry.)

    Required reserves aren’t necessarily held in the form of reserve balances.

    Reserve balances are not necessarily required reserves.

    (Which is why I would prefer a better term than “reserve balances.” Fed deposits?)

    Ramanan: “Steve Roth should have instead said “ban the phrase reserves” ”

    Ha. Got me. But the confusion arising from “reserve” usages is fairly simply and clearly resolved (above). And many/most discussants seem to understand these distinctions. That is decidedly not true for the “money” confusion.

    • Nick Edmonds says:

      Got you (I think).

      In the UK, reserves (in this context) usually just means commercial banks’ holding of reserve balances at the central bank, so reserves and reserve balances are pretty much the same thing.

      Re the balance sheet comment on required reserves. You are just saying that this is not a figure identified in the accounts aren’t you? You’re not saying these assets are off-balance sheet?

      I don’t know about the Fed, but reserve balances with the Bank of England are not quite the same thing as deposits with the BoE as there are other deposits that are not reserves (cash ratio deposits for example).

      • @Nick: “Re the balance sheet comment on required reserves. You are just saying that this is not a figure identified in the accounts aren’t you? You’re not saying these assets are off-balance sheet?”

        Correct.

    • Steve,

      Ok good point.

      But …

      “The banks’ reserve balances necessarily increased, but not the banks’ “reserves.” (This is a very widespread source of confusion.)”

      US Treasury capitalizing banks pays by transferring funds from the Treasuries accounts at the Fed to the banks’ account at the bank.

      Both reserves and reserve balances increase.

      Don’t know why you say reserve balances increased and not reserves.

      • @Ramanan: ““The banks’ reserve balances necessarily increased, but not the banks’ “reserves.” (This is a very widespread source of confusion.)””

        I should have said reserve balances increased but not required reserves.

        I’m sort of saying: excess reserves aren’t really reserves in the sense of “funds that are ring-fenced so 10% of people people can withdraw their money.” They’re reserve balances (fed deposits).

  34. BTW .. anyone interested … full of quotable quotes:

    “Why Britain Should Join The Euro”, 2002:

    Layard, Buiter, Huhne, Hutton, Kenen, & Turner and a priceless foreword by Volcker.

    http://www.fh-brandenburg.de/~brasche/EU/k3/k32/k325/whybritainshouldjointheeuro.pdf

    • Krugman’s latest:

      “There is not, I’m told, any formal Treasury backing for the Fed; there were informal assurances early in the crisis, and everyone takes it for granted that any losses that exceed the interest the Fed normally turns over would be made whole, but in principle the Treasury could let the Fed go bust. I guess that after President Rand Paul appoints his father as Treasury secretary this becomes a real concern.”

      LOL

      enter MMT

      • So what would happen if the fed went into a negative capital position and the treasury said we are not going to give you currency anymore and are not going to give you any gov’t bonds either?

        I hope I said those right.

        • not gonna happen (negative capital)

          I don’t make many outright 100 per cent forecasts, but that’s one

          • Let’s say the fed funds rate needed to rise to 3.5% and other treasury rates went up too (no yield curve inversion).

            The fed would lose “money” if they sold the bonds they hold. Paying IOR would probably mean paying more on their liabilities than their assets are earning. Is the fed going to go “crying” to congress and say give us treasury bonds that yield more?

            If so, I may be forced into debt by these other entities?

            • If the Fed wants to increase interest rates, it will simply pay more interest on reserves and/or do reverse repos.

              The Fed values bonds at amortized costs for held to maturity assets and these won’t fall due to fall in market value of assets. It will let maturing assets simply mature instead of rolling them over as it does in normal times and this will help to reduce the size of its balance sheet.

              • amortized cost instead of costs.

              • “If the Fed wants to increase interest rates, it will simply pay more interest on reserves and/or do reverse repos.”

                I’m going to focus on IOR. What is the fed’s holdings (mostly bonds?) yielding? I’m guessing less than 3%. If the IOR rate goes above it, what happens?

  35. Strange way of saying it by Krugman.

    A more direct way of saying it is the Eurosystem cannot finance the current account deficits of weak nations indefinitely because a pure fiscal support without any conditions i.e., an independent fiscal policy of the supported nation’s government will lead to its national debt and public debt (as a counterpart of the national debt) rising forever relative to their output. An increasing share of the rise in debt will be held by the Eurosystem and cannot continue.

    If this happens but they realize it cannot continue later, the nation leaves but the Eurosystem can easily be recapitalized by the governments. The trouble wouldn’t be this but the deterioration of the international investment position of the Euro Area.

    An alternative is coordination of fiscal policies by governments where creditor nations are encouraged to increase expenditure but this is difficult to do politically. So the only way out is to change the Treaty and give higher taxing and spending powers to a supranational fiscal authority. So taxes have to be sent directly to this authority. Low output regions will have lower income but will be recipients of net transfers in its favour.

    Or else prey for a miraculous large increase in net exports of the Euro Area.

    • On the Krugman quote, I seem to recall from looking at the Eurozone last year that the EZ members are on the hook for ECB losses according to their capital keys. That seems to imply “recapitalization” by way of loss coverage. What’s your understanding?

      • Don’t recall this. Any links. Did the members worry about their losses and complain to the ECB?

        My point was that even if the members reiterate and work out legal details on loss sharing etc and a plan to recapitalize, that won’t be a solution to the problem. The issue that the claims on the weak nation by the rest of the Euro Area can keep rising if the ECB underwrites the nation’s fiscal expansion.

        This can be seen by thinking of the Euro Area as just two countries in a monetary union in which the complication of the members of the rest of the Euro Area doesn’t arise because there is only one.

        • yeah – don’t disagree on your point

          I’d have to go back and look re mine, haven’t been there lately, but I think its part of the ‘constitution’ for the ECB – I’m pretty sure, although that would contradict Krugman et al – but its stated as I recall in terms of paying for losses, which can only mean automatic rolling recapitalization IMO, otherwise I’m not sure what it can mean in the sense of a meaningful monetary (and fiscal) effect for the EZ member capital holders

          That’s just bookkeeping of course

          Similarly, Krugman’s point about the Fed ‘going bust’ is pretty nonsensical – negative capital doesn’t mean ‘going bust’ in the case of the Fed (hence my summoning of MMT), not unless Congress shuts it down, which it would no more do than not recapitalize the commercial banks in the midst of the crisis – you don’t redesign the financial system in the midst of a financial crisis for the banks (which I think is a point you made elsewhere) OR for the Fed

    • On Fed losses (which I don’t think will happen), the question is whether Treasury would recapitalize it (which means in effect that Treasury swaps newly issue bonds that become Fed assets in exchange for paid in capital to cover losses – or as an alternative the Fed leaves a negative capital position to accumulate, evolve, devolve over a longer period of time, which would be offset by incremental interest paid on reserves (and compounded as a reserve liability) in an equal amount. The generation of negative capital has to show up as (compounding of) interest on reserves over time.

      Either way, it would probably take Congressional hearings to hash out the alternatives and their implications – as those sorts of hearing would no doubt precede any such situation ultimately materializing, as it could be detected in advance, depending on the actual path of monetary policy and interest rate levels. I think that would happen, legal niceties or not.

      But the economic point of it all is that eventually the process would catch up with itself and start producing positive margins for the Fed once again, simply due to the passage of time. This is assured eventually simply due to the gradual maturing of the bond portfolio and the buffer effect of banknote seigniorage over a long period of time.

      There are a number of reasons why I think the Fed won’t go into a loss position. One is that I expect net banknote expansion will be at least $ 50 billion a year over the next 10 years or so. That banknote increase will result in equivalent reserve destruction and saving of interest paid on reserves. Another reason is the effect of bond maturities over that period, which also destroys reserves. Another is the effect of bonds rolling down the yield curve as they age, which means less in the way of losses as they approach maturity, if they are sold before maturity. Another is that rates will likely move up gradually. And so on. Passage of time has a powerful effect on bond portfolios and interest margins in this sort of workout situation of an “offside” portfolio.

      • Excellent points especially passage of time.

        Even in drastic scenarios I think it can be shown that the Fed’s is able to sail smooth. The initial intuition that rises in interest rate will easily lead to a negative capital isn’t right I guess.

        Also, isn’t the fact that the assets the Fed holds are valued at amortized costs important in the discussion? People seem to implicitly assume market valuation.

        • yes – VERY important – should have mentioned that

          time is on their side in the way it isn’t for a volatile trading portfolio

          that whole issue was very important during the financial crisis – its the obsession with marked to market accounting that almost buried the entire financial system

          • “that whole issue was very important during the financial crisis – its the obsession with marked to market accounting that almost buried the entire financial system”

            I’m trying to remember this from something I read somewhere. I thought mark to market accounting was about needing to set aside less bank capital?

  36. Nick Edmonds June 7, 2013 at 11:05 am

    Very good question.

    Some generic comments, for simplicity:

    The first principle is that the purpose of capital is to absorb losses. Losses can occur across a range of define businesses. The risk in each business is evaluated, and capital is allocated across businesses accordingly.

    So it’s natural to divide capital up internally in such a way as to distribute it to defined business units on the basis of risk taken or risk limits allowed. That way, risk versus return performance can be measured for each business unit.

    Capital is basically the bank CFO’s or treasurer’s portfolio. That function is responsible for making sure the allocation gets done “properly”.

    As part of this, it is important that all forms of capital be accounted for.

    The area is too complex to become too precise in giving boilerplate general answers, but suffice to say that it makes no sense to use any particular form of defined capital in such a way as to dedicate that form to any particular business. The entire capital position, as defined, needs to be allocated out on a basis which does not discriminate in favor of or against different business units according to the weighted cost of capital. So capital is defined, and then allocated out in pieces of homogenous composition – with common equity and non-common components, whatever those latter parts might be (e.g. preferred equity, qualifying debenture capital, etc.) All those pieces have a homogenous transfer price interest rate (as earlier described on a source of funds basis), and then the combination of that plus the effect of business unit leverage and margin otherwise will determine the unit’s overall return on capital.

    It’s important that the capital attribution requirement be fully coherent with the way in which risk is measured – because the purpose of capital is to absorb the risk of unexpected losses. So the analysis and measurement framework for risk and capital allocation must be fully integrated – so that capital units of measure are allocated to risk units of measure in a consistent way.

    Banks are subject to regulatory requirements for minimum capital levels. One interesting aspect relating to your question is how banks deal with excess capital. And it is interesting that banks will strategically position their excess capital level in such a way that they do not in the case of reserve management. Banks do not want to be caught with deficient capital positions, and they want some measure of excess capital on hand in case of business opportunities. And banks don’t want to raise external capital suddenly, simply because of prior poor management and being offside on their ratios. And they have to wait for the passage of time to generate internal capital. This of course is not the case with reserves, where a solvent bank can raise reserves at a moment’s notice simply by going to the money market. This sort of difference is what it means for banks to be capital constrained and not reserve constrained. The whole sad saga of the money multiplier can be interpreted as a case in which neo-classical economics has confused the roles of reserves and capital.

    It is conceivable that different bank managements may make slightly different evaluations of capital requirement at the level of individual business lines, but in such a way that constraints on aggregate regulatory requirements are not violated. This point shouldn’t be overemphasized, but it is possible for different bank managements to tweak their own view of risk in a way that may be slightly different than the way in which regulators do for individual pieces – but not in such a way as to violate aggregate regulatory requirements. However, I think this is a minor point.

    The more important point is the difference between actual capital and required capital – that difference being excess capital. That excess is temporarily “parked” by the Treasurer in risk free assets, those assets being the asset offset to excess capital funding, but requiring no capital allocation for risk, until later used in some new business venture or in supporting the growth of existing businesses.

    • Nick Edmonds says:

      Thank you for that.

      I wonder if you could clarify one point. Aside from the fact that the bank will want to keep a buffer of capital, I think you are saying that the business RoE is generally calculated on the basis of actual ex-post capital usage (whether calculated on normal BIS rules or an internal basis), as opposed to a predetermined allocation, which is fixed regardless of whether the actual usage is higher or lower.

      Is that right, or am I reading too much into your reply?

      • That’s going to vary among banks. But again in general, the capital utilization of trading businesses would be calculated on a daily basis. If trading businesses tended to have a track record of high volatility of intra-day risk, this would be taken into account in some way through limits on both intra-day risk and potential intra-day capital exposure even if the latter is not actually captured in ROE calculation, but I’m sure that would vary among banks. The utilization of non-trading businesses would be captured depending on a frequency that would depend on the granularity of such changes and their materiality and frequency. Trading businesses draw on capital on a daily business, non-trading businesses if anything slower than that. I’m guessing most banks might have something recording utilization weekly for the entire bank, but some might have a daily estimate for the entire bank.

        Trading businesses in particular are subject to upper risk limits that determine upper limits on capital utilization (and vice versa I suppose) on both daily and intra-day bases. I think its the limits that play the role of predetermined allocation (allocation of risk limits) there more than in the sense of the allocation of actual capital at risk at a point in time.

        The entire framework is not just for regulatory purposes. The bank does want to know its risk exposure as best as possible, and its corresponding position in terms of capital adequacy to absorb related losses.

        • Nick Edmonds says:

          Thank you again, and I’m sorry we’ve drifted so far from the subject of Steve Keen and his accounting.

          I have my own stuff on the sorts of things SK is looking at. I’m thinking of maybe posting a few bits somewhere. If I do, any comments that you’d care to make would be gratefully received.

  37. Back to “demand”:

    You object to GDP + Borrowing = Demand (Effective, Aggregate, whatever)

    Isn’t your objection based on the implicit assumption that GDP = Demand? That assumption renders Keen’s formula ridiculous a priori in “accounting” (or simply arithmetic) terms.

    But isn’t that assumption falling prey to 1. the ex-post fallacy and 2. an insufficiently theorized notion of “demand”?

    I could certainly be wrong here…

    • Hi Steve,

      Aggregate demand is properly expressed in terms of the national accounts identity for income and expenditure flows:

      C + I + G + (X – M) = C + S + T

      Where S is private sector saving

      That’s standard, and expressing aggregate demand in those terms is standard and stock/flow consistent. It’s all expressed in terms of expenditure/income flows over a given accounting period.

      And NIPA maps to that consistently as an ex post calculation in terms of expenditure/income flows.

      Aggregate demand is simply a scenario for that equation. For example, increasing G provides a Keynesian stimulus that may increase aggregate demand. And as that occurs, the above equations must be satisfied AT ALL TIMES, continuously. That sort of continuous stock/flow consistency is essential in Godley/Lavoie.

      There is no analytical issue in terms of ex post / ex ante. That is a red herring that is the result of a failure to understand the full depth and time scope of the application of accounting to economics. The WHOLE point (really, the main point) of the Godley/Lavoie book is that ex ante simulations of and scenarios for the economy must be stock flow consistent from an accounting perspective. This is forward looking.There is no “ex-post fallacy”.

