The IMF recently published a working paper (August 2012) “The Chicago Plan Revisited”:
The authors are IMF staffers Jaromir Benes and Michael Kumhof. The views expressed are those of the authors, and not of the IMF. Accordingly, we will refer to this paper as the “BK Chicago Plan”.
The original “Chicago Plan” is a proposal (or a family of similar proposals), put forward in the 1930’s, with ideas for fundamental banking system reform. The BK paper updates the general framework with a version of its own. The key elements are 100 per cent reserves against deposits, a massive “debt jubilee” for bank borrowers, and a greatly restricted scope for bank lending. It is specific with respect to tangible steps that are required in order to transform the existing system into the proposed Chicago Plan system. This should be of interest to those who are curious about actual and/or potential structural change for the financial system. For that reason, the subject seems like a good fit for the “Contingent Institutional Approach”:
The “Contingent Institutional Approach” is a general framework for comparing various monetary system designs, from the perspective of monetary and fiscal policy. The framework was applied in a follow up essay examining the design of the Eurozone system:
The assessment of the Chicago Plan will include some points of comparison with standard Federal Reserve operations, quantitative easing, and certain proposals associated with ‘Modern Monetary Theory’. So the examination of the IMF Chicago Plan, while detailed on its own account, is part of a broader analysis.
The monetary and the financial system is a design of double entry bookkeeping. It may be other things as well, but it is certainly that. But even with the financial crisis, many mainstream economists still seem unconvinced as to the importance of accounting and even the monetary system to the study of economics.
With this fundamental characteristic in mind, the overarching feature of monetary design is the relationship between the state and the “non-state”. I’ll occasionally refer to the “non-state” as the private sector, indicating the same general understanding, allowing that “private-sector” for this purpose includes its extension into the open economy foreign sector. The detailed treatment of the foreign sector is not critical in this essay.
The institutional arrangement for banking is critically influential. This includes banking arrangements for state monetary and fiscal policy. The contingent institutional approach is a framework for covering related design issues. A key element is the conception of a “Central Treasury Bank” (CTRB) as a specifically designed multi-purpose institution that offers full flexibility in the arrangement for both fiscal and monetary policy operations. In this essay, we refer to the CTRB concept and each of the central banking (CB) and Treasury (TR) functions, as naturally embedded within the fused institutional structure (CTRB) and occasionally as separate institutional entities as they are now in today’s bifurcated system. For the purpose of this essay, the banking operations that distinguish the fused institutional structure from the bifurcated institutional structure will turn out to be not so important, because the concepts of the Chicago Plan transcend the question of whether the architecture is fused (CTRB option) or bifurcated (today’s actual system). Nevertheless, it becomes convenient to assume a fused CTRB structure in order to streamline the description of transaction sequences assumed in the Chicago Plan. This is also the case because the BK paper itself is not particularly forthcoming with respect to providing operational detail for the envisaged Chicago monetary system.
This essay is an analysis but not an endorsement of the BK Chicago proposal.
The BK Chicago Plan
The “contingent institutional approach” recognizes the option of creating the institutional structure of a “Central Treasury Bank” or CTRB, which merges the functions of a government treasury operation and its central bank into a single state entity that issues fiat money. The CTRB can issue reserves, currency, and bonds for the purpose of either deficit financing or asset acquisition. There is full flexibility for any desired asset-liability profile. Deficit financing becomes a direct banking function, and there is no longer a separate treasury function that is an operational user of funds issued by either/both the central bank and the private banking system.
The core BK proposal combines several ideas as a variation on the generic Chicago Plan. The first is 100 per cent reserve banking. The second is a one-time “debt jubilee” (although that term does not appear specifically in the paper). The third is the establishment of a reformed private sector banking function that allows lending only for specified productive investment purposes.
The BK paper acknowledges the post Keynesian theme of stock/flow consistent accounting in its preliminary description of the monetary system as it exists today. But it seems to slip somewhat in describing monetary operations as imagined under the proposed system. A number of important operations appear to be omitted entirely, potentially mischaracterizing in particular the nature of the proposed system with respect to the aspect of endogenous money. So part of the purpose here is to explore these operational aspects in more detail. The Chicago Plan is an interesting example of institutional transformation – change that is absolutely massive. Achieving clarity on the nature of that change is useful. Hopefully the end result is a better understanding not only of the ideas contained in the Chicago plan, but also of the characteristics of the contrasting system that we have in place now.
Preliminary: The Role of Bank Reserves
Bank reserves are held as an asset by commercial banks, either in the form of banknotes issued by the central bank (and coins issued by Treasury in the case of the US), or reserve balances held as deposits with the central bank.
The US system has a regime of “required reserves”:
For the most part, the reserve requirement is 10 per cent of demand deposits – a relatively narrow subset of total deposits in the US commercial banking system. By contrast, the Canadian system has a zero reserve requirement.
The most relevant component of reserves for purposes of this discussion is that of reserve deposit balances, which are held at the Federal Reserve in the case of the US. These are the balances that serve as liquidity in clearing and settlement of payments among those institutions that participate in the Fed inter-bank clearing system. In normal times, this component is smaller than the inventory of Federal Reserve notes and Treasury issued coins held at bank branches. But it is the component that is most critical to inter-institutional liquidity and therefore to most of the transaction volume that flows through the financial system. It is also the component that as a stock measure has undergone massive expansion since the early stages of the financial crisis in 2008.
The normal mode of Federal Reserve operations includes a modest excess reserve buffer, supplied by the Fed in order to enable payment clearing and related reserve demand to exist at around the level of interest rates targeted by the Fed. This buffer is typically quite small, since the clearing system is normally very efficient at clearing funds. However, the Federal Reserve’s implementation of various forms of “quantitative easing” since the start of the financial crisis has resulted in an extraordinary increase in the level of excess reserves. This unusual excess level is not an accurate reflection of the role of reserve balances in their core function of providing basic liquidity to the interbank clearing and settlement system. Although this extra liquidity was helpful in the early days of the financial crisis, the size of system reserve balances now far exceeds the amount the banking system actually needs to perform this function. This reflects the Federal Reserve’s decision to use excess reserves as a source of funding for its own balance sheet, in order to support the expanded asset profile it has taken on in the management of monetary policy at the zero lower bound for interest rates. Reserves are a natural source of funding because the Fed’s own asset expansion creates reserves as funding for its own balance sheet as part of that process. This is the central bank version of the well-known post Keynesian theme that “loans create deposits”.
