Treasury and the Federal Reserve – Platinum/QE Integration

This is mostly for the potential benefit of Treasury and Federal Reserve operatives, in case they’re now seriously considering the operational implications of the platinum coin idea. I say this with tongue slightly in cheek, because both organizations obviously have highly qualified and brilliant experts in this area. This is just food for thought from an outside observer.

This platinum coin strategy has obviously been the subject of much recent public comment and debate. The platinum strategy more broadly is not the primary question for consideration in this post. The topic here is the potential for the specific operational integration of a platinum strategy with overall Federal Reserve balance sheet management.

Congress has the responsibility for approving spending and tax measures that Treasury implements through cash and funding management. Since spending and tax policy, once implemented, determine the budget deficit, Congress in effect approves the deficit that Treasury ends up incurring – before it incurs it. This interpretation naturally allows for the fact that there is an element of endogenous determination of that deficit due to uncertainties in overall performance of economy. But that doesn’t change the fact that Congress sets the policy dials for spending and taxation that together with economic performance end up determining the budget deficit.

The normal case is that Treasury must fund the resulting deficit with bills and bonds. And so, given Congressional authorization of spending and tax policy – and by implication the resulting deficit – Treasury depends on normally prescribed funding mechanisms using bill and bond issuance in order to implement policy instructions from Congress.

If Congress throws up a roadblock by impeding that normal funding process, it is reneging on what it has already authorized.

And that is the current situation regarding the debt ceiling. Treasury’s hands are tied in its responsibility to implement fiscal policy that has already been approved by Congress. If Congress refuses to increase the debt ceiling, Treasury will at some point be unable to fund the deficit in the way it normally does.

The possible scenario of a platinum coin solution to this operational impasse is generally understood by now. If Treasury deposits a platinum coin in its account at the Fed, the Fed will credit Treasury with funds and hold the coin as an asset. Treasury will use its funds according to what spending has already been approved by Congress, when taxes are insufficient to fund the required spending.

This operational adjustment can blend seamlessly with the Fed’s ongoing process of balance sheet management in the current economic environment. The coin is a relatively minor asset management tweak in this context. The Fed has an army of technical experts who can handle this adjustment within the constraint of overall policy consistency. That desired goal of policy consistency can be framed in the following way:

The essence of the quantitative easing concept (QE) is that it partially replaces the normal funding of the cumulative budget deficit – bills and bonds issued by Treasury – with bank reserves issued by the Fed. (The Fed also typically acquires bonds when it issues banknotes, but most of the ultimate funding for the cumulative budget deficit normally remains in the form of bills and bonds held by the public.) Currently, the Fed is operating a QE policy based on market purchases of Treasury and agency bonds in exchange for newly created bank reserves. At the same time, the Fed has been transparent in its intention to exit this strategy when the economy and the financial system gain firmer footing. The Fed will eventually eliminate the excess reserves it has created through asset sales and maturities. And it will be able to refine further the exit pace and the degree of associated monetary tightening by increasing the federal funds target rate and the interest rate paid on reserves. And it will also be able to tighten the immediate availability of bank reserves, if so desired, by auctioning off reserves as term deposit liabilities issued to the banking system. All of these elements point to a well thought out Fed contingency plan for the QE exit program. At present though, QE policy remains in full implementation mode.

The platinum proposal can be integrated with this plan in both the active QE implementation stage and the phase out or exit stage. If Treasury deposits a platinum coin with the Fed, the result will be a platinum asset held by the Fed and a deposit liability issued by the Fed to Treasury. Treasury will spend those newly created funds, and in that process Treasury balances will become bank reserve balances, still on the Fed’s books. The eventual result will be a platinum asset that is then funded by bank reserves. This is exactly the same funding result as occurs with regular bond QE.

In the context of its overall program for the QE related expansion of bank reserves, the Fed will now have two broad asset categories – Treasury and agency bonds; and platinum – funded by bank reserves created by each of bond QE and “platinum QE”. That augmented asset portfolio of bonds and platinum can be managed as part of a more broadly integrated QE strategy. The Fed strategy then consists of the optimal path over time for the size and composition of an augmented QE asset portfolio (bonds and platinum) held by the Fed, funded by bank reserves issued by the Fed.

In this regard, it should be recognized that the platinum component of this augmented asset portfolio is in effect a virtual QE bond portfolio. This is because platinum easing has replaced counterfactual (and normal) bond issuance – since the reason for the platinum asset on Fed books is that the Congress has not allowed bonds to be issued in the first place. In other words, the effect of platinum easing is very close to that in which bonds were issued instead and then purchased back in normal QE mode. Both techniques correspond to fewer bonds held by the public, compared to the relevant counterfactual without either bond QE or platinum QE. For this reason, the two modes of QE are directly comparable – not only in terms of the reserve funding effect, but in terms of the implication of the Fed’s asset portfolio for bonds outstanding and held by the public. Thus, the augmented QE program corresponds to a draining of bonds from the public sphere in two ways – purchasing bonds already issued (regular QE) and replacing bonds that should have been issued but that weren’t issued due to Congressional holdback of bond funding authority (platinum QE).

In both tranches of the augmented QE program, the cumulative budget deficit is being partially funded with reserves instead of bonds. In the case of regular QE it is technically a previously incurred deficit within the cumulative deficit that is being partially funded (re-financed) with reserves. In the case of platinum QE, it happens to be the current deficit that is being funded, which is effectively the front end of the cumulative deficit. Thus, in the case of platinum, the current deficit rolls out as Treasury spends its platinum sourced cash balances, creating new excess bank reserves as a result.