      As I said in the post:

      “The idea of accounting “closure” resonates throughout the book. It is at its core the simple idea that balance sheets connect to each other through double-entry bookkeeping over space/time, with the facility of consistent income and flow of funds accounting as the required linkages for change. It acknowledges and uses the powerful logic that accounting is not just a rear view measurement device – it is also a constraint on all forward looking projections of economic outcomes – meaning that it is an important condition in the substance and shape of good economic analysis.”

      The main problem with the Keen conversion, as I said in the post, is that Keen has attempted to create his own definition of aggregate demand in such a way that it can only be stock/flow inconsistent. Here’s what I said in the post:

      “Steve Keen also notes his objective to establish a new definition for aggregate demand by equating it to income plus the change in debt. This entails embedded accounting confusion. Notwithstanding the evidence of impressive historic correlations and causal connections between changes in debt and economic outcomes, it is nevertheless incorrect to add income and flow of funds (i.e. a change in debt in this case) in any expression deemed to be an equation or an identity. Such an expression at best can serve as a regression function, in this case relating current period income to prior period income plus the change in debt. It includes accounting variables that are simply incompatible in an additive sense for an actual equation to hold, either during a single period of time or at a single point in continuous time. Income and changes in debt are orthogonal accounting measures. Flow of funds accounting where the balance sheet equity account is not involved (e.g. increases in debt) cannot intertwine indiscriminately with income accounting in any fashion that could be considered stock/flow consistent. Moreover, there is the problem that changes in debt may well be a source of finance used, not for spending as captured in NIPA, but for asset acquisition as captured only in the flow of funds accounts. This component of the use of funds in the economy has nothing directly to do with income accounting. Finally, the notion that there can be such a strict relationship in terms of income and the change in debt overlooks the fact that aggregate demand can fluctuate at times due to changes in money velocity, without necessarily involving changes in debt. In either case, it is potential and actual spending rather than a change in financial stocks or the utilization of existing stocks (i.e. money) that is inherent in the idea of aggregate demand. I should note again that ‘Monetary Economics” by Godley and Lavoie is crystal clear and comprehensive in its appreciation of the differences between these two accounting modes, and of the importance of that distinction to the understanding of stock/flow consistent projections or scenarios for future economic activity.”

      In short, Keen’s error IMO includes rather massive inconsistencies in both time and space dimensions. That’s quite a lot. His equation properly can only be a regression equation. Otherwise it contradicts stock/flow consistent concepts of aggregate demand, income accounting, and flow of funds accounting. He’s just going to confuse himself and everybody else by creating personalized definitions that can never end up being consistent with a full world view of stock flow consistent economics.

      Steve, just to keep my own flows under control, I pray you don’t come back and say “I get all that” – because with respect I don’t see how this can be the case. It seems to me that it’s a similar question that keeps recurring. This idea that accounting must only be backward looking is simply wrong. Stock flow consistency means that accounting must be forward looking to the degree that economics is forward looking.

      • K let’s forget the whole ante/post thing.

        “Aggregate demand is properly expressed in terms of the national accounts identity for income and expenditure flows:

        C + I + G + (X – M) = C + S + T”

        I don’t see the term “demand” in that equation. What I *think* I see is an implicit assumption that “demand” = GDP = C + I + …..

        This looks very similar to G-L’s “Model SIM,” discussed very early on page 63, wherein they state, “These four equations imply that whatever is demanded (services, taxes and labour) is always supplied within the period.”

        IOW, in that early, simple model, “demand” = GDP.

        I’m digging more into this discussion and later in the book to suss out where they go with “demand” theory. Not making any statements about their demand theory here.

        My impression is that “aggregate demand” is *not* coherently or consistently defined out there in the world. It gets thrown around every which way from Sunday. And it’s far from simple and obvious. (Again: Nick makes valiant efforts to grok it, though I’m not at all sure he succeeds.)

        “Keen has attempted to create his own definition of aggregate demand”

        Can you tell me your definition?

        If there’s real-sector net borrowing in a period but the funds aren’t spent on (higher inflows to real-sector units but not higher income or expenditure), is there higher “demand” in that period than there would have been absent that borrowing?

        • My definition is as above – that its a scenario for that equation.

          I.e. a scenario for the values in that equation, coming at it from the expenditure side (obviously in the sense of demand as manifested in expenditure), but recognizing that there is an identity by which there are values for both sides of the equation

          Could be ex post, current period, instantaneous, or ex ante

          Depends how you want to use it

          But you can never throw in a change in debt

          There is no such equation in that last case

        • And yes, that equation does represent GDP

          So its a scenario for GDP as expressed through that equation

          But still could be ex post, current period, instantaneous, or ex ante

          Nick’s economics is far divorced from Godley/Lavoie – there’s no full reconciliation possible there

          The principal here of SF consistency is present in every GL model, and there will be a corresponding form for AD/GDP in every such model, acknowledging that each model represents a different composition of sectors

        • “IOW, in that early, simple model, “demand” = GDP.”

          Yes only in the early simple model. In later models, aggregate demand and aggregate supply from each other by change in inventories.

          There are two sense of the usage of aggregate demand. Aggregate demand is a sum of many things such as consumption demand + … and the consumption demand itself depends on the propensity to consume, expected income and so on. You can of course say ex-post that the aggregate demand would be sales and aggregate supply is output.

          But in no sense is the concept ex-post. For example, I can say that higher government expenditure will lead to higher aggregate demand.

          • “would be” bad grammar.

            “Aggregate demand was” is a better usage.

          • yeah, but inventory investment, intended or unintended, is accepted to be investment and puts the equation in equilibrium at all times in GL models right?

            I think that’s the best way, BTW

            that differs from the neoclassical models who have it out of equilibrium

            • note I said THE EQUATION ‘in equilibrium’, meaning that the equation holds

            • Point. I’d say that aggregate demand includes investment demand not including change in inventories or something of that sort if the definition of investment counts change in inventories.

              • really?

                but isn’t any gross increase in inventories (before accounting for sales) for example due to the decision of the firm to increase them?

                so even if it sells nothing in the period, the net increase in inventories is attributable to the firm’s contribution to aggregate demand?

                that corresponds to GDP and national accounting identities, and is what I would have thought to be stock flow consistent for aggregate demand

                • I’d say indirectly because the model is a sequence of income, expenditure, output, sales etc. So a firm may increase production and this would increase aggregate supply and output but not demand directly. That would come because increase in production will probably lead to more hiring etc and more wages which contributes to demand via increased consumption demand.

                  • I understand the process.

                    But it looks like you’re saying that your/the definition of aggregate demand does not include demand for inventory investment – i.e. what is actually purchased for inventory, regardless of sales from there – i.e. net intended or net unintended – that post Keynesian models do not consider that type of inventory investment as part of aggregate demand?

                    Just asking. I’ll have to check up on it again.

                    • Yeah not considered so aggregate demand and aggregate supply differ by “change in inventories”.

                      It’s like imagine a weekend with people shopping and when there is almost zero production except taxi and train services. So output and aggregate supply is almost zero (during the two days) but not aggregate demand and sales. So two differ and delta inventories is what adjusts. It is a bit difficult to have definitions which aggregate demand and supply are equal to one another.

                    • I like this Hicks quote:

                      “… The traditional view that market price is, at least in some way, determined by an equation of demand and supply had now to be given up. If demand and supply are interpreted, as had formerly seemed to be sufficient, as flow demands and supplies coming from outsiders, it is no longer true that there is any tendency over any particular period, for them to be equalized: a difference between them, if it were not too large, could be matched by a change in stocks. It is of course true that if no distinction is made between demand from stockholders and demand from outside the market, demand and supply in that inclusive sense must be equal. But that equation is vacuous. It cannot be used to determine price, in Walras’ or Marshall’s manner. For what matters to the stockholder is the stock that he is holding: the increment in that stock, during a period is the difference between what is held at the end and what was held at the beginning, and the beginning stock is carried over from the past. So the demand-supply equation can only be used in a recursive manner, to determine a sequence (It is a difference or a differential equation); it cannot be used directly to determine price, as Walras and Marshall had used it.”

                    • OK, thanks.

                      Haven’t been looking at it that way.

                      I’ll go back to GL ME in a few weeks time for another run through and look for that more closely.

                      Actually, I can’t say I like it. To my thinking, an unintended increase in ex post inventories is still the result of aggregate demand from firms; and an ex ante expected adjustment might involve a reduction in the demand by firms for new inventory investment. And the national income equation holds at all points.

                      whatever … circle back on this down the road maybe

                      (meanwhile, Steve Roth snickers off stage as JKH flounders with aggregate demand … JKH prepares for full body blow from Roth, forthcoming)

                      :)

                    • funny, I read that Hicks quote, and I get consistency with what I’m saying

                      just me, I guess …

                    • Yeah think understand what you are saying.

                      But described in a slightly different way in G&L. As in, seeing more demand, firms will increase production and the way they do it by looking at the inventories/sales ratio.

                      So it is higher demand reduces inventories/sales ratio which will lead firms to increase production to bring it back to the original. So more like aggregate demand and supply keeping up with each other but differing with each other in general.

                    • I understand what you’re describing in terms of household demand.

                      But I just skimmed GL for aggregate demand references, and I’m not seeing your aggregate demand definition yet.

                      whatever … later

                    • Oh Chapter 9 first page:

                      “… Third, we are about to break decisively with the standard assumption that aggregate demand is always equal to aggregate supply. Aggregate demand will now be equal to aggregate supply plus or minus any change in inventories. Hicks (1989) noted the high theoretical importance of this proposition because it destroys the market clearing condition which is central to general equilibrium theory, though he did so with a marked lack of rhetoric. For this reason our new model will be called Model DIS because it deals with disequilibrium (of a kind) in the goods market.”

                      the reference to Hicks is the earlier comment (from his book “A Market Theory of Money”).

                    • thanks!

                      Now I’m feeling mildly dumb.

                      So if aggregate supply = S, and the change in inventories is + 100, what is aggregate demand D?

                    • S minus 100.

                      So that in this case aggregate demand is less than aggregate supply.

                      Two things: if production is increased, that will increase the supply and change in inventories, not demand. Of course the latter will change in the next period because of the sequential nature of the description.

                      Also, on any other day/period, aggregate demand can be greater than aggregate supply, so change in inventories can be negative.

                    • Right, and that should be obvious, but just checking.

                      Because it seems like quite an awkward sentence in describing the signs involved – the change in inventories is “added” if the change in inventories is negative.

                      And your description of the inventory process is quite clear, and I think I understand the inventory process they describe in general, and I can now see the use of their language in describing aggregate demand.

                      What I don’t understand right now is why he would feel the NEED to use the language in that way. My own intuitive preference would be to describe an increase in inventories as a component of aggregate demand. And that intuition is immune to the change in the pricing mechanism – i.e. it has nothing to do with the difference between neoclassical and post Keynesian pricing theory. That difference in pricing explanation doesn’t suggest the need to me for a definitional change in the case of aggregate demand at all, from a purely personal intuition perspective.

                      So right now I’m puzzled.

                      Just means I need to spend more time with the book obviously.

                  • i.e. I understand it as a post Keynesian description of how the economy works

                    I don’t understand why it needs to become a language issue – because you could define aggregate demand to include change in inventories, without disturbing the analytics of the PKE model, its seems to me

                    • Oh if you include it will be difficult to think of describing things such as me purchasing a car. One would roughly say it has increased the demand and reduced inventories. But if one defines aggregate demand to include change in inventories, I don’t know how to translate it in normal language. The producer will notice a drop in inventories and increase production and so on but that is slightly a different point, but only slightly – highly related.

                      [Digressing, I think there are more issues that Nick Rowe keeps mentioning such as the car model not being available - I think this can be handled using notional and effective demand and things such as that but nothing to do with the previous paragraph].

                    • You may be right – although it seems to work OK in national accounts.

                      E.g.

                      If purchasing a car from the dealer lot is all that happens in the economy, then GDP is zero in that respect considered on its own, and inventory investment just converts to a consumer durable.

                      So net AD is zero in respect of that, which sounds weird, except all that’s happened at the macro level is that you’ve converted the classification of an asset that’s already been produced.

                      The way I would think of it is that household D is X and business D is (X) in that case.

                      Inventory unwinding of (X), whether intended or unintended, seems not incompatible with the concept of negative business demand to me (and positive household demand).

                      But the important point is that none of this (optional) use of language really changes the importance of the GL treatment of inventories and loans in terms in substance of their analysis.

                    • i.e. in my example, a bit like the “asset swap” idea – between firm and household in this case

                    • Oh I see what you mean and possibly that’s the reason why you see no inconsistency with Hicks because he says both definitions are possible.

                      “… a difference between them, if it were not too large, could be matched by a change in stocks. It is of course true that if no distinction is made between demand from stockholders and demand from outside the market, demand and supply in that inclusive sense must be equal”

                    • What does the scenario look like if GM built a car in April 2009 to sell for $20,000, but the car dealer could not sell it until April 2011 for $20,000?

        • I think you’re right that it gets thrown around – but the throwing around is hopefully minimized in Godley/Laovie and SFC accounting

          The usefulness of using both AD and GDP is that AD is pleasant verbiage for playing around with Keynesian “autonomous injections” like government spending and investment – which are intended to induce additional demand and multiplier effects, etc. – those things cause expansion of GDP from current levels, other thing equal

      • @JKH: “My definition is as above – that its a scenario for that equation. I.e. a scenario for the values in that equation”

        I’ve read this over a dozen times, thinking as hard as I can, and I just can’t understand what it means. I can’t figure out what you mean by “demand.” (Except: total purchases, which doesn’t explain anything. It’s just another word for the same thing.)

        “Aggregate demand” is not an accounting term, is not defined in (or even part of) the accounting lexicon (as equalling GDP or anything else). It’s a very poorly defined and variously used economics term. (cf Ramanan and JKH not knowing/agreeing whether to include inventory changes.) Depending on how it’s defined, in a given period aggregate demand could be higher than, lower than, or equal to GDP.

        You’re defining it as identically equal to GDP, because…I don’t know why.

        “I’ll go back to GL ME in a few weeks time for another run through and look for that more closely.”

        I’d be very interested to hear your thinking. In particular to some of these passages and ideas (condensed not for rhetorical purpose but for brevity and clarity):

        “…accounting matrices [absent "behavioral (transaction) matrices"]…can say nothing at all about how the system…works.”

        IOW, the balance sheet and transaction matrices only provide the “backbone to the model.” They are insufficient to specify the model (or, I would add, to provide a definition or understanding of “demand”).