(Regarding the use of this language of ‘funding’: While loans create deposits, it is appropriate to describe deposits (reserves in this case) as a source of funds when viewed as part of the Fed’s balance sheet. For example, the quarterly US flow of funds report (produced by the Fed) classifies reserves correctly as a source of funds for the Fed from this perspective. The concept of reserves as funding assumes additional importance in additional discussion of reserves later in this essay.)
The Chicago Plan can be split into two time phases for implementation. The first is the transition stage, in which the existing banking system balance sheet is converted from a legacy portfolio status to a reformed portfolio status. The second is a subsequent, ongoing steady state lending operation, now and into the future, in which the banking system balance sheet is subject to policy control by various regulatory criteria assumed in the Chicago Plan, including rules for credit, liquidity, and interest rate risk management and capital structure.
The remaining part of this essay orders these various aspects into a logical sequence of operational implementation.
Chicago Plan Transaction # 1 – the reserve injection
The transition to a full reserve system would require operations by the CTRB (or Fed) on a massive scale. The system requires reserves equal to 100 per cent of reservable deposits. Because the CTRB (or Fed) is the only institution that can create bank reserves, it must produce them for the system. Accordingly, it lends the required amount of reserves to the banks (allowing for any legacy excess reserve balances already in place due to recent QE programs). Loans create deposits (in this case, reserve balances).
(Again, we use the Federal Reserve name here sometimes for convenience, at least as representative of the internal function that would exist as part of the assumed CTRB “fused” institutional structure. In this regard, it is ironic that the BK paper actually refers regularly to “Treasury” as the unit that produces the required level of reserves, which contradicts the prevailing institutional arrangements for today’s US monetary system. This is all the more reason to keep the formal and unambiguous CTRB institutional concept in mind.)
A full (100 per cent) reserve requirement, like any other, must specify the category of reservable money to which it applies. On this issue, the BK paper is aggressive in scope but somewhat vague on specifics. It identifies a reservable deposit base of far greater scope than is currently the case. This includes most of the balance sheet of both commercial and shadow banking systems. Specifically, the Chicago Plan assumes a total reservable deposit base of about two times nominal GDP, which would be a staggering $ 32 trillion in the US case. This order of magnitude then contemplates a first stage 100 per cent reserve injection of that same amount of $ 32 trillion, with a corresponding explosion in the size of the CTRB/state balance sheet.
Moreover, there seems to be a general vagueness in the paper regarding the vastly different liability structures of conventional commercial banking and shadow banking. This is a huge topic unto itself, perhaps only dominated by the sheer ambition of the Chicago Plan in terms of the scope of the proposed banking system conversion.
We are going to set aside this issue of commercial and shadow banking institutional differentiation. It is simply too large a topic for inclusion in this essay. Therefore, for illustration purposes only, we will imagine a “smaller” template, closer in size to the US commercial banking system, using some very round numbers in that context: $ 11 trillion in loan assets, $ 10 trillion in “deposits”, and $ 1 trillion in equity capital. Consider this then to be a sub-section of what the Chicago Plan contemplates in total. With that simplifying element of downsized scope in mind, we will still occasionally refer to the much large scope of what is actually found in the BK paper of $ 32 trillion in round number terms so as not to lose sight of the full ambition of this plan. Moreover, even in its own right, the treatment of the commercial banking scope alone includes a simplified representation of what in fact is a more complex liability structure than just deposits and equity capital – but the simplification helps us focus on the core characteristics of the Chicago Plan.
Thus, within that defined scope for purposes of this essay, the CTRB/Fed would lend $ 10 trillion to the banks in order to provide them with the reserves for their deposits. The banks would have a $ 10 trillion liability to the CTRB/Fed and a $ 10 trillion reserve asset. The nominal size of the Fed balance sheet would expand by $ 10 trillion.
(For simplification, we ignore the fact that in today’s actual environment, a large excess reserve position already exists due to quantitative easing programs.)
The 100 per cent reserve requirement is intended to protect depositors and the taxpayers alike from having to absorb losses that might otherwise result from the risk that banks under liquidity stress may be unable to repay deposits on demand.
Chicago Plan Transaction # 2 – the legacy loan transfer
The second stage of the BK Chicago plan transition identifies legacy banking assets that the regulatory authorities have judged to be unworthy of continued inclusion in a reformed banking system. The plan then constructs a massive transaction between the state and the commercial banking system, in order to extract these undesirable assets from the banking system and move them onto the state balance sheet.
Assume that $ 10 trillion in legacy bank assets (bank loans) are deemed undesirable as portfolio holdings in the reformed banking system.
Recall in our simplified template development that as a result of the first stage reserve injection, the state CTRB state balance sheet now includes $ 10 trillion in funding provided to the banks (as part of the reserve injection). In this second stage, the CTRB swaps its first stage funding asset of $ 10 trillion (which is a liability of the banking system) in exchange for a “payment” of $ 10 trillion. The banking system makes this payment in the form of the transfer of legacy assets to state books.
In effect, the banking system uses its $ 10 trillion loan portfolio to pay down its $ 10 trillion loan (which injected reserves) from the CTRB.
The final result is that the banking system now has a clean, post-transition balance sheet consisting of $ 10 trillion in reserves, $ 10 trillion in deposits, $ 1 trillion in residual loans, and $ 1 trillion in equity funding.
The combination of the first stage reserve injection and the second stage loan transfer can be construed more simply as a single composite transaction: The banking system swaps $ 10 trillion of its legacy loans in exchange for $ 10 trillion in reserves. And in mirror reflection the CTRB “self-funds” its $ 10 trillion loan acquisition by issuing $ 10 trillion in new reserves.
The section of the reformed banking system balance sheet not yet discussed then is the residual $ 1 trillion loan portfolio, funded by $ 1 trillion in equity.
“Instead of leaving this in place and becoming a large net lender to the private sector, the government has the option of spending part of the windfall by buying back large amounts of private debt from banks against the cancellation of treasury credit. Because this would have the advantage of establishing low-debt sustainable balance sheets in both the private sector and the government, it is plausible to assume that a real-world implementation of the Chicago Plan would involve at least some, and potentially a very large, buy-back of private debt. In the simulation of the Chicago Plan presented in this paper we will assume that the buy-back covers all private bank debt except loans that finance investment in physical capital.” (Page 6)
Accordingly, our simplified template assumes that the residual $ 1 trillion loan portfolio consists of “loans that finance investment in physical capital”. These are the loans that qualify as legitimate assets to be carried forward in the reformed banking system.