Finally, perhaps the most integrating characteristic of this augmented QE approach is that both regular QE and platinum QE tranches are intended to be temporary, and their prospective amortizations are captured within specified exit contingency plans. Regular QE begins to wind down and then reverse when the Fed decides enough is enough. Platinum QE begins to wind down and then reverse when Congress finally acknowledges its responsibility to permit the normal bond financing of spending that it has already approved. At that point, in both cases, the Fed and Treasury will work in concert to allow the Fed balance sheet to shrink and to replace bonds and platinum held by the Fed with bonds held by the public. In the case of regular QE, the Fed will either sell bonds to the public, or it will mature existing bonds that Treasury will refinance with the public. In the case of platinum, the Fed will return the platinum asset to Treasury and Treasury will fund the effective repurchase of that asset from the Fed by borrowing in the market. In both cases, there is an effective category conversion of Fed held bond and platinum assets to publicly held bond assets, with an accompanying withdrawal of excess reserves from the banking system.

The Fed can manage all of this on a fully integrated basis. As described, the augmented portfolio consisting of bonds and platinum is in effect a portfolio of actual and virtual bonds – because the same amount of bonds has been withdrawn from counterfactual public holdings – either by Fed purchase or by Congressional hold back. So the augmented portfolio is – in full effect – an actual/virtual bond portfolio. Thus, in this platinum easing context, platinum held by the Fed is equivalent to a bond in terms of its monetary policy effect. It is an effective operational monetary substitute for QE bonds – in the context of the problem now faced by Treasury in funding the budget deficit.

And because of this, the strategic management of the Fed’s balance sheet can become a seamless exercise in which platinum is simply a minor operational adjustment to current QE bond portfolio management by the Fed.  If the Fed begins to accept Treasury/Mint platinum on deposit, it is in effect accomplishing the equivalent of quantitative easing by buying bonds. And therefore, it can at its own option offset such platinum easing by adjusting downward the residual quantitative easing that it might have completed otherwise in the absence of the platinum modified strategy. And this sort of seamless strategic effect extends to the exit strategy. In the unlikely event that platinum easing proceeds long enough to interfere potentially with what might have been a more straightforward exit from the regular QE strategy, the Fed can simply step up the pace of exit on the bond tranche of the augmented portfolio. And again, the Fed has a rich array of tools in order to implement this exit strategy more comprehensively, including increasing the Fed funds rate and the interest rate paid on reserves. Moreover, a powerful tool that is sometimes overlooked is that at any point the Fed can proceed to auction off reserve balances as fixed deposits held by the banking system. This would have the intriguing effect of providing duration hedging product for banks in their own liability portfolio management process, and would also provide solace to some of those outside of the banking industry who (erroneously) believe that banks lend reserves when they make loans to non-bank customers.

Comments
  • Bill January 9, 2013 at 7:45 am

    Question – by minting the coin and thereby “printing money”, wouldn’t this result in a devaluing of the currency? Just wondering what the actual impacts of a lower exchange rate for the dollar would have (ie on energy and other imports, thereby affecting the economy).

    • JKH January 9, 2013 at 9:19 am

      That’s a big question which is much larger than the effect of platinum in this context.

      You can think of regular QE as the Fed targeting the size and composition of an asset portfolio that it is funding with reserves. The platinum piece just changes the composition of that portfolio from bonds to bonds plus platinum. The size and the funding can be managed in the same way it is with regular QE.

      The issue you raise relates directly to the size of the funding portfolio. That’s the part that raises the question about “devaluing” the currency. And that’s the part where there’s considerable debate obviously about the best way to think about that risk. And it’s really inflation risk that’s the core of that issue – whether experienced through the domestic price level or through the foreign exchange market or both. That’s a big, separate discussion.

      But platinum is only a marginal piece of that discussion in this context, as per the post. And the core of that discussion is where does inflation come from? And even if were the case that inflation could be linked to the size of a QE reserve funding portfolio – which is highly debatable – the Fed has all sorts of tools to manage the exit from augmented QE in order to control that risk – even if the risk has more to do with perception than reality.

  • Joe Franzone January 9, 2013 at 8:20 am

    “Moreover, a powerful tool that is sometimes overlooked is that at any point the Fed can proceed to auction off reserve balances as fixed deposits held by the banking system. This would have the intriguing effect of providing duration hedging product for banks in their own liability portfolio management process,”

    This is interesting. Is there any possibility we could see a post that elaborates on this tool and its effects? Thanks.

    • JKH January 9, 2013 at 9:33 am

      That’s also a big, separate discussion.

      In general terms, the presence of a $ 1.5 trillion reserve portfolio has a substantial effect on the interest rate sensitivity position of the banking system. Risk managers in banks have to judge the effect of that interest rate position on their books. The nature of the interest rate sensitivity of a reserve portfolio from the perspective of a commercial bank risk manager is that the interest rate (currently 25 basis points) can be expected to be dormant for as long as the Fed takes before beginning to tighten monetary policy. When it starts to tighten, it will start increasing IOR. And you never know about these things – it may be the case that after a long time at the zero bound, the Fed may tighten in a series of steps which have a non-trivial effect on IOR. Even moving IOR from 25 basis points to 1.25 per cent for example would have a material effect on the interest rate risk positions of commercial banks.

      In that sort of scenario, you could expect short dated yield curves of all types to steepen. If the Fed at the same time starts auctioning off reserve balances in the form of term deposits, in may provide an opportunity for bank risk managers to lock in fixed term yields on reserves and offer fixed term deposits to customers (customers might switch out of demand deposits). Or, risk managers might hedge short dated liabilities that are already on their books at lower rates and that have been left as open positions.