        “…beyond displaying the universe to be considered, these matrices do no more than show how all columns and rows must in the end ‘add up.’ By themselves they can say nothing at all about…how the system works.”

        “it may not be assumed in advance that the horizontal entries sum to zero; we have to specify the mechanisms by which these equalities come about.”

        What’s required is G-L’s “Behavioral (transactions) matrix,” which necessarily makes assumptions that are beyond the scope of accounting.

        As G-L point out, in their Model SIM equations:

        “…what is demanded…is always supplied within the period. These…are strong assumption implying, obviously enough, that we are describing an economy that has no supply constraints of any kind.”

        That’s the only “scenario” (“obviously enough”) in which demand = GDP — one where all demand is satisfied.

        They formulate the “how does it work?” question very well here, emphasis mine:

        “Starting from a situation where production could be different from supply, and where supply could be different from demand, how do we arrive at the equality between sales and purchases? There are several mechanisms that could lead to such a result…”

        (They cite four possible mechanisms, explained via: neoclassical price adjustment and market clearing, rationing and constrained equilibrium, inventories, and Keynesian/Kaleckian quantity adjustment.)

        I think we’ll all agree that (loosely stated, as per Ramanan) real-sector borrowing from the financial sector increases “demand” for real goods. (Yes, with “leakage” back into financial-asset markets.)

        But if demand = GDP, and inflows from borrowing (unspent) are not part of GDP, that’s impossible. How does that work?

        (Still like to hear an answer to this: “If there’s real-sector net borrowing in a period but the funds aren’t spent (higher inflows to real-sector units but not higher income or expenditure), is there higher “demand” in that period than there would have been absent that borrowing?”)

        Keen’s trying to answer and understand that question formally, and arguably getting the answer/understanding wrong. Nick also tries to answer it, perhaps also getting it wrong.

        But just pointing to the GDP equation and saying “that’s the scenario” doesn’t answer it either. A useful answer to an aggregate demand question must define aggregate demand beyond (implicitly) making it identical to expenditure.

        Much more fascinating conceptual and theoretical thinking to wrestle with on pp. 59-65 alone, going far beyond just “getting the accounting right.”

        I can find no place in the book, BTW, where G-L directly tackle the issue of debt’s contribution to demand, or lack of same. Their later models generally treat debt as funding inventory expansion, reflected in household saving. But that’s a fairly limited and stylized view, it seems to me. Much good economics work to be built on their accounting-based frameworks and models… Time to make the book required reading for every econ student, perhaps beginning with 101.

        • “I can find no place in the book, BTW, where G-L directly tackle the issue of debt’s contribution to demand, or lack of same.”

          Addressed to JKH so I won’t interfere with the rest of the discussion but in their models households do finance consumption by borrowing. In addition to working capital finance, firms finance investment by equities or loans in simpler models but you can think of them financing by issuing bonds. The contribution to demand is from the demands such as consumption demand , investment demand etc. This can be financed from income or by borrowing. As I mention in another comment, it is a bit like looking at the wrong side of the balance sheet – in this case wrong flows.

          • “As I mention in another comment, it is a bit like looking at the wrong side of the balance sheet – in this case wrong flows”

            To be clear asking debt’s contribution to demand is a bit like looking at the wrong side of the balance sheet – in this case wrong flows.

          • Nick Edmonds says:

            G-L have something in their growth model which deals with personal loans, but I don’t think they give this as much space as it deserves.

            Interestingly, the behavioural relationships for personal loans in this model do not appear to be based on a target stock-flow ratio. The steady state debt income ratio seems to be determined by the repayment rate on loans, rather than by any concept of acceptable debt burden. This seems to me a bit of an anomaly, given the overall approach in the book.

        • First, I should have acknowledged more directly that I think you have a good point that AD is something where two much vagueness surrounds the definition.

          My approach in the context of this entire discussion is to have a general idea (my own idea I guess) about it and at the same time start to weed out some ideas about it that are unacceptable, IMO.

          The first general idea is that it be expressed in terms of expenditure – as in the type of expenditure which appears in the national accounts identity for income – where expenditure equals income. That’s the reason for associating income with aggregate demand as well as expenditure. But expenditure is the more direct and intuitive one.

          Next, since the national income identity is intended to fit with GDP, yes there is a connection there.

          Third, the first thing that the definition of AD cannot include is non-income flow of funds elements, such as a change in debt.

          As Ramanan points out, that is double counting of a sort – although it’s actually more than double counting – because it includes flow of funds such as change in debt where the flow is used to finance non-GDP purchases – such as existing houses. That’s more than double counting things that are correctly in AD.

          In any event, that’s the one thing I insist not be included in the concept of AD, which SK does directly and I believe that Nick R may do indirectly some times.

          I would also insist that Keen’s equation be expressed as a regression equation with appropriate timing subscripts to indicate the time sequence of any inferred causality.

          On a tangent now, Ramanan and I have been having a discussion about the relationship between AD and GDP, which has to do with inventories. Not sure it’s so much a disagreement, as it is my failure to understand why GL is using the term in quite the way it does. I need to go back to that. As I suggested to Ramanan, I don’t see why it’s necessary to do that.

          Fourth – my definition then basically amounts to scenarios for GDP. You’re right – that doesn’t seem to add much. Except that the aggregate demand terminology is useful IMO when assessing for example a projected marginal increase to AD – such as Keynesian “exogenous injections” like government spending and investment, and their effect on GDP. This is forward looking stuff. Where I take your point in particular about the vagueness of it all is that it seems quite redundant especially to refer to AD ex post – to my way of thinking that is just GDP ex post, which is along the lines of your thinking about it as a redundancy. But I do think it’s a bit more useful ex ante as above. That said, we have to adjust all that apparently for the GL use of AD in a way that excludes changes in inventory investment, which is part of GDP.

          At the end of the day, I tend to agree with you that it’s too vague.

          Actually, I really think about the term somewhat in the same way as I do about ‘money’. I think the exact definition is really not terribly important. You can do economic analysis without either of those terms, IMO, because you’re always going to require more specificity to understand what’s being discussed at the appropriate level of detail. I don’t mind referring to bank reserves as money if I’m describing things from the perspective of a bank reserve manager, but I want to know I’m talking about bank reserves as opposed to a commercial bank deposit liability. That doesn’t require me to reject either one of those things when using the term money. Same sort of thing with AD I think. I don’t feel a pressing need to define it perfectly – although in partial opposition to that I can’t say I like the GL adjustment identified by Ramanan. As a final example, I think its much more important to define saving perfectly, for example, than AD. Because saving is definitely a national income concept.

        • going far beyond just “getting the accounting right”

          there’s that straw man again – it never goes away

          • “just “getting the accounting right”

            there’s that straw man again – it never goes away”

            Okay I promise I’ll never assert that in the future. (Pointed in your direction.) Though I felt that reciting C + I +… was getting perilously close…

            I think you and I crave different things: you want to know “what does it say?” I’m always driving past that (sometimes with only a glance) to “what does it mean?”

            Character flaw or not, I really feel a need for a better definition of (aggregate, effective, …) demand. Don’t feel like I can understand how it all works without it.

            1. See Nick’s comment, and a couple that follow including a link, about “notional” demand:

            http://www.asymptosis.com/is-the-elasticity-of-labor-demand-at-zero.html#comment-2739

            2. Clower on flow demand/supply and stock demand/supply. This seems essential thinking to me and seems never to have been theorized beyond this exploratory paper.

            http://www.jstor.org/discover/10.2307/1907357?uid=3739960&uid=2&uid=4&uid=3739256&sid=21102377905657

            3. I would *really* like someone with better supply-and-demand-curve chops than I to give a careful vetting to this:

            effectivedemand.typepad.com

            It’s extremely formalized and precisely defined, and from my surface viewing seems compelling.

            And he makes predictions. i.e.: short financial stocks, they will dive by or about the end of the year. That is only a virtue, of course, if he’s right…

            IOW, even on its most basic constructs, I think the economics community — maybe even G&L — have yet to carefully define what’s arguably the most fundamental term in the discipline.

            (Oh: besides “money.”)

            So maybe this is why I tend to defend Keen. My cravings for understanding are similar to his, and we trip over the same conceptual constructs.

            I am never, ever going to get cogent discussion of this stuff in a University of Washington econ or accounting class. (And don’t even get me started on their Shakespeare department.) So thanks for listening.

            • “I think you and I crave different things: you want to know “what does it say?” I’m always driving past that (sometimes with only a glance) to “what does it mean?”

              I don’t think you quite mean that the way you say it (ha!) because I don’t think that’s fair.

              I think I agree with your desire for a clear definition here. I’m just more patient than you.

              A couple of things:

              Just looked at my 30 year old macro textbook, and I see no reference to the elimination of change in inventory investment in the general definition of aggregate demand.

              I do see this:

              “GDP refers to actual spending whereas aggregate demand may be described as either planned or actual spending”.

              That’s what I was getting at with “scenarios”.

              Also, I see the Keynesian cross throughout the chapter on aggregate demand. This is the planned dimension – or what I refer to as scenarios. That’s what I was getting at with “Keynesian exogenous injections”. I rarely read Nick now, but I recall him saying something about there being a number of ways to teach this stuff, including Keynesian cross, ISLM, etc.

              And then what I see is that aggregate demand CAN include unintended inventory investment BUT that that there is no unintended inventory investment ONLY if EQUILIBRIUM is reached via Keynesian cross, ISLM, etc.

              I.e. aggregate demand CAN include unintended inventories, as per my assumption. And the scenario idea includes OUT OF EQUILIBRIUM scenarios.

              Second, have a look at the Wiki:

              http://en.wikipedia.org/wiki/Aggregate_demand

              I don’t know if its “right” but it begs the question as to whether ANYBODY IS “RIGHT” WHEN IT COMES TO DEFINING AGGREGATE DEMAND.

              A couple of things jump out.

              Somebody has put in the Keen approach as if it’s the generalized post Keynesian approach. (This should drive Ramanan through the roof.)

              Second, I see the assumption of static inventories.

              I’d be interested in Ramanan’s view there. Is the assumption of static inventories considered standard, so that there’s nothing special in GL about the treatment of inventories except model SIM? Because that’s not what my old macro textbook says. My textbook seems to restrict that to equilibrium, which is really on one scenario out of many for aggregate demand (can the multiplier be viewed as the evolution of aggregate demand toward equilibrium?).

              Is there no such thing as broad agreement on this subject (which would be one reason for my being somewhat loose about it), since in using the term it generally requires a qualification about what one means? Steve, does Nick usually qualify what he means when using aggregate demand in any in-depth discussion? Because that would lend credence to the view that the definition always has to be specified – simply because economists all over used different definitions? That’s my guess on the thing.

              • Interesting paragraph from my macro textbook:

                “Since the value of unplanned inventories is the difference between aggregate supply and planned aggregate demand, it can be seen that unplanned inventories is the item which makes actual aggregate demand necessarily equal to aggregate supply. That is, actual aggregate demand and supply will always be equal regardless of the level of national income. However, there can only be one level of national income at which planned aggregate demand is equal to aggregate supply: this is the equilibrium level of national income.”

                I think that roughly describes how I’ve been approaching this intuitively and using the term aggregate demand – i.e. in the sense of actual aggregate demand, where only in equilibrium does actual aggregate demand equal planned aggregate demand and is there no unintended inventory investment. Unintended inventory investment only happens at equilibrium, where planned and actual aggregate demand are equal. My use of the term is wider than such an equilibrium condition – hence my use of the term ‘scenarios’.

                BTW, I find the elimination of inventory changes in entirety to be totally unsatisfactory (if that is claimed anyway), because at least there is a secular increase in desired inventories as the economy grows.

              • starting to see the distinctions between use of the terms:

                -aggregate demand
                -planned aggregate demand
                -actual aggregate demand

                “aggregate demand” may be too loose a term for purposes that are deceivingly specific

              • How about this?

                http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/01/the-war-on-demand-and-the-short-side-rule-.html

                “Quantity actually bought and sold is whichever is less: quantity demanded; or quantity supplied. Q=min{Qd,Qs}. That’s the short side rule. The short side of the market determines the quantity traded.”

              • @All:

                Have to go out the door soon so this response will be disorganized, rambling, and repetitive. Apologies. (Twain: “I’m sorry I wrote you such a long letter. If I’d had time I would have written a shorter one.”)

                And I just can’t address all the good stuff being discussed here, so I guess rather random responses…

                @JKH: “I don’t think that’s fair.”

                To you or to me? Probably both.

                “I’m just more patient than you.”

                Undoubtedly. But in my defense: some centuries into this “economics” enterprise, and economists haven’t sorted out what demand, supply, and money mean? Just sayin': grounds for impatience? (Ramanan: it’s like physics before Newton, certainly before Einstein, medicine before germ theory. Keen is flailing around in epicycles before heliocentrism, but so are we all, and he’s at least trying to break out. He’s no Newton but could he help provide the intellectual space for one? I think his book is a big step that way.)

                “have a general idea (my own idea I guess) about it and at the same time start to weed out some ideas about it that are unacceptable”

                Seems like a sensible approach given the confused state of play.

                I’m not going to wade into the inventory thing except to say that 1. G-L highly favorite it as a theoretical model to bring S and D into equality, and 2. I wonder if they overemphasize it given that in the U.S. at least, we have an 80% service economy in which (labor) capacity, rather than inventory, is the key buffer. If a service isn’t purchased, it isn’t produced. Full stop.

                “does Nick usually qualify what he means when using aggregate demand in any in-depth discussion?”

                Yes, in multiple places and ways. Here the stellar JW Mason encapsulates in brief, responding to commenter JCE giving Nick Rowe Keynesian Cross links:

                http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/01/understanding-the-keynesian-cross.html

                http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/02/but-where-will-the-demand-come-from-in-praise-of-older-keynesians.html

                JW MasonApril 20, 2013 at 12:55 AM
                JCE-

                Nick Rowe’s stuff is great. I wish I could write half as clearly as he does. But “aggregate demand” when he says it, means something different from when Keynesians say it.

                For us, a “demand problem” means that someone has chosen to reduce their expenditure relative to their income, and because in modern economies there is a strong causal link from expenditure –> output –> income –> expenditure, the effect of this is to reduce total activity in a self-perpetuating way.

                Whereas for Nick, it means there is excess demand for means of payment, and excess supply of currently produced goods, which will last until the supply of money increases, or the price of currently produced goods falls.

                Two ways of looking at the same phenomenon, but two ways with different implications for policy and that invite a different set of natural next questions.

                I think Nick roams broadly in his thinking (within limits…), and is far from consistent or coherent over the range of his discussions. If find this to be a (frustrating) virtue. He asks questions and worries at these issues while most economists would claim that it’s obvious while at the same time also being vague, ambiguous, inconsistent, and incoherent in their thinking.