With regard to the residual $ 1 trillion equity funding, some further adjustment is required. The BK paper suggests that the state tax away most of the legacy bank equity capital position, leaving a smaller amount required as capital support for the remaining $ 1 trillion loan portfolio.
From a capital management perspective, it is perfectly rational that regulatory policy should require less equity capital for a qualified $ 1 trillion loan portfolio under the reformed system than it should for an $ 11 trillion portfolio under the pre-reform system. For example, with average loan quality improvement and tougher risk weighted capital requirements, it is still reasonable to expect that the capital requirement would drop from $ 1 trillion to something more in the area of $ 100 billion in these circumstances.
From a tax policy perspective, it seems exorbitantly onerous that $ 900 billion in value in this template example be taxed away from the equity capital owners. This is confiscatory to the point of being inconsistent with some reasonable vestige of capitalism remaining in the system. It is one reason why, although the BK Chicago proposal is interesting from an analytical perspective, it is hardly a pragmatic approach unless one also assumes fundamental upheaval in the fabric of society and its government. I will not be exploring this implicit dimension in the proposal. But from an operational perspective, an alternative mode of balance sheet adjustment is certainly available, which could be used in conjunction with equity capital downsizing, without such confiscatory methods in its execution.
Chicago Plan Transaction # 3 – The Debt Jubilee
In our example, the state has transferred $ 10 trillion of legacy banking system loans onto its own books. It now has a choice about what to do with that portfolio. One option is to treat the portfolio as a performing asset, and manage the expected cash flow due from private sector borrowers as part of the state budget. A second option is to implement a policy of mass debt forgiveness – a debt jubilee in effect. This latter option is the one selected by the BK Chicago proposal.
In effect, the state purchases the debt from the banks as described, and then writes the value of the debt down to zero on its own books. The BK paper includes detail about borrower accounts being credited to pay down the debt, but the end result is the same as just described:
“by transferring part of the remaining treasury credit claims against banks to constrained households and manufacturers, by way of restricted accounts that must be used to repay outstanding bank loans.” (Page 34)
The BK paper:
“The principal is instantaneously cancelled against banks’ new borrowing from the treasury, after part of the latter has been transferred to the above-mentioned restricted private accounts and then applied to loan repayments.” (Page 36)
The “transfer” is the debt jubilee funding mechanism. It equates to forgiveness of bank debt through a purchase and write-down of the debt by the state. The debt jubilee is arguably the most interesting part of the entire BK proposal from a monetary analysis perspective. As noted above, it is an option exercised by the state for the management of the legacy loan portfolio. As an option, it is separable from the policy initiatives of 100 per cent reserves and resetting of credit criteria for bank lending. Those changes could have been instituted instead with the bank loan portfolio that is acquired by the state left on state books as a performing asset – i.e. still owed and serviced by the original borrowers. As described above, the CTRB has purchased the loans in exchange for reserves, in effect. Had those loans been left as performing assets, the effect would have been to nationalize the legacy loan portfolio. The BK Chicago methodology forgives these debts instead – a debt jubilee. This means that the value of the loans is written down to zero – because the portfolio as held by the CTRB will no longer earn income.
The micro distribution of macro debt forgiveness depends on the relative distribution of the pre-reform debt burden across different borrowers. Those who are deeply in debt benefit directly; those who are net creditors don’t. The BK paper avoids dealing with such issues of inequity by assuming homogeneity in the wealth effect:
“In the above analysis we make the simplifying assumption, common in macroeconomic models, that there is no heterogeneity in debt levels across constrained households, or across manufacturers. This is certainly not true in the real world. But it is straightforward” (Page 35)
It seems a larger socioeconomic agenda is in play:
“A flat per capita transfer is a natural starting point for this thought experiment. But it should be clear that the transition to the Chicago Plan would represent a unique opportunity to address some of the serious income inequality problems that have developed over recent decades, by making larger transfers to lower-income households.” (Page 35)
This theme of income redistribution is somewhat repressed within the structure of the BK Chicago paper. It seems more like a stealth creature whose shadow suddenly appears. This fundamental economic motivation deserves more prominence in the paper as a matter of stated ideological orientation and policy intention.
The state’s choice to forgive the debt has a fundamental implication for balance sheets from a sector financial balance perspective. In fact, a number of interest monetary design aspects intersect at this nexus of contingent institutional design. This will be the subject of further discussion below.
Chicago Plan Transaction(s) # 4 – New Lending Activity
With the major balance sheet transactions that form the transition to the new system complete, fresh lending by banks can now go forward on the basis of revised credit criteria as specified by the Plan.
The notion of a broad regulatory framework for bank risk taking, including credit risk, is obviously not new:
“The government affects the price of lending through its control of the interest rate on treasury credit. It can also affect the volume of lending through capital adequacy regulations. But unless those regulations are tight, banks retain considerable power to determine the aggregate quantity of credit. And of course they are completely in charge of choosing the allocation of that credit. There is therefore nothing in the monetary arrangements of the Chicago Plan that interferes with the ability of the private financial sector to facilitate the allocation of capital to its most productive uses.” (Page 34)
However, the bar set by the BK Chicago proposal is higher in terms of permissible lending categories, based on the nature of the expected impact on economic activity. Specifically, the state now prescribes that bank lending be only for purposes of productive investment activity – i.e. “loans that finance investment in physical capital” (Page 6).
“After the large-scale debt buy-backs made possible by the government’s initial seigniorage gains, bank credit to households can in net aggregate terms be completely eliminated, as can short-term working capital credit to ﬁrms. This is because credit is no longer needed to create the economy’s money supply, with both households and ﬁrms replacing debt-based private money with debt-free government-issued money. The only credit that remains is lending for productive investment purposes.” (Page 19)
Thus, the BK plan proposes the elimination of a range of lending that is absolutely astonishing in its scope. This includes not only lending for speculative type activity, which one might expect from this sort of proposal, but also consumer lending in its entirety. The paper does offer a tentative escape valve from this restriction with the possibility of allowing special non-bank “investment trusts” to facilitate such additional consumer lending. But the general thrust is that consumer lending as a category of credit is treated as undesirable and/or unnecessary. The magnitude and scope of the restriction is alarming. Reading between the lines, this seems to be fostered by a strong ideological disposition toward macro state-managed income distribution. The BK Chicago core proposal assumes that all future requirements for both household and business borrowing are eliminated – with the exception of funding for longer term capital projects. In other words, everybody has a self-sustaining liquid asset position from this day on, for the rest of time, with the exception of those investment projects that pass the test for productive lending. Surely this must involve extreme assumptions about uniform future income distribution patterns that could only be achieved through comprehensive state involvement in the economy.