      That’s just one example. The general point is that the Fed can transmit yield curve effects through the banking system if it wants to by terming out reserves.

      And that may provide at least the comfort of perception to those who become increasingly concerned about the level of excess reserves in the banking system – even if those concerns of perception are based on rather wobbly notions about the sources of inflation risk.

      • Sergei January 13, 2013 at 10:02 am

        “When it starts to tighten, it will start increasing IOR” and so on

        I think that there is more much here and life is not that easy. What does it mean to tighten? Normally it means increasing interest rates. Given that assets are longer than liabilities, margins on liabilities get squeezed and ALM tries to passing new margins to assets. If BLs have both assets and liabilities and try to reach their profit targets then they react at well be re-balancing their internal margins. That is the normal course.

        But what does it mean to tighten when the CB pays IOR and there is plenty of excess reserves around? Collectively banks would prefer to stop any lending, create a mother of credit squeezes and wait until they get more reserves for which they get more money from the CB.

        • JKH January 13, 2013 at 10:12 am

          The immediate mechanism:
          - rates go up on all floating rate bank assets – that includes prime rate lending; which is tightening
          - the anchor for the bank based yield curve increases, and its likely much of the curve will move to a higher level as well, reflecting what normally would be an expected tightening cycle that doesn’t stop with the first increase; which is tightening, and that ripples through all fixed interest rate markets and their pricing
          - wrt to reserves, you’re right that on that portion of bank balance sheets, margins may improve – because the liability rate structure will be slower to move off the zero bound; but in total the impact of rate moves on banks is complex and shouldn’t be oversimplifed

      • Sergei January 13, 2013 at 10:06 am

        And IOR is a perfectly floating position. It alone does not create any IRR. Given the volume of MZM in the banking system excess reserves should reduce the model risk. Banks should generally feel happy.

  • wh10 January 9, 2013 at 8:52 am

    JKH,

    “In the case of platinum, the Fed will return the platinum asset to Treasury and Treasury will fund the effective repurchase of that asset from the Fed by borrowing in the market.”

    Who is going to direct the Treasury to repurchase the coin? Congress? The Treasury? The Fed (would that be over-stepping its bounds)? If this is ultimately not within the Fed’s purview, it appears a decision regarding monetary policy needs to be outside of the Fed. For those that cherish Fed “independence,” would this be problematic?

    The coin seems to be this weird fiscal-monetary hybrid, in which its initial purpose is to enable fiscal policy while its exit requires “unconventional” monetary policy. I think this is highlighted by Krugman’s quibble with Joe Weisenthal: “Joe Weisenthal says that the coin debate is the most important fiscal policy debate of our lifetimes; I agree, with two slight quibbles — it’s arguably more of a monetary than a fiscal debate…”

    Also, this is slightly off-topic, but you’ve always acknowledged, in operational detail, the Fed’s intention to exit its QE policy sometime in the future (e.g., your ninth paragraph in this essay). However, from your perspective, do you view this exit as necessary for the Fed to carry out effective monetary policy in the future? It seems like QE has enabled us to transition to an IOR-based monetary policy system. Why not just continue with that?

    • JKH January 9, 2013 at 9:57 am

      It’s a process of operational co-ordination between Treasury and the Fed – the same type of operational joint effort that we saw at the start of the financial crisis when Treasury issued a special tranche of treasury bills in order to sop up excess reserves that the Fed had started to create, before the Fed had the go ahead to pay interest on reserves. There was a short delay in getting that authority.

      Treasury knows what its job is in the normal course. Once Congress stops blocking the issuance of bonds, things can return to normal as far as retiring the platinum piece and returning to bond funding is concerned. And as I said in the post, the Fed can accommodate platinum exit as a subset of the entire QE program.

      “However, from your perspective, do you view this exit as necessary for the Fed to carry out effective monetary policy in the future? It seems like QE has enabled us to transition to an IOR-based monetary policy system. Why not just continue with that?”

      Yes.

      It is necessary in the current institutional configuration – i.e. in the absence of something like a Central Treasury Bank, which would provide a more formal institutional platform for co-ordination of mixed reserve/bond funding on a more permanent basis.

      The idea that QE has “enabled” transition to a permanent IOR system is inadequate on its own, IMO. It twists the defined purpose of bank reserves in the existing institutional framework – which is to provide transaction liquidity for a competitive banking system – not permanent funding where banks become effective funding subsidiaries for government.

      That sort of change shouldn’t be random outcome of a temporary financial crisis. If that’s what people want for a future banking system, somebody needs to think it through more than any proposal I’ve seen floated so far – just IMO, of course.

      (See my post on the Chicago Plan for some general thoughts on the functional role of reserves.)

      • wh10 January 9, 2013 at 10:23 am

        Per the first part – makes sense, in that there is precedent for these types of operations.

        Per the second part – makes hazy sense, at a high level. I’ll have to revisit your papers. Though I would be curious to know if you’ve identified any actual problems or operational roadblocks that could arise if we continued with IOR without further reform. In other words, are there things monetary policy would like to do, in terms of affecting the economy, that it won’t be able to do because of this random transition?

        • JKH January 9, 2013 at 10:47 am

          Well, I think that from an institutional perspective, you would have to have a permanent structure and process for active co-ordination of funding strategy as between bonds and bank reserves.

          At a more fundamental level, it’s a great question.

          So far, I’ve just offered a template (CTRB) for considering this kind of funding question at a unified institutional level. In the Chicago post, I considered that funding question a little more deeply I think, but from an analytic rather than a policy perspective.