                “starting to see the distinctions between use of the terms:

                -aggregate demand
                -planned aggregate demand
                -actual aggregate demand”

                Nick pounds his spoon on the highchair repeatedly on this subject, that “demand” is not just “spending.” (One reason I was confident enough to push back atcha on it). Just in example, here from an email he sent me once, which he gave me permission to quote:

                Take a very simple model. Assume there is no debt at all. Assume a monetary exchange economy in which people use shells as money. Assume an excess supply of all goods (except money), so that people’s planned expenditures are always realised. Therefore actual income is always equal to planned expenditure in aggregate. And, it doesn’t matter what actual income was last period, since last period might have been nighttime for all we know, and maybe people don’t trade at night.

                It is perfectly possible for planned expenditure to exceed expected income. Because every individual is planning to reduce his stock of shells. And if that’s the case, people will be surprised to learn that their actual incomes were higher than they had expected. And will be surprised to discover that their stocks of shells did not fall as planned.

                So I can get planned expenditures exceeding expected income even without debt increasing. (And I could also get debt increasing with planned expenditures equalling expected income too). And *actual* income will always equal planned expenditures, unless there is excess demand and people can’t buy as much as they want.

                I laid all this out in my post “The very short run”. http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/08/the-very-short-run.html

                Gotta go, no more time right now for this rather bottomless topic…

                • “it’s like physics before Newton”

                  Let’s not give Sir Isaac too much credit, at least in economic affairs. His fix-and-forget approach to monetary policy resulted in incalculable human misery for about 200+ years. http://en.wikipedia.org/wiki/Isaac_Newton#Royal_Mint

                • “So I can get planned expenditures exceeding expected income even without debt increasing.”

                  Not sure why that’s mysterious.

                  If planned expenditures materialize, then income must match by accounting tautology.

                  And no debt is required by the same accounting tautology.

    • Steve Roth check here at the bottom:

      http://www.concertedaction.com/2013/05/02/erroneous-use-of-the-sectoral-balances-identity/

      I hope I say this right.

      Spr = Ipr with balanced budget of gov’t and CA balance = 0

      I’ll use S and I for the rest.

      S = Y – C

      I = Igoods plus financial assets including the MOA/MOE

      Y – C = Igoods plus financial assets including the MOA/MOE

      Example:

      I earn $100, consume $55, buy Igoods of $20, buy financial assets of $15, save in the MOA/MOE = $10.

      $25 has leaked.

      I think Keen is really trying to say Y “minus” financial assets including the MOA/MOE = C + I

      • “I think Keen is really trying to say Y “minus” financial assets including the MOA/MOE = C + I”

        Make that = C + Igoods.

  38. Steve,

    Steve Keen confuses. We say higher debt increases aggregate demand which is okay as a statement in English but that doesn’t mean we go and add change in debt to aggregate demand. This is because aggregate demand already has things such as consumption demand, investment demand and so on. The change in debt is one way to finance things such as consumption or investment. Just like looking at the wrong side of the balance sheet, he is looking at the wrong flow and worse double counting.

    • right!

    • Y = C + I + G + NX

      Let’s assume G = 0 and NX = 0

      Y = C + I

      I think the I’s should be Igoods.

      Let’s look at this from a MOA/MOE viewpoint.

      Entites can spend on C with savings, current income, and/or issuing bonds or issuing stock.

      Entites can spend on Igoods with savings, current income, and/or issuing bonds or issuing stock.

      Notice the past, present, and future time periods there.

  39. Steve Roth from above:

    ““Isn’t bank capital “stored” as an asset somewhere?”

    No that’s the whole point that I’ve known forever, but that I’ve just had drummed into me yet again, some more, quite rigorously:

    Shareholder’s equity, aka capital, is just a residual: assets minus liabilities. (Basically net assets.)

    You can’t point to any particular assets that are “capital” (or “required reserves” for that matter).

    If assets increase or decrease (due to income or market-price changes of the assets), capital goes up or down.

    If liabilities increase or decrease, capital goes up or down.

    If those asset/liability changes cause capital to decline to the point that it’s not sufficient under regulations, or it’s gone completely, that’s spelled T-R-O-U-B-L-E.

    Even if there’s still a big pile of assets.”

    I believe equity and bank capital do not have to be the same thing.

    I believe an asset could be written down that is not bank capital or affects bank capital.

    • good stuff, couple of observations:

      – last sentence is questionable and probably not right depending on context

      – people shouldn’t get overly carried away with the residual idea – its more or less true from a stock perspective, but its also true that retained earnings are more residual than new capital issues from a flow perspective

      • I was thinking of a scenario like this.

        Start a new bank, sell stock of $150,000, and spend $50,000 to buy a building for the bank.

        Assets new bank = $100,000 in treasuries and $50,000 building
        Liabilties = $0
        Equity = $150,000

        Something happens to write down what the building is worth to $25,000

        Assets new bank = $100,000 in treasuries and $25,000 building
        Liabilties = $0
        Equity = $125,000

        Bank capital is still $100,000 in treasuries. If there were loans, I would think they would be unaffected too.

      • @Fed Up: “I believe an asset could be written down that is not bank capital or affects bank capital.”

        I think this is almost certainly true given the accounting shenanigans allowed out there — SIVs and such “off-balance” sheet practices. But at a basic conceptual level such practices should bear the burden of justification, be viewed as a priori “non-kosher.”

        @JKH “retained earnings are more residual than new capital issues from a flow perspective”

        Reminds me that I did not mention the other things affecting capital beside income/expenses and market price changes: 1. stock issuance/buyback, and 2. dividends distributed. Yes agree that retained earnings are more (doubly!) residual (income-expenses-dividends distributed). But: what does that mean? ;-)

  40. “Given this acknowledgement of the full span of the QE bond transfer, it must accordingly be the case that the Fed’s activity in outright purchases of bonds in QE is most often associated with banking system reserve creation AND bank deposit liability creation, since for the most part it is non-bank portfolios that are ultimately affected by these bonds transactions.”

    Let’s say I took a gov’t bond to the fed. They gave me currency for it. I take the currency and deposit it at the bank. The bank has no need for extra vault cash/currency. The bank exchanges the currency for central bank reserves. Does the accounting end up the same?

  41. Steve,

    There are debates/discussions which are helpful but:

    “So maybe this is why I tend to defend Keen”

    Keen’s error are very amateurish.

    I hate economics analogies to physical sciences but will try:

    Exerting force on a body increases its velocity or adds to its velocity but I don’t add force in the mathematical definition of velocity.

    I can have expressions in which the force appears. Such as the velocity of a body in free vertical fall is given by

    v = v(initial) + (earth’s gravitational force/particlemass)*time.

    But velocity is dx/dt in vector form and there is no force in its fundamental definition. If I were Keen, I’d say “velocity is speed plus force” or “speed plus acceleration” or some slogan like that. Funnily, he collaborates with a mathematical physicist who doesn’t seem to get the point.

  42. This questioning on aggregate demand by Steve Roth is very useful.

    Examining the chapter on aggregate demand in my macro textbook fairly closely for the first time in decades, I’m now recalling the following picture:

    I think there are two reasons why aggregate demand is used instead of using GDP for the same idea:

    a) There is a distinction between planned aggregate demand and actual aggregate demand. That distinction can be framed in the context of the Keynesian process for how planned aggregate demand evolves to actual aggregate demand, step-wise, and to “equilibrium” aggregate demand at the end of that defined Keynesian process. At any step in that process, you can identify actual aggregate demand, which is equal to GDP. But until the multiplier type equilibrium is reached, planned aggregate demand will be different from actual aggregate demand. So the process can be envisaged as a sequence in which planned aggregate demand moves in steps toward an eventual point of equilibrium GDP. At the same time, there is an identifiable actual aggregate demand, which is equal to GDP, at every step of the process, and which is out of equilibrium at every step preceding equilibrium. At every out of equilibrium point, the difference between planned aggregate demand and actual aggregate demand is unplanned inventories. But actual aggregate demand at every step is equal to GDP, and actual aggregate demand includes unplanned inventories (surplus or deficit inventories from plan).

    b) The other reason is that aggregate demand is essentially the sum of it component demands. So, for example, at the beginning of the process, it’s assumed that investment, government spending, and exports are all autonomous demand sources – component sources – of what in total is planned and/or actual aggregate demand at every step of the process. The Keynesian multiplier process moves that along with the stepwise addition of induced consumption demand. Anyway, the aggregate demand terminology is used because it is consistent with demand decomposition across component sources of demand.

    Now how that squares with GL, I’m not certain.

    Comments welcome if I have that right or wrong.

    • P.S.

      I’ve used the term “scenarios” earlier to refer to any of these situations of planned, actual, or equilibrium aggregate demand.

    • In G&L the concept of equilibrium is different and is defined as a state when stock/flow ratios are not changing. In the most general model, there is no tendency of economies to reach a state of equilibrium. In fact their models is disequilibrium economics.

      The primary act is the decision to produce something. The concept of aggregate demand can be understood if seen from producers’ viewpoint. On any given day it is the demand for their products and can be dependent on many things such as incomes of the buyers and their moods, producers’ competitiveness and so on. To the extent they sell everything demanded it is also equal to sales.

      GDP or output are different because that is what the producers actually produce. On any given day there may be more sales than what is produced on the day or the opposite and the level of inventories adjusts.

      There isn’t a need for expected aggregated demand because there is expected sales.

      Producers will adjust their level of production depending on signals such as the level of demand or sales but also other things. How much they produce can differ from expected sales as they may produce more. Also in G&L models, sales is always equal to demand because of the quantity clearing process. So it is demand which is driving supply because faced with low demand, entrepreneurs may not produce – quite unlike the neoclassical world where supply creates its own demand.

      The word aggregate demand is used by economists because there can be situations where demand is greater than supply and sales is the supply. This is Nick Rowe’s world which he considers to be general rather than special.

      But another reason is that aggregate demand is a sum of components such as consumption and depends on things such as the propensity to consume and it is better to explain something as a fall in the propensity to consume and things such as that.

      It is better to keep notions such that AD is not equal to AS because it is reminiscent of the neoclassical ‘equilibrium’ approach.

      Also the GT – a heavily classical book has some strange definitions of aggregate demand http://www.marxists.org/reference/subject/economics/keynes/general-theory/ch03.htm – perhaps the reasons which has confused everyone.

      • Should add that GDP is a special type of output because it is based on “value-added” and also things such as intermediate consumption etc are not calculated. Sure Keen doesn’t know these things.

      • I’m aware that the equilibrium concept is different, because it incorporates stock effects comprehensively.

        As I said, what I’m getting from reviewing the conventional description is that the key idea underlying the use of the term “aggregate demand” is the distinction between and analysis of the difference between planned aggregate demand and actual aggregate demand. Actual aggregate demand becomes GDP, but the reason to use AD terminology is to analyse that process of planned versus actual. Also, the aggregation of individual component demands lends itself to the aggregate demand terminology.

        Question # 1:

        Does your reading of GL suggest that the concepts of planned aggregate demand versus actual aggregate demand do not exist there?

        Question # 2

        What is the GL definition of aggregate demand?

        (because they use the term)

    • I don’t think that aspect really features in G-L.

      For a start, the only reason for unplanned inventories in their model is firm’s estimation error with regard to sales. If the estimation error is set to zero, planned inventories are zero. This appears to have little to do with what’s happening with demand.

      Secondly, I don’t think dynamic adjustment as a result of the short run multiplier plays much of a role in their models. In the simpler models, behavioural equations only reference current period flow variables (indeed, part of the whole philosophy is that the past is captured by starting stock values, rather than lagged flow values). This means the results for each period are all at equilibrium in terms of the short run multiplier.

      In the more detailed models, lagged flow values are used but as far as I can tell (and I honestly haven’t read the book cover to cover), this is only in the context of expectation formation. This will, of course, tend to lead to some form of incremental movement towards a short run multiplier equilibrium. However, as I said, I don’t think it plays any role in determining unplanned inventory accumulation.

      (I’m not sure if the term “short run multiplier” is a term of art – what I mean is the normal Kahn multiplier as oppose to the multiplier which determines the long-run steady state in G-L models).

  43. Still no time to engage here but wanted to share this recent post from the always excellent Josh Mason. Cuts to a logical/conceptual crux we’ve been worrying at, I think:

    Aggregate Demand and Modern Macro

    Start with a point on language.

    People often talk about aggregate demand as if it were a quantity. But this is not exactly right. There’s no number or set of numbers in the national accounts labeled “aggregate demand”. Rather, aggregate demand is a way of interpreting the numbers in the national accounts. (Admittedly, it’s the way of interpreting them that guided their creation in the first place). It’s a statement about a relationship between economic quantities. Specifically, it’s a statement that we should think about current income and current expenditure as mutually determining each other.

    This way of thinking is logically incompatible with the way macroeconomics is taught in (almost) all graduate programs today, which is in terms of optimization under an intertemporal budget constraint. I’ll avoid semi-pejorative terms like mainstream and neoclassical, and instead follow Robert Gordon and call this approach modern macro.

    More…

    http://slackwire.blogspot.com/2013/04/aggregate-demand-and-modern.html

    • What exactly or not exactly he saying about the definition of aggregate demand?

      I can’t tell.

      And, given what he is saying, what does he mean?

      • “What exactly or not exactly he saying about the definition of aggregate demand?And, given what he is saying, what does he mean?”

        Back to what I mentioned about “demand” (as a “quantity”) not being terribly meaningful? That you have to talk about demand curves (relationships) to say something meaningful?

    • “People often talk about aggregate demand as if it were a quantity. But this is not exactly right. There’s no number or set of numbers in the national accounts labeled “aggregate demand”. Rather, aggregate demand is a way of interpreting the numbers in the national accounts.”

      Second part about interpreting true.

      First part is wrong. Take the national account release and get the numbers output and change in inventories and subtract the second from first.

  44. It is sum of consumption demand plus investment demand plus demand for exports minus demand for imports. (govt is included in cons and invest).

    So it differs from output or GDP because it doesn’t include change in inventories. It is actually equal to sales. Aggregate supply is equal to output and not equal to aggregate demand.

    Now the question is why have a different name than sales if this is the case. This is because sales is just sales – it doesn’t tell you about components of demand etc and questions such as what happens if propensity to consume increases/decreases.

    Also there may be some shortages such as Nick Rowe coming home unsatisfied because he couldn’t get a rainbow-colored BMW. In such case sales is the supply and aggregate demand something which is higher. However, these are special cases rather than general and ignoring Nick Rowe, aggregate demand is equal to sales and aggregate supply equal to output with the difference equal to delta inventories.

    • Actual aggregate demand in the conventional definition includes change in inventories. It equals GDP.

      So are you saying that the GL definition of actual aggregate demand then differs from conventional, as you seem to describe – i.e. it excludes the change in inventories?