The Chicago plan submits bank credit origination processes to risk and capital management discipline. Banks determine the aggregate quantity and quality of credit in the proposed system, provided they work within such guidelines. This is not new in broad concept. But the guidelines for acceptable credit risk and reserve requirements for deposits have been severely tightened. Nevertheless, some essential operating characteristics of the monetary system remain in place:
“Under the Chicago Plan private financial institutions would continue to play a key role in providing a state-of-the-art payments system, facilitating the efficient allocation of capital to its most productive uses, and facilitating intertemporal smoothing by households and ﬁrms. Credit, especially socially useful credit that supports real physical investment activity, would continue to exist. What would cease to exist however is the proliferation of credit created, at the almost exclusive initiative of private institutions, for the sole purpose of creating an adequate money supply that can easily be created debt-free.” (Page 7)
Thus, the Chicago Plan appears to involve an enormous curtailment of bank credit creation. And the effectiveness of such a system rests on the implicit assumption that such credit simply won’t be needed.
Chicago Plan Transaction(s) # 5 – State Funding of Bank Lending
The Chicago Plan in generic form focuses on the objective of eliminating liquidity risk across the full liability structure of banking. First, the reservable deposit base is fully reserved. This ensures a particular bank can always make payments in respect of the withdrawal demands of its depositors. Second, funding for the bank lending activity that does proceed forward must not be in the form of reservable deposits. It must consist of “non-monetary” funding liabilities, and it must adhere to matching criteria that ensure it remains secure as funding until the contractual maturity of the lending. There can be no convertibility of such funding into liquid deposits prior to maturity of the loan. The objective is near perfection in asset-liability cash flow matching:
“The key feature of this plan was that it called for the separation of the monetary and credit functions of the banking system, first by requiring 100% backing of deposits by government-issued money, and second by ensuring that the financing of new bank credit can only take place through earnings that have been retained in the form of government-issued money, or through the borrowing of existing government-issued money from non-banks, but not through the creation of new deposits, ex nihilo, by banks.” (Page 4)
The BK form of the generic plan ratchets up the level of liquidity protection further by proposing that the state also be the exclusive source of funding for the lending section of bank balance sheets. This is intended to offer even greater protection against bank liquidity risk and bank runs.
But there is a point to be made here. Even with deposits fully “hedged” by reserves and lending “hedged” by state funding, a lending bank under the Chicago Plan may still experience a particular type of liquidity risk. The BK proposal overreaches when it makes the statement that the risk of bank runs is entirely avoided in the proposed structure. Bank runs can be fast or slow, and the slow ones can be just as deadly. Even though locked in funding as proposed may not be free to run prior to contractual maturity, it may cease to become available for continuing operations in the event of perceived balance sheet insolvency. And if actual asset losses emerge in excess of equity capital, the funding provider is subject to the risk of non-repayment.
If the lending bank that is taking on credit risk happens to become insolvent from a balance sheet perspective, it will eventually be unable to repay its funding to the state. And in the formative development of this type of exposure, the state should rationally impose pre-emptive discipline just as the private sector would with its own funding at stake – at some point the state should discontinue funding operations that otherwise would support a financial institution that is at now at risk of insolvency from a balance sheet perspective.
The BK paper:
“Bank runs can obviously be completely eliminated, as bank solvency considerations are no longer an issue in the safety of bank deposits, because money is now debt-free and therefore independent of banks’ performance in the credit part of their business. A run on the credit part of banks’ business is impossible because banks’ debt liabilities are held by the government.” (Page 51)
The state in theory has the option of overlooking the usual dynamic of bank liquidity risk as it might otherwise apply to the funding that it provides to the banks for their lending. In effect, the BK paper takes the position that there is no bank liquidity risk, because the state will absorb such risk that might otherwise have been present after depositors have been secured with 100 per cent reserves. But that seems to suggest that the state can afford to overlook and passively absorb or ignore such risk indefinitely – perhaps because of its unique “currency issuing” capacity. That does not seem to be the best way to look at risk and its potential costs. Private runs on bank deposits can be eliminated by full reserve backing, and private sector runs with respect to other types of funding can be eliminated by ruling out private sector funding altogether and assigning responsibility for it to the state. But liquidity risk is still an essential consideration for the relationship between the bank and the state as it relates to the ongoing operating viability of the lending bank. The state can “shield” the bank from residual liquidity risk, but in doing so may continue to fund a bank that might otherwise have emitted signs of solvency concerns much earlier had the system been subject to greater private sector market discipline. The result might be greater losses in the long run, due to the protracted death spiral of a bank that is still obtaining secured funding from the state.
The Ubiquity of Endogenous Money
The standard modern monetary system includes commercial banking activity in which “loans create deposits”. Banks create money “out of thin air”. This basic point is emphasized in particular by post Keynesian writers. As a fact, it is undeniable.
Solvent banks in principle are not constrained in lending by the supply of reserves – whatever the reserve ratio requirement. Solvent banks can always borrow reserves from the central bank in its lender of last resort (LLR) role. Anything to the contrary would impose the risk of gross systemic dysfunction, since the central bank can’t rationally declare a solvent bank to be insolvent simply because the bank happens to be short funds in a particular operational situation. And the central bank supplies the reserves that are needed by the entire system on an ongoing basis in order to manage interest rate levels according to a policy target.
Endogenous money is an automatic consequence of double entry bookkeeping. Loans create deposits within the accounting structure of commercial bank balance sheets. However, that statement alone is a simplification of a more refined reality. The form of bank liability created from lending also includes variations on the basic demand deposit, depending on the variety of liability types offered by the bank that issues them. For example, the recipient of a borrower’s funds may ask his bank for a (non-reservable) fixed deposit rather than a demand deposit. Or, the bank itself may issue new debt or equity securities, with that form of funding sourced partially from new loans or from the conversion of demand deposits already held with the banking system but originally associated with new lending. Thus, a more general form of “loans create deposits” in banking might be the more sweeping “assets create liabilities and equity”. But the standard “loans create deposits” is most often the relevant fact at loan origination and adequately expresses the intended meaning.