          So I don’t have a strong view on structural reform right now, except that I think its interesting to consider the possibility of more permanent Treasury and Fed coordination, and maybe even along the lines of the CTRB.

          There’s one other important analytic perspective that I haven’t really tried to develop yet in sufficient detail, but which I mentioned it toward the tail end of the Chicago post. That has to do with the distinction between liquidity risk and interest risk in all of these questions about funding.

          Interest rate risk needs to be dealt with properly from a system architecture perspective. And that starts with the specific choice of active interest rate management policy by the monetary authority – as opposed to permanently zero interest rates, for example. I don’t think that the latter choice has a hope in hell of progressing in our lifetimes – just IMO, but I don’t see a sufficient thoroughness in the development of that idea. Once you take that out of the equation, other things follow – including the fact that interest rate risk must be considered in the macro institutional design of the system. Recognizing that a currency issuer doesn’t face liquidity risk just isn’t sufficient.

          The other thing is that IOR forces the distribution of cumulative deficit funding through the banking system, and cuts off funding opportunities in the sense of what might be provided directly by non-banks outside of banking system channels. This is a huge issue, which again has not been treated seriously enough in proposals I’ve seen. But of course, the elephant in the room is that such proposals that steer towards IOR funding on a permanent basis are heavily driven by ideology on the role of banks. In any case, I haven’t seen sufficiently detailed proposals that reconstruct the financial system along such lines, ideology or not.

          • wh10 January 9, 2013 at 10:51 am

            Thanks, lots to chew on here.

          • Clonal Antibody January 9, 2013 at 3:57 pm

            Why is a $1 Trillion coin any different than a Trillion $1 coins? It costs the US mint $0.18 to mint a $1 coin. So there is $0.82 profit on every $1 Coin. This $0.82 is Revenue to the US Treasury (same as taxes) So if the US chose to honor its obligations in $1 coins, it is free to do so. The $0.82 would go to fund activities that would not be covered by tax income. Theoretically, you should be able to run budgets 5.5 times tax revenues as long as the cost of production holds – which with recycling the coins should be quite doable.

            What do you think of this idea?

            • JKH January 9, 2013 at 5:41 pm

              First, the pure math of the situation:

              Set aside any taxes.

              If Treasury pays $ 180 billion to produce $ 1 trillion in coins, that increases the deficit by $ 180 billion.

              Treasury then issues $ 1 trillion in coins. That results in a $ 1 trillion flow from bank reserves to the Treasury account at the Fed.

              (Let’s just set aside timing issues, given the massive amount we’re talking about here)

              That $ 1 trillion flow of funds in itself has nothing to do with the deficit.

              But when Treasury spends that $ 1 trillion, the result is a deficit of $ 1 trillion.

              You can then consider the $ 180 billion cost of producing the coins as part of the $ 1 trillion deficit that the coins fund. So the coins fund a net $ 820 billion deficit, in addition to the $ 180 billion deficit that is the cost of producing them.

              An additional financial issue is that of seigniorage:

              The correct financial accounting is that the seigniorage effect of issuing such coins will show up in future years – in the form of a reduced cost of paying zero interest on $ 1 trillion in coins compared to some positive rate of interest on $ 1 trillion in bonds. That is an opportunity cost/benefit expressed in terms of the reduced interest cost of future deficits. But it is all in the future. There is no immediate financial accounting effect from seigniorage.

              This financial accounting treatment is something that is poorly understood by the economics profession and the blogosphere – because the relationship between economics and accounting is poorly understood.

              There are many other practical issues, such as the capacity for retail distribution, production cost in comparison to large denomination coins, etc. etc. But that is enough for now on the trillion coin scenario.

              • Clonal Antibody January 9, 2013 at 6:12 pm

                But Diehl says

                The Mint strikes the coin, ships it to the Fed, books $1 trillion, and transfers $1 trillion to the treasury’s general fund where it is available to finance government operations just like with proceeds of bond sales or additional tax revenues

                This 1 Trillion is booked immediately, and after deducting the cost is considered revenue to the Government, and therefore does not count towards the deficit. Tell me where I am wrong?

                • JKH January 9, 2013 at 11:41 pm

                  I’ve never “booked” $ 1 trillion in coins.

                  But suppose for a minute there’s no cost to producing any of coins, notes, or bonds.

                  Now, bonds are the only instrument that pays interest there.

                  And suppose for a minute we’re looking at a single government transaction in which $ 1 trillion is spent on – say – defense.

                  And let’s look at the unified government budget deficit.

                  So we forget about internal transactions between the Mint, Treasury, and the Fed, because on their own, they have no effect on the unified deficit.

                  First, consider the case where the coin is sold to the public.

                  Then, the government swaps the $ 1 trillion coin for reserves.

                  The private sector is now down $ 1 trillion in reserves.

                  The government spends $ 1 trillion on defense. Reserves are returned to the banking system.

                  Then the government then has a liability of $ 1 trillion, in the form of the coin it has issued.

                  Now go through the same exercise, using a bond instead.

                  Then the government has a liability of $ 1 trillion, in the form of the bond it has issued.

                  The only difference between the two is that there are future interest payments owing on the bond, which will have the effect of increasing future deficits. As I said above, that is where the seigniorage effect is captured from a financial accounting standpoint for the unified budget – as a comparative saving in the future budget deficit due to the reduced cost of issuing zero interest coins, compared to interest bearing bonds. The same applies to bank notes.

                  Now, in the case of the bond, you don’t treat the fact that the bond has been swapped for $ 1 trillion as $ 1 trillion in revenue.