      I think so.

      Then what about the conventional definition of planned aggregate demand – does a counterpart to this exist in GL?

      • Yes differs from conventional definitions because in the latter AD=AS

        Don’t know about planned aggregate demand because each sector is doing its own thing and planning but nobody is planning in the aggregate. Producers can only have sales expectations and production decisions, so I do not know what planned sales mean.

        • “Don’t know about planned aggregate demand because each sector is doing its own thing and planning but nobody is planning in the aggregate.”

          There’s definitely an interpretation of this in the conventional explanation, from what I’m seeing.

          BTW, I’m asking all these questions, because I’m trying to get establish a better “anchor” on the conventional approach, before going back to GL. So your answers are helping me do that.

          • Oh try doing the opposite!

            Sec 3.3.2 of G&L very useful but careful about the context because in SIM there are no inventories and in that model AS=AD and this can add to confusions and I guess Steve Roth seems to be gotten into that. Perhaps some discussion in 3.3.2 added too early and should have been written in Chapter 9.

            • Yes – I recall SIM from the book, and Steve’s pointing to that – but it is a special case

            • did the opposite before

              now doing the opposite of the opposite

              rinse, repeat

              • A lot is about cleanly separating past and future. I think neoclassical models suffer that and as Joan Robinson said, economists somehow think that the future has already occurred. Steve Roth suffers that and as does the other Steve (are you listening Steve?). Same cannot be said about the two Nicks – the one here seems to know Keen’s errors thoroughly!

                One silly thing about Keen is that he thinks nobody has thought about these things before him and on top of it ends up with his stuff!

                In this latest http://socialdemocracy21stcentury.blogspot.in/2013/06/steve-keen-on-his-book-debunking.html (21:00) Keen names names of economists who agree with him. Soon we will have 5-10 PKEists with the slogan “aggregate demand = GDP plus change in debt”.

                • exactly right on past and future

                  you noticed the WIKI entry on aggregate demand, re post Keynesians?

                  • He he! Hadn’t seen it earlier.

                  • One of the things Keen and his fans have done is to unnecessarily assert that those who reject him believe that economies are not debt driven or that they believe in Say’s law or they do not understand endogenous money and all that.

                    • right again – the first part most directly, in the case of the equation

                      straw man reflexive criticism

                • @Ramanan: “economists somehow think that the future has already occurred. Steve Roth suffers that and as does the other Steve (are you listening Steve?). ”

                  Please please explain how I’m going wrong there. Don’t understand exactly what you’re saying, want to.

                  (But of course it has occurred [provisionally] — in our heads! That’s the whole point of expectations affecting present behavior. Thermostats don’t do this, but insects and even amoebas probably do, and primates and especially humans do in spades. Sorry, semi-tangent…)

      • Also neoclassical models assume that there can be more demand than supply – as if these things are normal, such as Nick not able to find a car. In that case, one has to be careful.

        • more planned demand than supply, from what I’m seeing

          which is the conventional Keynesian cross with excess planned demand

          but not excess actual demand – that doesn’t exist

          • “Also neoclassical models assume that there can be more demand than supply”

            “more planned demand than supply, from what I’m seeing”

            Doesn’t the most common definition of economics mean planned aggregate demand is unlimited?

        • Nick as in Nick Rowe – there are two Nicks here. As is the case with Steve.

        • I’m not sure the basic planning/actual sequence for demand is all that different in GL

          I just think the accounting closure is more complete, as well as the identification of what’s endogenous

    • You seem to be saying here that actual aggregate demand excludes change in inventories – which is not the case in the conventional definition (from my reading of it, because it equals GDP) – but seems to be the case for your GL interpretation:

      http://monetaryrealism.com/the-accounting-quest-of-steve-keen/#comment-19447

      • If it’s of any interest here’s an approach that I’ve used with good results in some recent empirical modelling of the UK.

        This involves a concept equal to the sum of actual sales plus planned stock-building. Although I don’t tend to use the term aggregate demand, if I had to identify something in the models with that label it would be this.

        I assume that planned stock-building is based on a target inventory ratio, actual sales and actual inventory levels, with some incremental adjustment mechanism (like G-L, but without expectation error). I then make the further assumption that actual production is a function of this aggregate demand measure and prevailing capacity constraints (proxied by prior production levels). Since this results in a difference between this aggregate demand and actual production, there will be a corresponding difference between planned and actual stock-building.

        (Obviously I can’t measure planned stock-building directly but it can be inferred from the regression results.)

        • “This involves a concept equal to the sum of actual sales plus planned stock-building. Although I don’t tend to use the term aggregate demand, if I had to identify something in the models with that label it would be this.”

          Oh that is a good way to use it. Interested in your approach.

        • Nick,

          How do I contact you? Can you send me your email as a comment on my blog. I don’t publish comments so don’t have to worry about exposing your email to robots.

      • I can’t explain this fully and coherently, but again: For G-L (p. 64), inventory is the third of four theories (they consider it “the most realistic one”) that answer the question (p. 63) “starting from a situation where production could be different from supply, and where supply could be different from demand, how do we arrive at the equality between sales and purchases?” That seems to be a crux question of the whole book. In a sense, they seem to be trying to define supply and demand rigorously, though they don’t actually say that’s what they’re trying to do… (The whole point of the behavioral matrix is that it has supply and demand subscripts.)

        This is not asking “what do supply and demand equal?” That would be attempting to define economic constructs/quantities as equal to accounting constructs/quantities. (Is this what Josh means when he says AD is not a quantity?)

        It’s asking, “what’s the relationship between supply, demand, production (and sales)?” (Think curves, not quantities?) If supply and demand are out of equilibrium, not balanced, by what mechanism are sales and purchases equal? Hence their name for the inventory theory?: the “general disequilibrium approach.”

        • You are mixing various approaches and even approaches within the book. Sec 3.3.2 ends with “It is best not to confuse these two assumptions.”

          In simple (but not complicated) G&L models, supply is equal to demand due to instantaneous production but this is not true in general models. There the adjustment happens via changes in inventories. Supply is *not* equal to demand, later on. You can think of simple models as barber models where haircuts are produced on demand.

          Also, theirs is not a story where supply and demand curves meet. I do not know the background of Josh but he looks like a mix of heterodoxy and neoclassical.

          “In a sense, they seem to be trying to define supply and demand rigorously, though they don’t actually say that’s what they’re trying to do”

          But they are precisely doing that. And because there is no reason AD is equal to AS, we ask what happens to bring them equal – in simple models by instantaneous production and in complicated models there isn’t any equality and firms have a buffer of inventories.

          demand is sales and supply is output. It is going to be my slogan to counteract the slogan “aggregate demand is gdp plus change in debt”. But more importantly demand while it turns out equal to be sales, it is not its definition. Its definition is the component demands. Households consume a proportion of their income and it is their consumption demand, for example. AD is a quantity and so is AS.

          But more generally I don’t know why you bring this complaints of accounting and past and future.

          • i.e., you are reading G&L with the Josh lens of having supply and demand curves and supply and demand not being a quantity which can easily be obtained from national accounts.

          • “in complicated models there isn’t any equality and firms have a buffer of inventories”

            my reading of the Keynesian cross is that this is true there as well

            the difference between planned demand and supply is the inventory buffer

            e.g. it is an inventory surplus when supply exceeds planned demand

            and in terms of definitions:

            supply = actual demand = GDP in that model

            and this is a bit of a stab, but it looks to me like sales = output = actual demand when planned demand exceeds actual demand, and sales = planned demand < output when actual demand exceeds planned demand in the Keynesian cross

            again, I'm not sure the demand relationships between GL and the K cross are different – just that GL provides comprehensive stock flow accounting for the time process

            • “and this is a bit of a stab, but it looks to me like sales = output = actual demand when planned demand exceeds actual demand,”

              Does planned demand exceeding actual demand get resolved in the MOA/MOE market?

              “and sales = planned demand < output when actual demand exceeds planned demand in the Keynesian cross"

              Can actual demand exceed planned demand?

              • “Can actual demand exceed planned demand”

                yes – where there is a withdrawal in the context of the Keynesian cross, setting a negative multiplier in motion, so that there is an inventory buildup, with an unplanned inventory increase being the difference between actual demand and planned demand.

                • I’m still not getting that part. If I decide to spend only $35,000 and make $50,000, my planned demand and actual demand are both $35,000.

                  Is there a need for the terms planned demand, actual demand, actual sales, planned sales, and potential output? If so, can planned demand = actual demand = actual sales = planned sales < potential output?

                • Theory works something like this:

                  A firm pays workers $ 1 million in the process of making an investment.

                  Workers start saving and spending, with a marginal propensity to consume of 2/3. Through the K multiplier, workers will eventually spend a total of $ 2 million on goods and services while saving $ 1 million.

                  In the course of spending, until the full $ 2 million is spent, planned (or desired) demand will always be greater than supply, and that shows up as a running inventory shortage in respect of planned (desired) demand.

                  Through the process, actual GDP reflects actual amount sold and actual inventory build. That is actual supply and actual demand – where, as NR would say, quantity supplied equals quantity demanded.

                  So, there is always some unsatisfied planned demand until the full process has been completed and the full $ 2 million spent.

                  Eventually, planned demand converges with actual demand/supply/GDP and that’s the point where the full $ 2 million has been spent.

                  • As an example, suppose the $ 1 million is an incremental investment in an auto factory, with wages paid to those who produced the investment goods.

                    One of the wage earners then has $ 100,000, and goes to a car dealer to buy a new car – but can’t find the car he wants because it hasn’t been produced yet. So he orders it.

                    That’s a case where planned demand of $ 100,000 (due to the K multiplier) exceeds actual demand of $ 0, and the difference is an inventory shortage. Actual demand is what actually happens in the transaction and which contributes to GDP in the current accounting period.

                    So that order will eventually be filled, at which time plan will converge to actual in respect of that phase of the multiplier process.

                    And so on until the multiplier is exhausted.

                    Inventory fluctuations are handled both in the conventional K cross and in GL.

                    GL is just more fully precise about it from a monetary economics standpoint.

              • I don’t know if the phrase “planned demand” has any meaning. One can plan production and the government can manage demand but what is “planned demand” – planned by who?

                • its standard Keynesian terminology in the textbook I’m looking at

                  as in the sense that the marginal propensity to consume puts multiplier in process on a planned basis until equilibrium is reached, which is where planned demand and actual demand converge

                  its the curve above the supply curve on the Keynesian cross

                  the supply curve is also called actual demand as in actual GDP

                  (terminology here – not say that deposits create loans yet :) )

                  • and I do think that this SORT of thing is the key to the confusion that Steve is referring to

                  • book also uses ‘intended’ or ‘desired’ as synonyms, but sticks with ‘planned’

                    • @JKH: “book also uses ‘intended’ or ‘desired’ as synonyms, but sticks with ‘planned’”

                      Yes I took Nick Rowe to task at one point for using all these terms (and as I remember, two or three others) in a post semi/quasi-synonymously. Suggests that he’s not coherent about how (in G-L terms) the transactional and balance-sheet matrices connect to the behavioral (transactional) matrix.

                      “and I do think that this SORT of thing is the key to the confusion that Steve is referring to”

                      Uh, yup. Given the controversy over these words/issues across the economics community (those who actually stop to think deeply about it, like NR), and their widely disparate usages, I really have wonder whether S and D have been coherently theorized at a fundamental level. (At the same time I acknowledge that I may just be incapable of understanding what is in fact coherent theory, or haven’t done the work to achieve that. But: whose coherent theory?)

                      It certainly is all about time (has the future happened yet?), and Ramanan is right that I’m profoundly confused about that dimension. I’m also quite certain I’m not alone in that.

                      I’m going to keep re-re-re-reading G-L, assuming they’re most likely to be coherent.

                      Just to add to the mix, see Scott Fulwiler’s “The Sector Financial Balances Model of Aggregate Demand–Revised”

                      http://neweconomicperspectives.org/2009/07/sector-financial-balances-model-of_26.html

                      Which concludes: “aggregate demand is set by the government’s deficit relative to net savings **desires** of the non-government sectors”

                      (I’d really like to see this theorized re: the net savings desires of the real sector, with the financial sector represented as a different desires hence a different curve).

                      Interestingly, I don’t remember seeing “net saving” when going through all the G-L index entries for “demand.” I’ll now have to go through all the (undifferentiated, goddamit) index entries for “saving” and see what I turn up, see if they discuss it relative to “demand,” desires, intentions, plans, etc…

                    • AD may be a little difficult to extract surgically and cleanly from GL, which is one reason why I went back to an old textbook for contrast purposes, where there is a very rich chapter titled aggregate demand. And that adventure back in time has taken me into Nick Rowe land – he’s quite prolific at dealing with the “quantity demanded” types of nuances. Of course, you were the one that really drove me to my textbook – would have gone mad otherwise. Been several weeks since I scanned GL, but I’m pretty sure I saw “net saving” in there.

                    • I keep saying (sort of) that the definition is a modal framework analogous to that for the case of money. There are different modes of money and of aggregate demand. I think you have to think about the definition as a set of modal applications – as in bank reserves versus bank deposits; planned AD versus actual AD …

                      As in James Joyce … “the ineluctable modality of the visible”

                      :)

                      I suspect this will not be entirely satisfying for you.

                    • “I suspect this will not be entirely satisfying for you.”

                      Very satisfying, as it cuts deeply to the level of meaning. Especially cause I think you’re talking both conceptual and linguistic modality, and you know my obsession with those issues — carefully and coherently theorized concepts, expressed using precisely defined words.

                      Might this be an example of what you mean?:

                      When thinking about (Personal) Saving, you need to think in both NIPA and FOA modes simultaneously, drawing on a deeply internalized and coherent understanding of those two modes’ accounting and conceptual definitions (what the words mean) and relationships.

                      ??

                      Also helps me characterize the nature of my confusions. I tend to get flummoxed where modalities meet — “real” vs. “financial/monetary” being a perennial of mine. (See: “Saving.” )

                    • @JKH: “you were the one that really drove me to my textbook”

                      Well, if only by driving you to distraction… ;-) Glad something useful came out of my confusion.