It remains bewildering that the economics profession still does not uniformly acknowledge the fact of endogenous money as a broadly accepted proposition. Some progress has been made over the past few years (including educational efforts from various post Keynesian writers, and the IMF itself in this and some other papers). But the economics profession shouldn’t find itself in a position of debating the basic facts of banking – and yet it does so still, even with a monumental financial crisis to help focus the collective mind on such issues.
An important point of emphasis here is that whatever form “loans creates deposits” assumes, this is a statement of loan and deposit origination and creation. What happens subsequent to creation is a different and much deeper conceptual and operational beast that involves robust competition among individual banks for a desired share of the totality of banking system deposits and other forms of funding. There is a difference between an event of loan/deposit creation, and subsequent events that alter the status of such original loans and deposits through further banking transactions. The acknowledgement of this difference evidently has been problematic in blogosphere discussion and interpretation, where participants sometimes claim that “loans create deposits” implies the stronger but misleading proposition that banks “don’t need” deposits. Nothing could be further from the truth in the actual operations of the modern banking system. There is considerable competition for deposits, just as there is in lending markets. Banks do not rely indefinitely on lender of last resort privileges as a chronic source of funds in today’s system. If they did, they would lose their banking licences. The fact that a lender of last resort facility (LLR) is available should never be expanded to a general proposition along the lines that banks “don’t need” deposits – at least not unless this includes reform to the point of eliminating competition for such deposits. The banking system we have is set up for competition and capitalism in deposit gathering. The Chicago Plan seems to want to lobotomize this normal competitive reflex as far as bank deposit gathering is concerned.
The BK paper also seems to stray on the subject of how the Chicago Plan relates to the fate of endogenous money flows in which “loans create deposits”. It seems to imply the demise of that basic dynamic. But this cannot be the case.
By logic, a 100 per cent reserve requirement has no unique effect insofar as the creation of endogenous money is concerned. 100 per cent is just another number – like 10 per cent (e.g. US) or zero per cent (e.g. Canada). Beyond the Chicago Plan in particular, it seems that many proposals that include a requirement for 100 per cent reserves fail to recognize this fact.
The US banking system has a basic reserve requirement of 10 per cent of reservable money (mostly demand deposits). Canada has a 0 per cent requirement. Both of these systems generate endogenous money with equal operational ease. Suppose that a banking system existed with a 90 per cent reserve requirement. Such a system would still generate endogenous money, no less so as a fact than the two systems noted above. And if such a system converted from 90 per cent reserves to 100 per cent reserves, it would still generate endogenous money. The concept of endogenous money does not suddenly disappear at the 100 per cent reserve level. Full reserves are not a discrete inflection point in this regard. In particular, endogenous money continues to be created and to endure (at least in part) in the BK Chicago Plan.
A private sector commercial banking system cannot originate new loan assets without establishing offsetting private system liabilities of some sort. This is the case because the sequence by which borrowed funds are transferred to subsequent agents will at some point include an event where an agent will deposit the borrowed funds back into the banking system. The act of lending generates required offsetting liability bookkeeping somewhere in the system. That liability is at least contingently a deposit, although the system may issue a non-reservable deposit or a non-deposit liability in its place, as noted above (e.g. non-demand deposit forms, debt, equity). If a private sector banking system lends to the non-bank private sector, it will create endogenous money as the inevitable result.
Such a result could only be precluded if the recipient of funds created by bank lending were to deposit them in a state banking institution (e.g. Fed, Treasury, CTRB), in which case the state would become a banker to the general public. But the Chicago Plan proposes no such institutional arrangement. Therefore, some private sector recipient of borrowed funds will deposit them back into the private banking system – likely in the form of a reservable deposit – and loans will have created deposits.
We’ve said that the BK Chicago Plan specifies state funding for new lending. And now we’ve said that new lending generates new deposits. How can this be? Both of these things – state funding and deposits – are liabilities of the banking system. How can both of these things be happening at the same time in respect of the same amount of new lending?
The answer is that state funding for the loan also results in an identical increase in the amount of reserve balances in the account of the lending bank.
The lending/funding combination in total looks like this:
a) Bank loan
b) State funding of bank loan
c) State funding results in new reserves for the lending bank
d) Loan creates reservable deposit(s) (fully reserved via c))
The amounts reflected in each step above are identical. And we will see that the complete means by which state funding “results in” new reserves for the lending bank in c) can vary. The way in which this happens is important to the interpretation of endogenous money in the context of the Chicago Plan. This is described in more detail below.
Note the crisscrossing of asset and liability associations inherent in the steps above. The originating flow of state funding results in an offsetting reserve asset. The originating flow of bank lending creates an offsetting deposit liability. But the steady state balance sheet conceptually matches state funding with bank lending and the reserve asset with reservable deposits.
In fact, similar asset-liability management techniques are practiced more generally in commercial banking today. A central treasury and risk management function assembles the full sleight of incoming liability formations and outgoing asset formations – and matches off the risk characteristics of each to the other in a process of natural balance sheet hedging. This rearrangement of original flows into revised interpretations for risk management – with additional hedging transactions initiated as necessary – is basic to bank treasury and asset-liability management.
Two additional points deserve emphasis. The first is that the timing of the reserve injection associated with endogenous reservable money creation is irrelevant to the fact of endogenous money creation. The second is that endogenous money creation does not preclude the contingency that the relevant state institution (CTRB/Fed) can “drain” such endogenous money along with matching reserves, by undertaking reversing operations such as is done today in the case of Fed system reverse repos, where the effective ultimate counterparty is a non-bank. But the fact that a monetary authority can “drain” reserves along with bank deposit money does not negate a natural endogenous money creation process. With those two provisos in mind, there are a number of statements in the BK paper that risk mischaracterizing what should be acknowledged as the continued existence of an endogenous money process under Chicago Plan conditions. And the reader may wish to work through the next two sections of this essay in order to see the picture that is gradually being revealed from this perspective.
Meanwhile, some further co-ordination of references in the BK paper:
“The control of credit growth would become much more straightforward because banks would no longer be able, as they are today, to generate their own funding, deposits, in the act of lending, an extraordinary privilege that is not enjoyed by any other type of business.” (Page 5)
This warrants qualification. The act of lending still creates deposits. The state provides funding to match lending, and in so doing brings reserves to the lending bank. The increase in reserves matches the reserve requirement associated with newly created reservable deposits.