                  If it were revenue, it would be a marginal surplus.

                  But it’s not a surplus. If it were, it would have an ADDITIONAL effect on the government’s balance sheet, like taxes.

                  For example, an actual tax of $ 1 trillion quite apart from what I’ve said so far would reduce reserves by $ 1 trillion.

                  And the combined effect of a coin liability of $ 1 trillion and reserves of $ (1) trillion would be a net liability of $ 0 on the books of the CTRB.

                  But the net liability isn’t zero – its $ 1 trillion in the form of the coin.

                  And the coin is correctly categorized as a liability in view of this.

                  This is really an issue of distinguishing between accounting for revenue and expenses versus accounting for the flow of funds. The issuance of a piece of paper or a coin, both of which cost 0 to produce, is a flow of funds transaction – not a revenue/expense transaction.

                  But the cost to produce the coin isn’t 0; it’s $ 180 billion in the example.

                  That’s the piece that is the accounting revenue category – or – negative revenue or expense in this case, which creates an additional $ 180 billion in the deficit. (The $ 180 billion could be considered part of the deficit spending funded by the coin, with an additional $ 820 billion in deficit spending funded by the coin – or the $ 180 billion could be funded by something else e.g. bonds in addition to $ 1 trillion funded by the coin.)

                  Now, consider the case where the coin is sold by the Mint directly to Treasury, and where Treasury deposits the coin at the Fed.

                  That deposit has no effect on the unified deficit. Nothing has happened that affects the unified government budget relative to the non government sector (leaving aside the $ 180 billion production cost, which is a net deficit effect as in the first case).

                  When Treasury spends those funds the deficit increases by $ 1 trillion.

                  It’s the same result in terms of unified deficit accounting.

                  The only difference is the net effect on bank reserves.

                  The sale of the coin to the public absorbs banks reserves. It is a swap of a coin for bank reserves. But that is a flow of funds transaction – not a revenue transaction.

                  The internal sale or deposit of the coin among the Mint, Treasury, and the Fed has no immediate effect on bank reserves on its own. Bank reserves increase when Treasury spends. That is why I’ve categorized it in the QE camp.

                  The difference is that the coin is involved in a flow of funds transaction between government and non-government when it is sold to the public.

                  It is not – when kept in-house and deposited at the Fed. But both types of coin transaction have the same effect on the unified deficit.

                  The accounting between the Mint and Treasury, taken on its own, is irrelevant in terms of categorizing the effect on the unified deficit. Any transaction between the Mint and Treasury, taken on its own, involves a double entry bookkeeping reversal within unified books.

                  So, if Diehl says:

                  “The Mint strikes the coin, ships it to the Fed, books $1 trillion, and transfers $1 trillion to the treasury’s general fund where it is available to finance government operations just like with proceeds of bond sales or additional tax revenues”

                  … I have no doubt that’s true.

                  In all three cases, there is $ 1 trillion in Treasury’s account at the Fed.

                  But the coin and bond transactions are flow of funds transactions, not revenue transactions. In the case of the coin and bond transactions, the net effect on the unified budget deficit after spending that money is a deficit of $ 1 trillion.

                  In the case of spending funded by a tax transaction, the net effect on the unified budget deficit is zero.

                  Again, the internal booking of the Mint relative to Treasury, just in respect of the $ 1 trillion, is irrelevant to the unified deficit. The Mint may book a “profit” on its own internal books, but that won’t show up as “profit” or revenue on unified deficit books.

                  So what Diehl says above – on its own – doesn’t contradict anything I’m saying. There is definitely an “availability” of funds that is booked in the TGA account in all three cases – but that doesn’t mean that the $ 1 trillion is unified government revenue in all three cases – in fact, we know it isn’t in the case of bonds (and notes), and it shouldn’t be any different for coins.

                  • Philip Diehl January 10, 2013 at 8:50 am

                    This is a very interesting discussion that goes far deeper into the accounting interactions among Treasury, the Fed and the Mint than I know about from my time at the Mint. Moreover, accounting is not my strong suit. But by my layman’s reckoning, the practical effect of seigniorage falls between that of bond proceeds and tax revenue since a portion of the Mint’s coin production is never returned to the Mint and the seigniorage never reversed.

                    For these coins, the seigniorage is an interest-free loan in perpetuity. I would view a personal interest-free loan given to me by a generous relative as an asset rather than a liability (and I suspect the IRS would view it as taxable income if they discovered the terms). So, irrespective of the accounting treatment, the seigniorage from these coins seems to be more analogous to tax revenue.

                    Of course, this is irrelevant in terms of the trillion $ platinum coin since the coin would be returned to the Mint.

                    • JKH January 10, 2013 at 9:10 am

                      thx

                      I’ve got a short post in the works, based on the discussion started here

                      probably up sometime before or by Monday, if you care to check in

                      see if it makes sense to you then, or if it seems like at least a feasible interpretation from the Mint perspective

                      but on first reading, your interpretation is very interesting and probably dovetails nicely with my mine

                      its a more complicated issue than first assumed by many I think

                    • Ramanan January 10, 2013 at 9:44 am

                      Here is what I think.

                      The government’s deficit is the difference between its expenditure and receipts (mostly taxes). The coin (general coins not necessarily $1T platinum) is one way of financing this deficit and hence not equivalent to a tax receipt. Of course, assuming that the coin is not returned back to the Treasury and stays in the hands of the public till eternity, implies less interest outlays for the Treasury and this can be thought of as seigniorage and can even be presented in tables as a supplementary item – but not as a receipt.