                      Just one more thought re: Fulwiler: do we have to start talking about government’s “desired” deficit? (Ducking…)

                    • in that context – government deficit spending aims at a multiplier effect in the sense of planning or desiring a multiplier effect from that “injection” – which materializes in theory with a lag, over which time period actual demand needs to catch up to planned demand – i.e. planned AD exceeds actual AD = GDP = AS over that period, until the two finally converge once the multiplication is exhausted

                    • perhaps “no light, but rather darkness visible” is also suited to your view on the issue of defining aggregate demand

                      John Milton, as opposed to Uncle Milton Friedman

                      but you knew that

                      :)

                    • “no light, but rather darkness visible”

                      Shit: “where peace and rest can never dwell…”

                    • “When thinking about (Personal) Saving, you need to think in both NIPA and FOA modes simultaneously, drawing on a deeply internalized and coherent understanding of those two modes’ accounting and conceptual definitions (what the words mean) and relationships.”

                      yes, deeply internalized – as in, can you remind me what FOA is?

                      flow of funds?

                      but yes, along the lines of that idea – there’s an “adjacency” of NIPA and flow of funds that is quite difficult for some people to discern – I referred to these as “orthogonal” measures previously – sometimes they are simultaneous and coincident, but that doesn’t mean they’ve merged into the same thing

                      also the idea of time – GL brings time to front stage – but time is still embedded in the conventional stuff, just harder to extract, and sometimes strangely incongruent, as in ISLM perhaps

                    • “yes, deeply internalized – as in, can you remind me what FOA is? flow of funds?”

                      Typo, geez. FOFA.

                    • “there’s an “adjacency” of NIPA and flow of funds that is quite difficult for some people to discern”

                      Not surprising that it’s hard. I think most economists have to pick up this understanding on the fly after leaving school. There should be a rigorous, required (one-quarter?) course in every economics curriculum addressing exactly this: what are the NIPA and FOFA concepts, and what are their accounting relationships? Also: What economic assumptions underpin their different accounting constructs?

                      Ditto, on the fly, for me: I’ve developed a somewhat dim understanding by building spreadsheets combining data from the two, and sussing out how they relate. (Plus pestering you with questions.) Problems of that type would be the basic homework of this course.

                      But that would require economists to study…accounting!

                      I’m quite certain that course is not available at UW. (Or I’d take it!) The string “account” does not appear in the econ course catalog. The word “national” (except “international”) does not appear in the accounting course catalog.

                      How did you figure it out? Not just those few accounting course in college….

                    • “How did you ..”

                      Well, to the degree that I have …

                      heavy financial management background, with a natural and professional interest in the macro view, and a tremendous natural interest in macroeconomics (with no econ degree) …

                      plus…

                      given those two things, it is inconceivable to me how anybody could be interested in macroeconomics without being deeply curious about how the macro flow of funds works

                      the two most important courses in my finance MBA were macroeconomics and financial accounting … the finance was a given, and I was already quite familiar with it, because I twin tracked a CFA – the CFA has a very enlightened approach to economics for its value in context – e.g. economics is positioned alongside financial accounting, security analysis, portfolio analysis, etc.; it ranks pari passu in the educational structure of the overall program

                      finally, macro flow of funds has a constituent element in micro flow of funds – meaning financial management of the firm/bank – so the relationships can be mapped macro to micro and vice versa, which is especially neat if your micro financial management responsibilities have a big macro scope

                      the rest is just thinking about how it must work

                    • Art Patten says:

                      “the CFA has a very enlightened approach to economics for its value in context ”

                      But seems to me that nearly all CFA charterholders are dyed-in-the-wool “neos” when it comes to macro, and it soaks quite heavily into the curriculum?

                      Said another way, 99% of them are familiar with James Grant, while fewer than 20% have ever heard of Wynne Godley. :)

                  • hoping I will not be excommunicated from PKE land for referring to an old textbook – its a pretty good one actually

                    • oh which textbook is this by the way?

                    • Economic Analysis and Canadian Policy
                      Third Edition
                      David Stager
                      University of Toronto
                      1979

                    • Oh seems out of print but I guess used copies should be available. 1979 is more Keynesian and likely to more right than most modern books, especially considering someone writing Keynesian economics is likely to not have been hypnotized by Friedman!

                    • First time I’ve looked closely at it since 1981 macro course

                      I read the chapter on aggregate demand, which includes K cross and injections/withdrawals approach, as well as following chapter on ISLM

                      Seems very well explained, notwithstanding weaknesses noted by modern PKE; certainly understand it now better than I did back then

                      Also seems soberly balanced between Keynesian and monetary

                      Funny thing – I did a thesis for that course, which I also haven’t looked at in 30 years, and I think in effect I constructed something similar to a GL transaction matrix approach – I.e. attempted to combine NIPA and flow of funds – will be reviewing that

                  • Nick Rowe might even endorse it

                    :)

                  • Yeah think it should be okay to use it.

  45. Steve Roth June 11, 2013 at 11:27 am,

    Good points. There are just too many issues around these things and a slight modifications of few points of theory can lead to VASTLY different interpretations of how the world works. But also means it is very interesting.

    For example economists are introduced to simple models of AS/AD but quickly “unlearn” it and instead concentrate on “price clearing”. If you see G&L, pricing is somewhat a different thing than what is going on. If you see the models of G&L and other similar models, entrepreneurs are making and changing production plans depending on “quantity signals” and gradually supply pressures are added in the model. The role of prices is also different.

    But in neoclassical economics “price signals” become more important along with the “production function”. So it is difficult for fiscal policy to do much because it cannot change the production function and since the government’s action is supposedly inefficient, it necessarily produces higher inflation and so on. In addition there is a “natural rate of interest” and the central bank’s task is to move policy toward the natural rate. So it is the production function which moves economies as opposed to demand which is important in PKE.

    The connection of demand with “saving net of investment” is indirect. Government deficit doesn’t appear directly in aggregate demand – only the government’s expenditure appears. However because of the dynamics, government’s tax rates also becomes important. The deficit is an outcome of the dynamics. It is a loose statement to say government deficits adds to demand or sets demand.

  46. “The deficit is an outcome of the dynamics. It is a loose statement to say government deficits adds to demand or sets demand.”

    Right. Deficit, in a real sense, is endogenous, not causative. I think you’ve said that.

    This brings me back to Keen’s Minsky modeling, which I like because it makes the cause–>effect assumptions (and feedback loops) explicit and obvious (and requires the modeler to specify them precisely — hard-coding presumed reaction functions by different sectors). Then you can just run the model and watch the system behavior over…TIME. Does a pretty good job of displaying three dimensions per graph. Makes the time dimension much less mysterious, compared to two-dimensional comparative statics (S/D diagrams) in particular, in which time is a very difficult concept to grasp.

    • lol. Totally agree. Economists draw strange diagrams. It is like fashion or something. Drawing such diagrams looks like establishing some supremacy of understanding or something over other colleagues. And (most, not necessarily all) these diagrams are themselves wrong. Some curve is moving but a lot of things are changing such as output, money supply etc – meaning if one curve is moved the remaining curves in the diagrams should move as well instead of being fixed.

      • “Some curve is moving but a lot of things are changing such as output, money supply etc – meaning if one curve is moved the remaining curves in the diagrams should move as well instead of being fixed.”

        Plus feedback loops. Keen makes this point hard in Debunking. Price/wage changes alter distribution of income. And distribution changes result in wage/price changes.

        S/D diagrams only make sense in context of other S/D diagrams, for substitutable “goods.” (Substitution is, after all, the sine qua non of demand curves, as expressed in the indifference maps from which they’re derived.) So S/D diagrams are asking you to hold all of that in your head, imagining all those other S/D diagrams behind it, understanding all the complex interactions at play when one curve, or one’s location on a curve, changes.

        Could somebody do a simulation, please, and let me watch it run? Oh, wait…

  47. “it is inconceivable to me how anybody could be interested in macroeconomics without being deeply curious about how the macro flow of funds works”

    Ditto, of course. It’s where thinking about this stuff has led me (Joyce: ineluctably).

  48. JKH
    the rest is just thinking about how it must work

    The best way. Much better than just reading what someone else thinks (in economics, more than anything).

  49. JKH – Wonderful post. I have learned alot from you and this blog. Monetary Economics’ by Wynne Godley and Marc Lavoie has been very enlightening for me as well, although I must admit I still have not finished reading it (my third son was just born and I’m buying a house – so I’ve been busy). I even stole the theme for the title of my blog – http://stockfloweconomics.blogspot.com/
    A few relevant passages:
    “Every transaction has both a seller and a buyer, or more generally, there are at least two sides to every market transaction. This indisputable fact, although extremely general and abstract, is one of the only true economic laws but one which is often forgotten. One reason why this fact is often overlooked is that its ramifications are most easily seen in the world of accounting – transactions are measured via income statements which lead to changes in balance sheets. Every transaction creates at least 4 accounting entries (two for each party). Accounting relations, in themselves, say little about what caused things to happen or reveal much about key economic variables like relative scarcity or utility, but they quantify a true constraint that exists in our market economy and provide a pseudo scientific framework with which to study economies. This methodology, stock-flow consistent macroeconomic modeling, has been developed by multiple Post-Keynesian economists such as Wynne Godley and Mark Lavoie1.”

    “However, my recommendations will not simply entail money flow redistribution. As I have discussed previously, the level and efficiency of real stock utilization (current standard of living) and the rate and wisdom of real stock production (future standard of living) is a function of current money flows, expectations of future money flows and the institutional structure and functioning of markets (which equate money flows with real flows via value judgments). My policy recommendations will therefore address all of these factors – simple recommendations such as more progressive taxes (which shift money flows to the less wealthy) or a workfare program (like the MMT job guarantee proposal) can potentially address the output gap issue, but by treating the symptom instead of the underlying cause.”

    • thanks

      and good stuff – interesting set of graphs across your posts

      keep going!

    • Looking at your blog, I see:

      “Real output flows of the US are below potential as evidenced in the low labor participation rate, high unemployment rate and low factory utilization rate (i.e. we’re not using or maintaining a portion of our real asset stocks, Chart 1a, 1b & 1c). This “output gap” …”

      It seems to me you are assuming real AD is unlimited?

      And, “Throughout this paper I use the term money to refer to items such as physical currency, bank deposits (including savings/checking/CD’s etc.), reserves and Treasury bills/bonds. The last item is included due to the ease by which Treasuries can be exchanged for cash and the debtor’s ability to pay (i.e. risk free). I will assume that all financial assets not backed by a bank or the government cannot function as a money flow. I will ignore such items.”

      I like to refer to reserves as central bank reserves. I don’t consider central bank reserves to be “money” because I don’t believe they are MOA/UOA and I don’t believe they are MOE except in the banking system covered by the fed.

      Here is one place where I agree with Scott Sumner. I don’t believe Treasuries are “money” because they usually need to be exchanged for currency or demand deposits. The reason they need to be exchanged is because they are not expected to be 1 to 1 convertible if sold before maturity.

    • “Money is created by the government and banks.”

      The way the system is set up now, I don’t believe the gov’t creates “money”.

  50. I’m trying to get the whole flow of funds (FOF) and NIPA accounting.

    Let me try this:

    Start a new bank and buy treasuries.

    Assets: $100,000 in treasuries
    Liabilities: $0
    Equity: $100,000 in bank stock

    Limited 20 borrower’s balance sheet

    Assets: $0
    Liabilities: $0

    FOF , no NIPA?

    Now apply for mortgages.

    Assets: $100,000 in treasuries plus $2,000,000 DD in its own account
    Liabilities: $2,000,000 in DD
    Equity: $100,000 in bank stock

    Each of 20 borrower’s balance sheets

    Assets: $100,000 mortgage loan
    Liabilities: $100,000 mortgage loan

    No FOF , no NIPA?

    Approve the loans and “barter” DD for loans.

    Assets: $100,000 in treasuries plus $2,000,000 loans
    Liabilities: $2,000,000 in DD
    Equity: $100,000 in bank stock

    Each of 20 borrower’s balance sheets

    Assets: $100,000 in DD’s
    Liabilities: $100,000 mortgage loan

    FOF , no NIPA?

    The 20 borrowers spend on 20 houses.

    No FOF , NIPA?

    As the borrowers pay down principal, both the loan and the demand deposit liability get marked down?

    FOF , no NIPA?

    Thanks!

    • You got it pretty much.

      One qualification – the 20 houses have to be newly built houses in order to fall into NIPA, because only houses that are new to (newly built in) the current accounting period are part of GDP and NIPA

      (there’s some macro level sector accounting where roughly the actual 20 house investment would show up as a business sector contribution to GDP/NIPA, and the household sector then buys it from the business sector)

      the purchase of an existing house (one built before the current account period) would show up in flow of funds – as an asset swap of money for a house, where some of that money might have been created by a new mortgage.

      The sale of an existing house is the inverse of the purchase of the house from a flow of funds perspective, so technically a flow of funds report wouldn’t capture this unless it drilled down to flow of funds or sources and uses of funds at the level of individual households – which is theoretically possible but of course doesn’t show up in fact in macro sector flow of funds reporting due to consolidation of individual households into the sector view.

    • Interesting.

      The neoclassical profession is nothing but a church of dogmas and tenets or a “politburo of correct economic thinking” as Jamie Galbraith put it. Even Krugman conceded recently that anything Keynesian is discouraged in journals. They occupy a position of power because they are also chosen to advise the government and exploit this power subtly to put down anything which is against their religious beliefs.

    • This book with Marc Lavoie as one of the editors is a good one and talks about several issues as this. http://books.google.co.in/books?id=YAb3nMDoK8wC&printsec=frontcover#v=onepage&q&f=false

    • If that Journal of post keynesian economics was an electronic free access journal then they wouldn’t need to worry about there being a conspiracy against them in Universities or whatever. The authors would pay a small fee for publication and then it would be out there for anyone to read.
      Some biology journals are like that. This is what PLOS Genetics says:
      http://www.plosgenetics.org/static/information

      Open Access

      PLOS applies the Creative Commons Attribution License (CCAL) to all works we publish. Under the CCAL, authors retain ownership of the copyright for their article, but authors allow anyone to download, reuse, reprint, modify, distribute, and/or copy articles in PLOS journals, so long as the original authors and source are cited. No permission is required from the authors or the publishers.

      Publication Charges

      To provide open access, PLOS journals use a business model in which our expenses—including those of peer review, journal production, and online hosting and archiving—are recovered in part by charging a publication fee to the authors or research sponsors for each article they publish. The fees vary by journal.

      PLOS is committed to the widest possible global participation in open access publishing. To determine the appropriate fee, we use a country-based pricing model, which is based on the country that provides 50% or more of the primary funding for the research that is being submitted. Research articles funded by Upper Middle and High Income Countries incur our standard publication fees. Corresponding authors who are affiliated with one of our Institutional Members are eligible for a discount on this fee. Such authors will be informed of the discount applicable after submission of their manuscript.

      Fees for Low and Lower Middle Income Countries are calculated according to the PLOS Global Participation Initiative pricing program for manuscripts submitted after 9am Pacific Time on September 4, 2012 (this program is not retroactive).

      Group One: Countries from this list will not be charged for publishing
      Group Two: Countries from this list will be charged a flat $500
      Our fee waiver policy, whereby PLOS offers to waive or further reduce the payment required of authors who cannot pay the full amount charged for publication, remains in effect. Editors and reviewers have no access to whether authors are able to pay; decisions to publish are only based on editorial criteria.