The BK paper:
“Credit is no longer needed to create the economy’s money supply, with both households and ﬁrms replacing debt-based private money with debt-free government-issued money. The only credit that remains is lending for productive investment purposes.” (Page 19)
Again, some qualification is appropriate. Credit (i.e. bank lending) does create new money. And credit is needed in particular to produce an enduring non-zero growth in the money supply, which is a central objective of the Chicago Plan on a controlled basis. We will describe this process in greater detail below. The fact that bank credit is eventually matched by non-deposit funding within an asset-liability management process doesn’t arrest the process by which credit creates endogenous money (deposits).
The BK paper:
“The government affects the price of lending through its control of the interest rate on treasury credit …. It can also affect the volume of lending through capital adequacy regulations.” (Page 34)
These points are not fundamentally different in concept from what is already in place for the existing monetary system. The Federal Reserve today sets the benchmark risk free rate as the expectational anchor for the full yield of government liabilities. That yield curve is used as the interest rate foundation for risk adjusted interest rates. The BK Chicago structure is not fundamentally different in principle in that credit risk pricing starts with the yield curve created by the state’s choice(s) of risk free rate(s) (or risk adjusted rates as may be the case) as the benchmark cost of funds that it lends to the banks and the (zero) rate it pays on reserves. Banks will continue to price their lending rates off the relevant cost of funds in order to generate an acceptable expected risk adjusted return on equity, allowing for the constraints of capital adequacy regulations.
The BK paper:
“The Chicago Plan provides an outline for the transition from a system of privately-issued debt-based money to a system of government-issued debt-free money” (Page 17)
“What would cease to exist however is the proliferation of credit created, at the almost exclusive initiative of private institutions, for the sole purpose of creating an adequate money supply that can easily be created debt-free” (Page 7)
Privately-issued debt-based money” continues to be issued. “Government-issued debt-free money” is present in the form of reserves, satisfying a requirement against reservable deposits – which is no different in principle than is the case for existing systems with non-zero reserve requirements (e.g. the US). The Chicago plan should not be characterized (in this dimension) as a transition from private issued to government-issued money, but as transition to a more onerous reserve requirement. Loans still create deposits (or some form of bank liability).
The BK paper:
“We assume that the deposit function of banks is perfectly competitive, and that banks face zero marginal costs in providing deposit services. In the pre-transition environment the government only has a single policy instrument, the nominal interest rate on short-term government bonds, to affect both money and credit… As we have discussed, the effect on each is indirect and weak. In the post-transition environment the government directly controls the quantity of money, while its control over credit is still indirect but potentially much more powerful. Furthermore the government, as the sole issuer of money, can directly control the nominal interest rate on reserves, and given our assumptions about banks’ technology, this rate is passed on to depositors one for one… To control credit growth, policy is assumed to control both the nominal interest rate on treasury credit and a countercyclical component of capital adequacy requirements… As the latter falls under prudential policy, it is discussed in the next subsection.” (Page 34)
Again, this description should be weighed in the context of persistent endogenous money creation. The government is not “the sole issuer of money”. It is the sole issuer of bank reserves, which is no different than the case today. Commercial banks continue to issue money in endogenous form (e.g. deposits). The state conducts reserve adjustment operations in parallel with endogenous money process – which in principle is also what happens in today’s system. The paper may confuse the dynamics of issuing money with subsequent asset-liability management adjustments that are required to hedge certain risks associated with that money creation process. In conjunction with this general objective of hedging such risks, the required reserve ratio is elevated to 100 per cent, which is obviously a stark difference of degree compared to today’s systems. But endogenous money creation continues as the consequence of lending, and controls over the volume of lending are indirect through regulation of credit risk and capital adequacy criteria, just as they are today, although with a narrower scope of permissible categories of lending.
We should acknowledge that there is one substantive technical feature associated with the state selection of a 100 per cent reserve ratio. The state can now “shadow” endogenous money creation with state money, dollar for dollar, through a matching reserve injection. This in turn allows for the option of withdrawing all or any of these amounts from the banking system through “draining” monetary operations. This becomes the lever for the control of money supply growth, as described in the next section below. But executing such a reserve/deposit withdrawal does require some variation on the type of open market operations that exist in the current system. Moreover, it requires a facility with an ultimate effect on both reserves and bank deposit liabilities, because that dual interface is needed in order to withdraw bank deposits from the system according to money supply control objectives.
The BK paper:
“This means that banks cannot lend by creating new deposits. Rather, their loan portfolio now has to be backed by a combination of their own equity and non-monetary liabilities” (page 34)
But banks do create new deposits in lending. Again, I have more detail on how this should work in the context of Chicago Plan money supply constraints, in the next section.
Chicago Plan Transaction(s) # 6 – State Liability Management and a Money Supply Rule
The BK Chicago Plan specifies a Friedman-like money supply control function. (Friedman apparently liked the generic Chicago Plan because of its effectiveness in this regard.) But the paper again seems to bypass required operational details, including the inextricable linkage between such a control function and the dynamic of endogenous money creation.
Lending under the reformed system still creates endogenous money (deposits). The state’s funding of bank lending results in the reserves required to back those deposits. The state funding is designed to hedge the liquidity risk otherwise associated with funding of incremental lending. This general configuration establishes the framework within which the state can take further steps to control money supply growth. The fact that deposits are created equal in amount to 100 per cent of new lending means that the endogenous money creation process must be largely offset by some state action that has the effect of reducing such deposits to a level that is consistent with its money supply control target. The state accomplishes this through liability management of its own balance sheet.