                      (Of course the Mint has its operations and costs and thinking of the Treasury and the Mint as one, there are costs such as wages paid, raw materials, electricity charges etc – which are part of the government’s expenditure so the supplementary item would be somewhat lesser)

                      A complication is that the profits of the Federal Reserve is counted as a receipt. This is so because the accounting of the Federal Government is separated from the Federal Reserve. So the interest paid on Treasury securities by the Federal government to the Federal Reserve is counted as expenditure and the profit as receipt.

                      So a bit counterintuitive since the profits of the central bank are counted as seigniorage but the interest saved on coins is not.

                    • JKH January 10, 2013 at 9:52 am

                      I think on first reading I agree with that Ramanan.

                      Except for the last paragraph – one man’s intuition is another man’s ..
                      :)

                      You’ve unpacked some of the components of seigniorge according to their presence in income accounting – which I think is the right way to do it – rather than present value accounting – which is what a lot of people seem to want to do, but which is fraught with pitfalls, IMO – the intuition thing again

                    • Ramanan January 10, 2013 at 10:26 am

                      Yeah that is a strange way of doing it – always reminds me of Buiter.

                      There’s a GAO on this:

                      http://www.gao.gov/new.items/d04283.pdf
                      “How the Costs and Earnings Associated with Producing Coins and Currency Are
                      Budgeted and Accounted For”

                      So in a nutshell, Fed’s profits remitted is counted as a receipt and the Mint’s profits as a reduction in the public sector borrowing requirement. (Difference between the deficit and the borrowing requirement). From a FoF angle however, both the profits are sources of funds.

                    • JKH January 10, 2013 at 10:33 am

                      Buiter – exactly

                      thanks for the link!

                    • JKH January 10, 2013 at 10:43 am

                      Preliminary Eureka (from Ramanan’s link):

                      “The profit earned from making coins, or seigniorage, is shown in the budget
                      as a means of financing the government’s borrowings and is not counted as
                      revenue in calculating the deficit or surplus for the annual budget.”

                      As I guessed, looks like – but preliminary.

              • Joe Franzone January 10, 2013 at 11:12 am

                “This financial accounting treatment is something that is poorly understood by the economics profession and the blogosphere – because the relationship between economics and accounting is poorly understood.”

                I find many in the accounting profession don’t understand this relationship either. I’ve tried to explain this to the AICPA when they teamed up with David Walker and the Peterson Institute to “Fix the Debt”. Are there any resources out there (besides this site) that you believe are helpful in understanding this relationship. Also, with regard to your comment on asset/liability management and reserve auctions do you have any recommendations for further reading? Thanks

                • JKH January 10, 2013 at 11:46 am

                  That’s a big one.

                  Ramanan, who looks in on this site regularly, may be able to help out – as he is miraculously aware of what seems to be an unlimited supply of information sources on an extraordinary range of topics.

                  On the general issue of economics and accounting, it is so fundamentally important as to define different schools of economic thought – those who get it and those who don’t. The ones who get it include post-Keynesian and related schools. The book ‘Monetary Economics’ by Wynne Godley and Marc Lavoie constructs a super impressive framework for how to think about monetary economics on the basis of stock/flow consistent accounting foundations. I suspect Ramanan has other suggestions as well, if he sees this.

                  The CFA program has always had an intelligent awareness IMO of the importance of accounting, finance, and economics taken together – as a well rounded education in this general area.

                  On the auctions and ALM, that’s really a specific example of the more general ALM function of banks. The point there is that if Treasury wants to auction reserves to banks, it will have no problem doing so. There will be a price, and banks will figure out that price as to what makes sense such that they can work the resulting term structure into their overall ALM interest rate risk profile.

  • Cowpoke January 9, 2013 at 9:25 am

    Fed can proceed to auction off reserve balances as fixed deposits held by the banking system

    JKH, can you alaborate a bit on this?
    Thanks

  • Steve Roth January 9, 2013 at 2:53 pm

    So much to say here but I’ll just do one thing:

    @JKH: “it’s really inflation risk that’s the core of that issue”

    I think: it’s really *stag*flation that’s (almost always) the core of the issue. Because we know the Fed can stomp on inflation if it needs to. Volcker showed us that. But he also showed us how painful it can be. If unemployment is already high when inflation start’s creeping (or leaping) up, the Fed’s between a rock and a hard place. And so are we.

    This is one reason I’m uncertain on NGDPLT, given the changes to the wage and profit shares over recent decades: rising NGDP is not guaranteed to bring commensurate employment growth.

    Oh another question: what happens if inflation remains a problem after the Fed has sold off its $3 trillion in bonds? What do they do? Do term deposits address that (adequately)?

    • JKH January 9, 2013 at 3:34 pm

      Hi Steve,

      - “that issue” referred to the reference to devaluation in the question, which I interpret as an inflation issue, at least superficially

      - “the issue” as you suggest may well be stagflation as a marginal effect, since in the tightening of policy there is typically some combination of a desired effect on inflation and an undesired effect on employment and real growth, as in the Volcker tightening

      - I can only encourage you to be highly sceptical of NGDP targeting or NGDPPL targeting as a policy. I think it’s a disaster – whether couched in market monetarist analysis or against a more intelligent fiscal policy approach – and for the reason you note. It’s as imprudent as a Friedman autopilot rule for money supply, IMO.