      • “If that Journal of post keynesian economics was an electronic free access journal then they wouldn’t need to worry about there being a conspiracy against them in Universities or whatever”

        Oh did John Hicks have the Apple Maverick with him?

        • Ramanan, all the backdated historical articles could be put online and all the new ones could from now on be online only so as to cut costs. Do many (any?) people read recent articles in paper form now?

          • Stone,

            Two things. Your point obviously don’t apply to Hicks’ times.

            Second it is simply easy to say put everything online and all that but everyone wants visibility and if I were the editor of a journal I want it to be displayed in the library so that people can take a look.

            You can simply handwave away and say everything is electronic etc but have university libraries with physical books disappeared altogether? In fact if everyone is rushing to electronic format, it is the best time to display physical copies not just for JPKE but for everyone.

  51. The Effective Demand guy seems to appear everywhere:

    Comment #1: “Submitted by Edward Lambert on Mon, 05/13/2013 – 1:44pm.”

    (no direct link, scroll down)

    http://ineteconomics.org/blog/inet/dancing-dark-creating-economics-21st-century

    • His work seems a tad complicated. What’s the insight there?

      • Don’t really know but has the same old natural rate of interest and production function stuff. Same old.

        • Output in terms of equations seems extraordinary.

          Did you win the twitter war on endogenous money yet? I would hope so.

          • Didn’t participate much actually.

            The thing is that they know they are wrong but don’t admit and pretend they knew all along and taught in Econ 102 and still make statements as if the central bank determines the money supply, long term or something. And then sometime later make the same mistakes.

            Btw Nick Rowe on Steve Keen: http://pragcap.com/theres-no-conspiracy-against-post-keynesians/comment-page-1#comment-146976

            • well, NR at least has the functionality and the timing such that it accommodates consistent accounting

              LOL – you were almost headlined in an MM post

            • This is absolutely classic, from NR in that comment:

              “If there is a war, it’s a war of all against all. Every ambitious academic wants to be the winner of that war, all by himself. And sure he’ll look for allies, but sometimes he’ll also look for interesting colleagues, even if they do seem to be saying something a bit different that isn’t what he’s spent his life doing. As long as the new guy doesn’t call him an idiot in public and doesn’t look like he would be a difficult colleague. By the time we get middle-aged and tenured, most of us don’t care much anyway. All the stuff we learned in grad school is now seen as hopelessly old-fashioned or hopelessly wrong. All our theory-specific human capital is next to worthless anyway. So who cares, as long as the new guy isn’t an asshole.”

              Now – that’s pragmatic!

              :)

            • BTW:

              QE does for the most part increase broad money supply, other things equal

              Would you describe that as endogenous or exogenous?

              • Not a completely specious response: that’s only because “the broad money supply” is defined as not including treasuries or GSE bonds. cf. Divisia M4+, which does include treasuries.

                • yeah

                  I was thinking of broad money supply directionally – as in bank deposit liabilities versus banks reserves – not extending it to include everything but the kitchen sink and Mosler’s business cards

              • That’s not exogenous though is it? The non-bank seller of the t-bond willingly sold the t-bond in favor of a bank deposit. The $ supply in broad money is still determined endogenously by demand for deposits even if the central bank is reducing the quantity of t-bonds. And if one is to argue that this actually matters then that person should clearly point out that the increase in deposits is offset by a reduction in T-bonds and therefore matters less than some presume….Even if we did consider this to be exogenous deposit expansion we should acknowledge the exogenous T-bond contraction….so we should be very specific about how QE changes the supply of assets.

                Personally, I think Sumner misunderstands his own ideas. He has said in the past that the Fed should buy anything and everything. But buying cars from individuals is way different than buying t-bonds. One is a swap of closely related financial assets while the other is a fiscal operation implemented by the Fed directly expanding the money supply in exchange for real goods. But this is fiscal policy in essence and I don’t think anyone would dispute whether it might be inflationary. T-bonds, however, are money-like so I think it’s less clear whether it’s accurate to say QE via non-banks is a form of exogenous money supply expansion….

                Either way, it’s a silly debate because the operational details are what really matter and the neoclassicals are all still hazy on the details.

              • Endogenous. The Fed doesn’t control it. Sellers of the bonds can reflux the money. There is also a portfolio balance happening. The Fed just announces a rate of purchase of securities such as $85bn per month and is not setting the money supply.

        • No it is definitely not the same old though he does use those terms. See his latest suggesting that there are different natural rates for capital and for consumption. Some serious unpacking to do in order to understand that.

    • Yeah I first saw Lambert in an SRW list of links, but he seems to be trying to get people to take a look at his stuff. I wish they would. I find it darned compelling, but distrust my judgment.

      • can you summarize the takeaway?

        • Art Patten says:

          “can you summarize the takeaway?”

          Leans very heavily on labor share of income. Beyond that, it’s over my head.

          • maybe some sort of “equilibrium” between labor income and capital income?

            • @JKH: “maybe some sort of “equilibrium” between labor income and capital income?”

              Yes, but I think I’d use the word “attractors” (might be better for econ in general…)

              When the system is in a given state, it naturally moves toward (is “attracted to”) a different state due to inexorable market forces.

              His basic notion is that effective demand is a function of capacity utilization and labor share of income, and that effective demand can (but doesn’t always?) exert a cap on growth. But there are a lot of moving pieces that I can’t encapsulate.

        • Sorry, I’m afraid I can’t without probably getting it wrong. I asked NR for his thoughts by email, and he said he’d he’d seen it, would need an hour to sort through it, plus writing a post, decided not to. It would take me far longer than that, and I just have a lot of difficulty thinking in that “mode.”

          BTW, I much appreciate Nick’s mini-post, think it nicely encapsulates a core view that I’ve been trying to wrap my brain around:

          Planned spending – expected income = change in bank money = change in bank debt

          Or in terms that make sense to me and suggest similar to Keen:

          Real-sector borrowing = planned spending – expected income

          They borrow because they want to spend more than their incomes.

          So: if they want to spend more than income, and intend to borrow to that end, isn’t that borrowing part of their (planned) demand? It may not “cause” that demand, but it certainly enables it…

          ??

          • This is basically defining:

            Aggregate demand = planned spending. Which is fairly standard, right?

            If that planned spending is based on expectations (or current actualities) of borrowing, then borrowed money is part of aggregate demand.

            • not necessarily standard, from my reading earlier

              planned spending = planned aggregate demand

              The idea from earlier that there is such a thing as actual aggregate demand (different from planned aggregate demand) is consistent with Nick’s repeated theme of highlighting the equivalence of quantity supplied and quantity demanded – as distinct from planned demand, which isn’t necessarily realized

              in that quote, IMO he’s saying people borrow early in the time period with intentions of realizing planned demand, but that doesn’t necessarily materialize

              something similar in GL I think

    • Art Patten says:

      “The Effective Demand [Edward Lambert] guy seems to appear everywhere”

      Anyone formed an opinion on his work yet?

    • This must be a spoof, surely.

  52. Absolute Gem of the day:

    http://noahpinionblog.blogspot.co.uk/2013/06/should-japan-default_12.html

    “So there might be some very good reasons for Japan to choose a sovereign default. But of course there would also be large costs. What would those costs be? I see three big ones: Human cost, inequality, and political risk.

    The human cost could be a jump in the already sky-high suicide rate. A large number of Japanese suicides are men who lose their jobs. The close family structure of companies means that these men essentially lose access to their entire social support network. Combine this with a culture that is not very forgiving of failure, and you begin to see why a spike in unemployment might cause a large number of self-inflicted deaths.

    Then again, this cost is not certain. A recovery of dynamism in Japan’s economy might ultimately save more lives than it took. And human psychology is a fickle thing; it might be that in the wake of a default, unemployment might be seen as a natural disaster rather than an individual failure, and the suicide rate might even fall.”

  53. @Cullen:

    “The non-bank seller of the t-bond willingly sold the t-bond in favor of a bank deposit.”

    And:

    “the seller is demanding a deposit over a t-bond….”

    But only because the Fed is offering an attractive price. “Demand” is at a given price.The Fed’s buy orders have to be at the top of the dealer’s/market-maker’s price-sorted buy-order trade book or it will never execute the buy orders; others will beat its offer.

    In that way, the Fed “forces” the banks in aggregate to accept more reserves in exchange for bonds. If a bank turns down a Fed offer that’s a one or three basis points better than “the market,” they lose out to their competitor who does take the offer. So they all have to take the offer.

    So the fed controls, decides on the quantity of, the reserves-for-bonds trade, in aggregate. New Treasury/GSE issues and retirements also affect the net flow of bonds into the market (and the stock over time), but the Fed’s effect on the flow is controlled by Fed policy. The banks just take the Fed’s offers if it’s selling above, or buying below, market.

    And @Ramanan: “Endogenous. The Fed doesn’t control it. Sellers of the bonds can reflux the money. There is also a portfolio balance happening. The Fed just announces a rate of purchase of securities such as $85bn per month and is not setting the money supply.”

    At least under these conditions:

    o IOR
    o ZLB
    o Untrusting credit markets

    The Fed does control the the reserves/bonds trade, hence the total “monetary base.”

    o Customers/banks control the currency/coin proportion of the base, but that’s immaterial.

    o Per the above, the Fed controls the level of reserve balances, by buying/selling bonds above/below market price.

    • Add to the list of conditions:

      o Big Fed balance sheet/high excess reserves

      I’d love somebody else to suss out whether any of those conditions could be removed, and the truism I assert (Fed controls reserve balances) is still true.

    • Cullen Roche says:

      I don’t think anyone in PKE would dispute that the Fed determines the level of reserves. The question is whether the central bank determines the supply of broad money. Cash and coins are determined endogenously by the demand for them by bank customers. Deposits are the same via lending. And I would argue that QE doesn’t alter that. QE with a non-bank starts with a t-bond seller who transacts with a bank and willingly demands a deposit in exchange for the T-bond. So, the point of sale is not a forced transaction by the Fed, but a bank whose book is probably well hedged en route to enacting QE for the Fed. The reserve increase and sale to the Fed is just QE as if the non-bank wasn’t involved in the first place.

      I’d still call the non-bank transaction an endogenous increase in bank liabilities, but I guess there’s some gray area here.

      • @Cullen: “The question is whether the central bank determines the supply of broad money.”

        @Ramanan: “Wasn’t talking of the monetary base!”

        Right. I went somewhat off that rail.

        But when the Fed buys bonds from nonbanks it is certainlyaffecting the broad money supply (increasing bank deposits, [hence banks' balance sheets, {hence Fed reserves}]).

        But the real point: is that increase in the broad money supply significant or material? Back to my previous semi-specious comment: QE only increases the broad money supply because the broad money supply is defined as not including treasuries. So the Fed swaps defined “money” for “non-money.”

        Does the real/nonbank sector have more “money” post-QE? Does it cause any wealth effect beyond the one we know about: driving up bond (hence equity) prices due to greater flow demand? (Much less any monetarist-style hot potato?)

        You could argue a la the FT Alphaville crowed that because reserves can’t be collateral-chained like bonds can (right?), there’s less “bank money” post-QE, compared to a non-QE counterfactual.

        • I am not sure of this FT Alphaville thing about collateral collateral collateral.

          A lot of vague analysis there. It is true that the repo markets observed the financial crisis more clearly than others … as if they were in the front seat of a concert … I really don’t know what FT Alphaville’s story is for recent times.

          QE does increase the broad money supply when the Fed buys the bonds. In a sense – as JKH mentioned – the counterfactual is no QE and you dismiss arguments about observed data because in the absence of it, broad money may have reduced.

          Hot potato process is to be distinguished from wealth effects of QE. The Wealth effect happens via capital gains which adds to the stock of wealth of households but the Monetarist hot potato process is supposed to happen at the flow level but in reality it doesn’t happen. Monetarists confuse income and money.

          • “you dismiss arguments” … sorry meant that the observed increase in the money stock measures such as M1, M2 seems to suggest that the Fed’s asset purchase has no effect but balance sheet arithmetic suggests that. So you have to compare it with the counterfactual. More generally there are other mechanisms such as agents reducing their indebtedness to banks when the sell the bonds to the Fed and portfolio rebalance and things such as that.

            • @Ramanan: “More generally there are other mechanisms such as agents reducing their indebtedness to banks when the sell the bonds to the Fed and portfolio rebalance and things such as that.”

              Right. Fed obviously doesn’t “control” the broad money supply, but it certainly affects it. But I think that affect is 1. pretty small and 2. pretty immaterial (at least in the current situation?).

              • I think it is difficult to prove one way or the other but IMO, the effect on equity prices is quite strong.

                • Right. Of course the effect of (temporarily?) boosted equity prices on investment and overall spending/NGDP is much less certain. See my yesterday wealth effect post.

                  • Yes agree.

                    Even I had written about this sometime back http://www.concertedaction.com/2013/05/15/central-bank-asset-purchases-and-its-connection-to-tobins-theory-of-asset-allocation/

                    Okay some general opinion without proof: I think the wealth effect is higher than most people think. I however think wealth effect due to capital gains may not have an immediate effect because people think the capital gains may not last. So only when they become confident that the gain is sort of permanent and there is less chance the asset price will fall, will they increase consumption.

                    • Seems like this would be measurable, which is something that real social/behavioral scientists would do. But you know: empirics are so…messy compared to theory… I’m guessing some have tried to measure this, I just don’t know of it and the mainstream ignores it?

                    • I think they have tried to do some econometric tests and observe something like 0.04 which is actually low.

                      I don’t know how they do it but my suspicion is that it is wrong. It is because in a model like G&L, in the equilibrium ultimately reached in *simple* models, propensity to save = 0. But propensity to consume is not 1 and there can be various combinations of propensities to consume out of income and wealth in which there is no saving in such a state.

                      Also, higher wealth effect leads to lower public debt. In the extreme with low growth and zero wealth effect, the public debt should be very high which is not the case. Third, fiscal policy works very fast and the time-lag has to do with the wealth effect parameter.

                      Something like that – a bit of hand-waving between pure theory and data.

                  • A further complexity with the wealth effect, at least in the UK.

                    Households hold very few government bonds directly, and a only smallish share of public equity. The majority of private domestic ownership is in pension funds.

                    A wealth effect can still occur from an increase in pension fund values, but it’s likely to be much weaker. There’s a real effect in that higher fund values translate into higher pensions, but that obviously takes a long, long time to come through.

                    The effect is also complicated by the fact that gilt yields are used to calculate annuity rates. So although asset values may go up, falling gilt yields depresses quoted pension rates. A fall in people’s expected pensions hardly leads to increased spending.