At this point, the “central treasury bank” concept (CTRB) becomes helpful. In this regard, it is interesting that the BK paper starts out by referring to a “treasury” function when discussing the first stage of the Chicago Plan implementation – the injection of bank reserves that is required to provide 100 per cent reserve back stop for the legacy deposit portfolio. Recall that this amount is $ 32 trillion in the case of the BK discussion, and $ 10 trillion in our sub-template discussion. This is interesting because a conventional government treasury function does not inject new reserves into the banking system. This aspect was described in the “contingent institutional approach”. And it was treated in great detail by Brett Fiebiger, here:
It is the conventional central bank function that injects reserves into the system when necessary. The BK paper seems to assume something else in its use of the term “treasury”. The framework for the analysis is more compatible with the formal assumption of a well defined CTRB institution, which has full flexibility of asset and liability instruments and the integrated effect on the bank reserve system. As noted in the introduction to this essay, such an assumption can always be reversed-engineered into the more complicated real world arrangement of a bifurcated central bank and treasury institutional arrangement. But we’ll go in general with an overarching CTRB type description in what follows:
The first thing to note is that conventional Federal Reserve operations have the capability of draining both reserves and deposits from the banking system. What is required is simply that the Fed must sell assets into non-bank portfolios. (This can happen with dealers as intermediaries.) This draws down both deposits of non-banks with banks and bank reserves. This reduces both narrow and broader measures of money supply.
In the context of the Chicago Plan discussion, the state CTRB can do the same thing, and can affect money supply in the same way. But the CTRB has an additional tool to drain reserves that is not typically used in the bifurcated institutional system. It can drain reserves and bank deposits by issuing debt.
(Such debt is issued by a separate treasury function in the bifurcated system. Standard operations are that that treasury function does not change the reserve setting in a significant way in actual operations. The converse is true only in a counterfactual comparison. The logic of distinguishing between factual and counterfactual descriptions of monetary operations was the motivation behind the concept of a formally prescribed CTRB institution, as suggested in the “contingent institutional approach”.)
The availability of both reserves and debt as instruments of active liability management for the CTRB is the key insight to how the Chicago Plan would manage the money supply in an endogenous money environment.
A Chicago Plan CTRB could issue bonds to non-banks, thereby draining both reserves and deposits from the banking system.
The Chicago Plan thus sets a target growth rate for reserves and deposits, regardless of the volume of private sector bank lending and the amount of endogenous money first created in that process. The state will drain any excess reserves and deposits down to the level of the residual target for money supply growth.
We now return to the earlier point where we left off on the discussion of lending and funding. The language used there was that state funding of bank lending results in an increase in the reserve account balance of the lending bank. This increase in reserves can now be related to the state’s CTRB liability management process directly. Liability management strategy determines where those reserves really come from.
Suppose, for example, that the state provides funding for a new $ 10 million loan. That funding will be offset by an increase of $ 10 million in the reserve account of the lending bank.
And suppose that’s it – there is no further operational adjustment by the CTRB.
In that case, the CTRB will have created reserves in the act of funding the lending bank. And the CTRB’s balance sheet will show a $ 10 million asset, and a $ 10 million reserve liability. As noted in earlier discussion, it is a reasonable use of language to say that the CTRB is funding its own balance sheet requirements with $ 10 million in reserves, acknowledging that this reserve funding has been created in the act of asset acquisition (the funding provided by the CTRB to the lending bank.) Again, this is “loans create deposits”, this time at the level of the CTRB.
In this case, the lending bank has created endogenous money (deposits) and the CTRB has created reserves. And we can now use the word “create” unambiguously as it applies to reserve creation in the sense of this specific example. At the margin, this transaction has increased broad money supply (deposits) by $ 10 million and bank reserves by the same amount. And in doing so, the credit criteria and liquidity management criteria of the Chicago Plan have still been satisfied.
This is an example of a Chicago Plan variation where in fact there is no control over the amount of endogenous money that is created, even though deposits are 100 per cent reserved.
But BK Chicago does impose control over money supply growth. How does it do this?
In the context of our construction here, it could add a specific turning of the dial (or twisting of a Rubik’s cube analogue) through the contingent institutional approach. It does this through exercising a CTRB option to issue debt.
Suppose under the Chicago Plan that it is determined that the state should accommodate a specific targeted level for money supply growth – say 3 per cent per annum.
And suppose the Chicago Plan’s economic forecasters project annual banking system loan growth, now under much more restrictive credit risk management criteria, at 6 per cent per annum.
Then it is possible for the CTRB to aim for 3 per cent money supply growth through a liability management formula that is directly linked to the endogenous money creation process.
Specifically, in response to the requirement of banking system credit growth for CTRB funding, the CTRB may strategize its own liability mix at 50 per cent reserves and 50 per cent debt.
The effect of this is that 50 per cent of projected loan growth will be funded ultimately by CTRB reserve creation. And 50 per cent will be funded ultimately by the CTRB issuing new debt. That debt issuance will result in the transfer of already existing reserve balances from the reserve account of the bond buyer’s bank to the reserve account of the lending bank.
Referring to our $ 10 million loan example, $ 5 million of the new reserves that show up in the lending bank’s reserve account will have been created by the CTRB as new reserves for the banking system as a whole. And $ 5 million of the reserves that are new to the lending bank will have in fact been transferred in from some other bank’s pre-existing reserve position. This is the case because the CTRB debt issue will, in effect, “destroy” both reserves and deposits on the books of the buyer’s bank. That is what happens at the microeconomic level through the process of interbank clearing and settlement when such a bond issue is sold ultimately to a non-bank.
And the result in terms of the CTRB’s balance sheet is that active liability management has produced a funding mix of $ 5 million in newly created reserves and $ 5 million in newly issued debt.
Meanwhile, the lending bank’s balance sheet has grown by $ 10 million in lending and funding, plus $ 10 million in reserves and deposits. Total balance sheet growth is $ 20, assets and liabilities. This includes $ 5 million in new reserves and $ 5 million in reserves cleared in from another bank.
However, the banking system balance sheet in total has grown by only $ 15 million. The $ 5 million difference corresponds to the reserves and deposits that were extinguished on the books of the bank of the bond buyer.
This is the case that is most representative of the Chicago Plan as a reformed system in which endogenous money creation still exists, constrained by a money supply control function.
Thus, the final liability mix of the CTRB will reflect both reserve and bond debt components, consistent with the objective of money supply control. (Banknote currency is left aside by the Chicago Plan, but this is an easy enough adjustment.) The reserve component of the CTRB liability structure will directly reflect the target for money supply growth, given the 100 per cent matching requirement against bank deposits. The natural endogenous banking dynamic means that deposits created by lending (with matching reserves created by the state) are partly offset through a reversing process of state intervention in draining some of the combined deposit/reserve growth that otherwise accompanies such lending. As another simplified example, suppose the state objective for money supply growth translates to 25 per cent of lending growth, on an ex post facto trend basis. Then it will reverse out 75 per cent of the endogenous money creation. This will leave 25 per cent deposit growth as a percentage of loan growth, that is 100 per cent reserved. This is an intricate but fundamentally important balance sheet dynamic that appears to have been overlooked by the BK paper. But it is required, given the intended control of (non-zero) money supply growth. And importantly, the core dynamic of money supply growth does in fact require endogenous money creation, just as it is today.