      - For either QE or augmented QE as I define it (bonds + platinum), the Fed only sells assets down to the level of residual excess reserves that provides the core requirement for transaction liquidity required by the banking system – that’s the limit of asset sales or net asset maturities. Any further asset sales would drive excess reserves negative and cause interest rates to spike uncontrollably. And the term deposit strategy only applies as a supplementary tightening measure DURING the period over which that asset decline is being managed. It is a diversification strategy for monetary tightening, so that the pace of QE exit does not depend entirely on just one element of the tightening mix – i.e. it does not depending entirely on selling assets; it does not depend entirely on increasing the target funds rate; and it does not depend entirely on issuing term deposits. This is a very wise and prudent and flexible contingency plan that the Fed has formulated. The Fed is far more astute at risk management than it is typically credited with.

    • Matt Franko January 9, 2013 at 6:06 pm

      Ha! yes “everybody” seems to be talking about it except the Fed and Treasury ;)

      (they are going to have to say something soon imo…) rsp

  • Stephen January 9, 2013 at 6:57 pm

    Great post JKH, informative and concise. Keep up the good work.

    • JKH January 9, 2013 at 11:42 pm

      thanks, Stephen

  • Ramanan January 10, 2013 at 7:27 am

    “Meet the Genius Behind the Trillion-Dollar Coin and the Plot to Breach the Debt Ceiling”

    http://www.wired.com/business/2013/01/trillion-dollar-coin-inventor/

    • wh10 January 10, 2013 at 7:43 am

      Hah! Not to be a sycophant, but congrats again Beo! Love how they have the snapshot of the post from Mosler’s (particularly the part about the anatomy of the President). (Someone should contact them about all the typos, though.)

      • wh10 January 10, 2013 at 7:49 am

        Really, as the article suggests, everyone involved deserves recognition, for their intellectual bravery and persistence. Quite remarkable, the internet is.

      • Ramanan January 10, 2013 at 9:03 am

        ” anatomy of the President”

        LOL. I hope he doesn’t read the Wired article and get upset about it and refuse to mint the coin!

  • wh10 January 10, 2013 at 9:37 am

    JKH,

    Do you think use of the platinum coin would hurt the credibility of the US bond market, like this commentator says (http://www.washingtonpost.com/blogs/wonkblog/wp/2013/01/09/the-platinum-coin-idea-is-idiotic-that-is-the-point/)?

    I agree with the idea that use of the coin would be very unfortunate from a political economy standpoint (as Cullen, Beowulf et al. say).

    However, purely looking at this through the lens of a bond investor, wouldn’t use of the coin make me even more confident that the US won’t default in the future? That should be beneficial for interest rates. (All that said, I don’t recall the debt ceiling fiasco moving rates much last year, so it seems the market is above all of this.)

    Perhaps you could spark some concerning inflation expectations, driving up interests rates on the belief that the Fed, in the future, would therefore hike rates. But the writer rules out inflation as a major threat.

    Basically, this article seems to display the hyperbolic, sloppy, sententious logic that Krugman decried a couple of days ago, the validity of lamenting our political system’s dysfunction notwithstanding. (I think he also gets the accounting wrong when he compares this to issuing bonds.)

    • JKH January 10, 2013 at 10:11 am

      I agree with Krugman’s reasoning on this.

      It’s not an idiotic idea in the current context of why its being considered.

      The article above lashes out at the coin; it should be lashing out at the conditions for which the coin is a potential legal operational solution.

      One should be more precise about what it is one is objecting to when calling an idea idiotic.

      This is not an idiotic idea in the context of operational resolution to a foolish and contradictory fiscal policy impasse with potentially damaging economic consequences.

      He’s wrong about the money supply effect in respect of a given amount of deficit being funded – assuming the coin is deposited at the Fed – the coin increases the money supply (other things equal); bills don’t.

      Default of any sort is a self-inflicted wound. I think the markets discount the probability of that self-inflicted wound at a low level by assuming resolution within some non-apocalyptic time frame, through some sort of messy process – but they discount the adverse effects of temporary deficit spending contraction over that same time frame more seriously. I think bond yields fell last time because of that.

    • Oilfield Trash January 10, 2013 at 12:50 pm

      WH10

      “However, purely looking at this through the lens of a bond investor, wouldn’t use of the coin make me even more confident that the US won’t default in the future? That should be beneficial for interest rates. (All that said, I don’t recall the debt ceiling fiasco moving rates much last year, so it seems the market is above all of this.)”

      You may be correct, but the question I am struggling with is what happens in the finance markets (global) if the USA stops issues risk free assets. It is not about the interest income these asset provide, more importantly they are also important to private sector repo operations, and leverage. You cannot repo or rehypothecate a TDC locked up in the FED’s vault.

      • wh10 January 10, 2013 at 1:07 pm

        This is a question for the experts, not me. My knowledge of these matters is very limited. I know Sankowski is big on this.

        What I personally don’t understand about this issue is that it’s not as if the USA would be issuing less assets if it just ‘printed.’ It’s just now cash instead of bonds. One of the main purposes of repos is to enable someone with bonds trade for cash. TPC replaces the bonds with cash in the first place. If you don’t consider interest income to be the issue, then what’s the major problem?

  • scepticus January 10, 2013 at 12:13 pm

    While the mutually exclusive instructions to the executive are indeed foolish I think at a more subconcious level whats really eating away at peoples psyche is the notion that PS NFAs can never under any circumstances (except perhaps momentarily during the height of a private sector bubble) be reduced.

    The reasons for this and whether it is in fact a genuine constraint or an arbitrary blind spot seem very badly understood by politicans, the public and by most economists. In fact in all the comments above it doesn’t seem that this most fundamental of all questions is actually addressed at all.

    Thats not to say that reducing PS NFAs is feasible at this juncture, the point is really under what circumstances it would ever be appropriate. A coin, psychologically, makes it seem as if the status quo is being set in stone. And I have sympathy with that view, especially if some member of the public effectively got to stamp their name on that coin. I mean, what is the method for reversing this transaction later besides confiscation?