                    The most likely source of a wealth effect in the UK is through the value of housing and, in my opinion, this is small but significant (notwithstanding your pie chart, Steve). However, the link from QE to house prices is an indirect one, given who holds what (funds holds gilts, but not housing; households hold housing, but not gilts).

    • “The Fed does control the the reserves/bonds trade, hence the total “monetary base.”

      Wasn’t talking of the monetary base! The Fed doesn’t control monetary aggregates such as M1, M2 etc.

      • Think of a model. Is the Money supply of the model determined by the Fed? Is it an input to the model? No. That is the result of the Fed’s actions and the private sector.

  54. Cullen Roche
    I don’t think anyone in PKE would dispute that the Fed determines the level of reserves.

    Is that right?

    Ch 10 of G-L on inside and outside money has the level of reserves determined endogenously, which fits with my understanding of how it works in the UK. It’s sort of like the money multiplier working in reverse. Is that not how it works with the Fed?

    Or did you just mean specifically in the current QE environment?

  55. Need for clarity on the criterion for money endogeneity:

    The criterion that the Fed does NOT control money supply (in the sense of broad money or bank deposit liabilities) via its control over reserves (i.e. via the so-called “money multiplier”) is quite different (it seems to me) than the criterion of who initiates a transaction in order to obtain money of any form.

    I see a mixture of these two arguments in discussion (except perhaps from Ramanan), and I’m not sure both are right.

    I think the first criterion is the one that applies to core discussions about endogenous money, as it is usually thought of in the context of broad money effects (i.e. bank deposit liabilities). I.e. the central bank does not directly control “the money supply” by its control over reserves.

    The second criterion above seems moot, in that every transaction requires a bid and an offer with two counterparties sawing off on the question of who takes “the initiative” (except in the refined way of who “hits the bid” or who “lifts the offer”). This applies to all of commercial bank loans, Fed OMO, and Fed QE, it seems to me. And it applies to both reserve effects and broad money effects in all cases (OK, Steve, broad money in the sense of bank deposit liabilities as the core idea of broad money). All transactions require two counterparties.

    What about reserve creation? Again, it is a case of the central bank wanting to control the level of interest rates. And once again, that has little to do with who “initiates” the transaction. The core point it seems to me is that the level of central bank reserves is determined by its desired control over interest rates.

    So it’s “the” interest rate (i.e. the rate that the CB wants to control) that becomes exogenous.

    Does that make central bank reserves themselves exogenous or endogenous?

    Does that question even make sense in a definitional framework that is balanced with respect to price and quantity effects criteria?

    And in the case of QE, the CB is extending the idea of interest rate control to the bond market.

    But it’s NOT controlling the bond rate – because it’s not establishing an unlimited two-sided market at the rate it “desires”.

    Is this all defined clearly enough?

    What’s the right definition of the criterion, as it applies to broad money and reserves, and as it applies to the policy interest rate versus QE transaction rates?

    Messy … just thinkin’ out loud here for now.

    Ramanan? Others?

    • Personally, I think that exogenous or endogenous is just a feature of models. In the real world, very little is truly exogenous. It’s easy to see something like a specific rate of tax as exogenous, for example, but in reality what happens in the economy has a lot of influence on what the government decides to do with that rate in the future.

      So, if we are asking about what is truly exogenous or endogenous, it’s more a question of which is a better model of how it works. This in turn is a question of what relationships do we think are strong and stable and which do we think are not. We may then choose to treat certain variables as exogenous, but need to remember that it’s always an approximation.

      So it’s not a question of whether money is exogenous or endogenous. It’s about whether we think a model with exogenous money determining money GDP reflects a strong and stable relationship, or whether we think a more reliable model is going to be one where other factors determine both.

      Just my thoughts.

      • Excellent point.

        Reminds me of Nicholas Kaldor’s point:

        “The only truly exogenous factor is whatever exists at a given moment of time, as a heritage of the past.” – 1985.

      • Sort of getting at something like that … exo versus endo is a bit of a mug’s game in terms of definitions, context and starting assumptions for the marginal transaction, unless one is quite specific about those

        Also, for example, as above, a choice between the interest rate and the quantity of reserves/money makes the criterion of who “initiates” the money transaction somewhat moot if the choice is the interest rate

        Specifying how everything works trumps constructing definitions

        • Totally agree with you that it is moot.

          But the initiation part may have a history. So consider a purchase of foreign exchange by the central bank. It creates settlement balances in the accounts of member banks. In neoclassical literature, the discussion is then presented as two things: non-sterilized and sterilized – as if the central bank has a choice in initiating. So it is as if there are only two possibilities: central bank initiation or central bank doing nothing. But banks themselves may get rid of the excess reserves by contacting the central bank and the central bank may sell T-bills.

          I however don’t know the context in which initiation is used in QE. The Fed announces dates and quantities in advance and does a reverse auction. So I don’t know what the initiation from the other side is about.

          • Exercising brain cells here in preparation for revisting GL – which has very interesting things to say about endogeneity – such as you point out on sterilization illusion.

            QE reverse auction? Don’t follow it that closely, but its an extension of OMO in terms of actual execution isn’t it? I.e. notwithstanding a pre-announced program, the Fed picks its spots and bids its price? Anyway, the choice to swap bonds for money is the seller’s – at a price – as is the choice to swap money for bonds by the Fed. The end-seller may not even be certain his bonds are the ones that are going into the Fed.

          • And putting this all back into context, SK has some things to learn about QE and the creation of deposits through flow of funds – before asserting differential equations that are stock/flow inconsistent.

      • Not to say it’s wrong to say something like “the policy rate is exogenous”; it’s just we mean it’s better to think of it that way.

      • No – not wrong at all.

        In fact, that’s clean origin of it.

        Just that following that trunk split – rate versus reserve balances – the rest of the tree gets a little tangled when attempting to clarify how the exo/endo distinction manifests itself.

    • 1. I don’t think it’s about who initiates a transaction, but about the Fed being willing to outbid the market. It’s not driven by profit motive. Banks are. So it can “force” the banking system to buy or sell.

      2. Been wondering for a while: when we talk about exo/endo, is it simply different (and arguably obscure) language for arguing about causes and effects?

  56. Ramanan,

    Check out brilliant insight:

    http://www.economonitor.com/lrwray/2013/06/12/warren-mosler-car-guy/

    Guess WM never thought about the Fed paying interest on reserves

    Or, more to the point, that paying a zero interest rate on reserves is a actually a choice, rather than a force of nature.

    Exogenous even.

    :)

    • Art Patten says:

      “Guess WM never thought about the Fed paying interest on reserves
      Or, more to the point, that paying a zero interest rate on reserves is a actually a choice, rather than a force of nature.”

      You’re kidding, right?

      • His “natural rate is zero” argument is logically premised on there being no alternative to paying zero interest on reserves. That’s obviously false – not only as evident in what actually happened in QE, but also as evident in various monetarist proposals for negative interest rates on reserves. Zero is just another rate, between positive and negative. How is that kidding?

        • I took Warren Mosler to mean that the risk free interest rate represents an artificial scarcity that central banks choose to impose rather than being some manifestation of the innate value of deferring consumption that stems from the innate human psyche above and beyond any institutional influence (is that what “micro foundations” are supposed to be?).

        • natural rate of zero is premised on goosing the system with reserves, not paying interest on them, and letting rates go to zero

          again it misses the logic that not paying interest is a choice to pay an interest rate of zero

          and its based totally on the institutional feature of bank reserves in the first place, which is a function of the architecture for a competitive banking system

          if the system consisted of one nationalized bank, there’d be no bank reserves, and the state would be faced with the choice of what rate to pay on bank deposits – as a function of monetary policy

        • Art Patten says:

          Mosler/Forstater: “If the central bank has a positive target for the overnight lending rate, either the central bank must pay interest on reserves or otherwise provide an interest-bearing alternative to non-interest-bearing reserve accounts…Our main point is, in nations that include the USA, Japan, and others where interest is not paid on central bank reserves, the “penalty” for deficit spending and not issuing securities is not (apart from various self-imposed constraints) “bounced” government checks but a zero percent interbank rate”

          WTF is wrong with you guys? MR = Mosler Rejection (at all costs)?

      • “You’re kidding, right?”

        Mosler’s contention – and in which he thinks is what sets his Mosler Monetary Theory apart from anything else – was that the government is in a supreme position that it has an open line of credit at the central bank – as if there is no law preventing it – and the only reason it issues bonds is to prevent interest rates from falling to zero. 2008 arrives and it is clear that it was not right. The Treasury continues to issue bonds in spite of the fact that with the Fed paying interest on reserves and with a floor system short term rates can fall to the floor. If what he said was true that the “purpose”, or the sole purpose was to prevent interest rates from falling to zero, then the US Treasury could simply have stopped issuing bonds as soon as the Fed started paying interest on reserves.

        It is true that the government is in a supreme position in the money creation process but Moslerism is full of overkills to drive this point.

        • Art Patten says:

          That’s a little better, but I stand by my BS call @ JKH. That was utter nonsense. (And your paper shredder comment was a low blow.)

          • Cullen Roche says:

            Art,

            No one is out to get MMTers. They’re not the victims of a conspiracy theory by MRists. They’re the victims of their own sloppiness at times. Mosler’s shredder analogy is a perfect example of such sloppiness. In trying to promote the idea that “taxes destroy money” Mosler loves to use the shredder example. Except that the only problem is this never really happens in any real sense. Taxes don’t “destroy money”. Taxes are a redistribution of bank inside money. There is a very clear flow of funds that occurs in the system from the time of deposit creation to the time of taxation to the time of govt spending. There are not a series of starts and stops here where banks create money, then the govt destroys the money and then creates it again when they spend. That’s all just MMT creating their own story for how things work. And the shredder comment is a dangerous oversimplification that has actually helped convince a good deal of people that the idea is right. It’s not.

            MMTers aren’t the victims of a “gotcha” campaign by MR. They’re the victims of years of sloppy work and a theory that doesn’t exactly apply to the way the US monetary system works. Other PKers point it out not because we’re out to get them, but because this type of stuff is actually hurting the progress made by other PKers in helping the public understand PK positions. That’s all. It’s not personal and never has been.

            • Art Patten says:

              Cullen, that’s fine insofar as it goes, but by the same token, JKH shold be properly seen as the victim of his own sloppiness re his assertion about Mosler and IOR. And again, Ramanan’s shredder comment was a lower blow than Mosler deserves. Problems with the shredder argument? Fine. Ripping him for a pretty innocuous video about his interest in cars? Come on, man.

              • Art Patten says:

                And please, don’t attribute any conspiracy nonsense to me. All I’m doing here is suggesting that people put their ugly sides under wraps. We all know firsthand how ‘well’ that plays outside an inner circle…don’t we?

                • Cullen Roche says:

                  I agree. Personal sniping should be left out of this. We didn’t start MR as a smear campaign against Mosler. In fact, I talk to Warren almost weekly and we’re more than cordial. He knows what we’re doing and he doesn’t take it personally. He disagrees with our position, but he doesn’t take anything personally. And I think that’s the way we should all view this. We’re all just debating ideas about how things work. Getting personal and calling people names just distracts from what we’re actually trying to achieve. So yeah, let’s put the ugly side under wraps. 100% agree.

        • I guess that Warren Mosler is taking the default state as being the system being “goosed” with excess reserves and the alternative state of bonds being issued and reserves being drained as being the active choice. To me that makes sense since bonds are an extra piece in the jigsaw so including them seems more active than not including them. Is paying interest on reserves more than just a payment to the banks? Is it much different from say providing all bankers with free rail travel?

    • He should also try a new business of making shredders and post a video on that. How he got interested in it and things like that.

      The worst evidence about his theory of taxes paid in cash being immediately shredded by the tax man is him having met a person who shreds cash!!!

      • :)

      • Art Patten says:

        Unbelievable. The guy, who doesn’t have a mean bone in his body as far as I can tell, realizes 20 years later that PR matters far more than he ever imagined. At the end of it all, an industry outlet interviews him about an innovative but ultimately failed venture. And that makes him what, a media whore? Is this still about Randy calling Cullen a retard, or what?

        F&#%ing ridiculous. All of it.

    • “Guess WM never thought about the Fed paying interest on reserves”

      Right. cf. My comment re: Soft Money Economics. Explanations about OMOs controlling interest rates via the interbank lending market are of purely historical interest. Doesn’t work that way any more, and won’t as long as banks hold (significant) excess reserves. It’s all IOR floor.

  57. somebody should read what they quoted:

    “where interest is not paid on central bank reserves”

    as in – didn’t think about it at the time – and what actually happens when it becomes a live issue

    its not about mean bones

    its about the thesis presented

    no need to get so emotional over facts

    • And no problem if Wray wants to advertise Mosler’s financial genius together with showing car videos. I didn’t watch the video myself, but he seems to make great cars.

      Apart from that, there are a few people around who have understood for some time that when the central bank supplies reserves earning zero interest in excess of what is demanded, that the market rate heads in the direction of zero. That’s been the working premise of standard OMO for decades.

      So as an extension of that, the entire premise of a zero natural rate becomes pretty questionable, as described in more detail above.

      And this aspect is connected very much to the topic here of the interest rate being exogenous. So that’s the point. Just the facts. I’ll leave to others to assess whether or not that’s sloppy.

      And given that I said nothing derogatory about WM at all, you’re WAY too sensitive about all this, and too eager to victimize a great financial mind and car builder.

      • Art Patten says:

        You should have been a lawyer. Maybe you and Carlos can switch careers? ;)

        • Not a bad idea.

          I think Carlos may be quite partial to zero interest rates as well – especially those earned by banks.

          :)

  58. Am I right in being under the impression that the fed (and bank of England and bank of Japan) does not now have the capacity to raise short term rates to say 6% for a sustained period. Doing that would entail paying out a massive amount of interest on reserves. Where would they get the money from to pay for that? If they sold of assets such as long term bonds to try and raise the money/drain reserves; the price of those long term bonds would fall so far that even if they were all sold there would still be a glut of reserves.
    If so, then I guess the central banks have lost the capacity to cause a classic “tight money recession” to keep wages down. Now it is down to speculators to cause a commodity price spike recession whenever the economy looks like picking up.

    • Related to that, would that high of an overnight interest rate cause entities to start redeeming currency for demand deposits?

  59. I thought Keen said income plus change in debt = aggregate demand.

    Could Keen be trying to say something along the lines of Y “minus” financial assets including the MOA/MOE = C + Igoods ?

  60. Steve Roth says:

    @JKH: watching my daughter graduate from u chicago. You will be happy to know that the commencement speaker is…an accountant!

    • A good day for you and your daughter!

      But please remember that accounting and economics are both too important to be defaulted into the hands of their respective professions.

    • Congratulations Steve, that’s a huge accomplishment for her – and for you!