  • Fed Up January 10, 2013 at 9:05 pm

    “Finally, perhaps the most integrating characteristic of this augmented QE approach is that both regular QE and platinum QE tranches are intended to be temporary, and their prospective amortizations are captured within specified exit contingency plans.”

    Sounds like an aggregate demand shock is being assumed. Is that correct?

  • Fed Up January 10, 2013 at 9:42 pm

    “Treasury deposits a platinum coin in its account at the Fed, the Fed will credit Treasury with funds and hold the coin as an asset. Treasury will use its funds according to what spending has already been approved by Congress, when taxes are insufficient to fund the required spending.”

    Don’t demand deposits need to be added somewhere?

  • Ramanan January 10, 2013 at 10:55 pm

    Krugman now in the print edition of NYT:

    http://www.nytimes.com/2013/01/11/opinion/krugman-coins-against-crazies.html?smid=tw-NytimesKrugman&seid=auto&_r=0

    “… Republicans go wild at this analogy, but it’s unavoidable. This is exactly like someone walking into a crowded room, announcing that he has a bomb strapped to his chest, and threatening to set that bomb off unless his demands are met.”

  • Robert Rice January 10, 2013 at 11:08 pm

    JKH,

    First, consider the case where the coin is sold to the public.

    Then, the government swaps the $ 1 trillion coin for reserves.

    The private sector is now down $ 1 trillion in reserves.

    The government spends $ 1 trillion on defense. Reserves are returned to the banking system.

    Then the government has a liability of $ 1 trillion, in the form of the coin it has issued.

    Now go through the same exercise, using a bond instead.

    Then the government has a liability of $ 1 trillion, in the form of the bond it has issued.

    You’re scheme seems to equivocate on the term liability. The “liability” generated by the sale of the coin cannot be ontologically equivalent to the liability generated from the issuance of a bond. When the Treasury mints a coin and sells that coin to the non-government sector (NGS), the two entities trade assets. The NGS receives a coin worth a trillion, the Treasury receives a credit of one trillion in its General Account. Critically, the Treasury hasn’t borrowed any money and hence doesn’t owe any entity any money. In the second case involving a bond, the Treasury has borrowed money, issued the corresponding IOU, and booked the liability (it has principal to pay to the NGS by/upon some maturity date). In what manner has the coin resulting in the Treasury owing the NGS some quantity of money (given this is the nature of the Treasury’s liability upon bond issuance)?

    I remember Kelton making a similar claim–money creation and its subsequent spending creates a liability on government–but the nature of that liability (a promise to receive the money back in settlement of a tax liability) is wholly different than what we generally regard as a liability (one entity owing some quantity of money to another).

    Perhaps I’m missing something.

    • Clonal Antibody January 10, 2013 at 11:30 pm

      Yes money is a special kind of IOU – A debt but one that cannot count against the Debt ceiling because of its nature. The only way that the IOU is extinguished is when the government receives a tax payment or other payment for services performed.

    • JKH January 11, 2013 at 4:06 am

      Robert,

      Excellent point.

      I think it may depend on how one categorizes ontological equivalence.

      In the case of both bonds and the coin, Treasury receives credit in its TGA account at the Fed.

      In the case of bonds, Treasury has borrowed in the normal sense of that term.

      In the case of coins, I think it depends on the circulating/non-circulating bifurcation.

      In the case of a circulating coin, Treasury is “liable” for redemption of that coin.

      In that sense, it has borrowed, using the coin as a borrowing instrument like a bond, but with the maturity date and maturity being a contingent liability.

      This seems very consistent with the off-budget treatment of circulating coins, in the sense that the Mint and Treasury treat these coins as substitutes for (bond) borrowing, rather than as revenue, akin to tax revenue.

      So I think it’s a reasonable logical extension on that basis to treat a circulating coin as a (contingent) liability and a funding instrument alongside bonds, where the coin in effect includes a short put option with respect to maturity date and maturity.

      In the case of a non-circulating numismatic coin or bullion, as I understand it Treasury is not liable for redemption.

      That seems consistent with the on-budget treatment of same, where these items are not treated as borrowing instruments, but as revenue, akin to tax revenue.

  • vimothy January 11, 2013 at 8:22 pm
  • Sergei January 13, 2013 at 10:22 am

    “It seems like QE has enabled us to transition to an IOR-based monetary policy system. Why not just continue with that?”

    Because, for instance, it would concentrate all interest expenses on the government deficit in the banks. Why should we prefer this outcome to the current one?

    • wh10 January 13, 2013 at 10:25 am

      I’m not sure, but it’s certainly not all. You don’t need to purchase every treasury to maintain an IOR system. The non-bank sector today still holds most of the treasuries.

      • Sergei January 13, 2013 at 10:31 am

        “The non-bank sector today still holds most of the treasuries”

        Sure, but even with current excess reserves we already create big disturbances to the financial system. Look for instance how money market funds perform. I am not defending MMFs but I am curious to know why everybody seems so much fond of IOR and how easy it is for the Fed to exit. Simply increase IOR and bingo! I do not think it will be easy at all and I find IOR given ever increasing volume of excess reserves very disturbing. It is a big subsidy to banks which they most likely will not pass over and do not deserve to keep.

        • wh10 January 13, 2013 at 10:33 am

          What do you think about Canada’s system?

          • Sergei January 13, 2013 at 10:35 am

            “What do you think about Canada’s system?”

            Sorry, I have no knowledge about Canadian banking system.