The state maintains tight control over money supply growth in this process because at its option it can hedge away any amount of reserve/deposit growth associated with lending by issuing debt. Debt issued by the state and purchased by non-banks drains reserves and deposits together. This is where the unique functionality of 100 per cent reserves comes directly into play – because it allows the state to choose its money supply control rate at any level up to the limit of draining 100 per cent of new deposit money created, meaning zero money supply growth.
Moreover, it is necessary that the process of reserve injection followed by reserve drain take place in two discrete steps, as described. This is because the lending bank requires state funding for the loan, and the state can only provide this funding in a discrete step that includes full provision of reserves as part of the transaction. And this is because the state transacts in exogenous money. The funding injection induces a high powered money injection (HPM) in the form of reserves. It is only when the funding is identified as a discrete step that the state can drain the associated reserves by issuing bonds. This activity may become somewhat bundled in terms of timing, but the discreteness of the required transaction components remains clear.
This is also a case in which the dynamic of seamless CTRB liability management becomes clear. As noted in the introduction, this could be reverse engineered into a comparable set of transactions under today’s actual bifurcated central bank and treasury structure, but the fused institutional structure offers a clean picture of the monetary operation effects.
The BK paper:
“The main reason why monetarism had to be abandoned in the 1980s is the fact that under the present monetary system the government can only pursue a rule such as for the narrow monetary aggregates under its direct control. This is only effective if there is a stable deposit multiplier that relates broad to narrow monetary aggregates. However, as discussed in section I, the deposit multiplier is a myth. Under the current monetary system, broad monetary aggregates are created by banks as a function, almost exclusively, of their attitude towards credit risk, without narrow aggregates imposing any effective constraint. On the other hand, under the Chicago Plan the rule directly controls the broadest monetary aggregate. In other words, monetarism becomes highly effective at achieving its main objective. This is the main reason why Friedman (1967) was in favor of the 100% reserve solution.” (Page 38)
Yes, the Chicago plan controls the “broadest monetary aggregate” – i.e. deposits. But it controls this in effect through quasi-open market operations that are required to drain endogenously created deposits down to the level of the desired growth trajectory. And yes, monetarism can work much more effectively under BK Chicago – as an operational concern. But whether it works effectively as an intended influence on economic growth and inflation is quite another matter – just as it is in the environment of the existing monetary system. The Chicago Plan produces a more controllable money supply function as a by-product of the 100 per cent reserve specification – which is presumably why Friedman happened to like it – for what is in effect an ancillary feature. But that operational advantage (for monetarism) addresses nothing in substance regarding the question of how relevant monetarism remains today. The advantages offered by the Chicago Plan in this regard presuppose advantages offered by money supply targeting – which itself is a highly debatable proposition.
In summary, the BK paper specifies that the state be the source of non-deposit funding for bank lending. It implies that this requirement displaces the creation of endogenous money. Endogenous money creation continues, and endogenous money remains, with the central bank holding an option to drain as much newly created money from the system as it desires. The 100 per cent reserve requirement is what gives the central bank that full flexibility. But even though that option (depending on how it is exercised) may change the equation for endogenous money endurance, it does not change the fact that endogenous money creation remains a natural feature of the system. And in fact, the central bank does allow for a targeted growth in money supply, which it can achieve only by NOT draining all the endogenous money that has been created. In other words, endogenous money remains – not only in the dynamic creation sense – but as a matter of sustained stock balances according to money supply growth targets.
Space, Time, and Required Reserves
The discussion of endogenous money thus far has focused on the spatial dimension of required reserve dynamics. But the timing dimension of required reserves is also critical to understanding the full context of endogenous money. The timing dimension pertains to the relationship between the timing of endogenous money creation and the timing of the appearance of reserves that may be required in conjunction with the creation of reservable money.
It is well known (by those who understand it) that in the existing fiat monetary system banks don’t need to find the reserves to support a new reserve requirement until the deposit that creates the requirement has itself been created. The BK Chicago paper actually does acknowledge this as a feature of the current system, which is good to see.
But the BK paper then seems to imply (but it is not entirely clear) that the Chicago plan requires that reserves be sourced in advance of lending and that this somehow equates to the belief that today’s endogenous money creation process has been eliminated under Chicago, generically.
There are a number of problems with such an interpretation. We have already addressed the spatial dimension of what must be a continuing endogenous money creation process under the Chicago Plan. We can now address the aspect of required reserve timing by examining the two alternatives of reserving sourcing – a focus on whether reserve sourcing precedes or follows the lending and deposit creation that it must associate with.
Suppose the state provides advance funding for bank lending. That results in a bank funding liability and bank reserves as an asset. So the bank has a surplus reserve position, and is now ready to lend. Note that this operational ordering seems to suggest implicitly that the bank is “lending reserves”. The bank is NOT lending reserves. This characteristic – that banks don’t lend reserves to non-banks – is well known (by those who understand it) as it applies to the existing monetary system. This does NOT suddenly change under the Chicago Plan. This is fundamentally important.
With that said, the bank makes its loan to its customer. Note that because the state has provided funding in advance, the bank’s funding liability will be designed specifically to hedge the liquidity risk otherwise associated with the loan/funding combination, as per the risk management objectives of the Chicago Plan. The loan and its matched funding are therefore in place, hedged for liquidity risk.
That leaves the question of how the reserves and the deposit that must be created by the loan will settle within the framework of the banking system balance sheet as a whole. And that depends on how the borrowing customer is credited with funds, and what he does with the funds.
The fact is that the loan must create a deposit, even under the Chicago system. There are two ways in which this can happen.
First, the lending bank may credit the deposit account of the borrowing customer. The immediate effect is that the new deposit is fully reserved, because the bank still holds the reserves created by state funding. This is an instance of the crisscrossed but logical asset-liability management process referred to earlier.
Alternatively, the lending bank may write a cheque/draft to the borrower, which the borrower (for some reason) deposits in another bank. In this case, the recipient bank will credit the deposit account of the borrower and claim reserves from the lending bank via the interbank clearing and settlement process. The result is that a new deposit is fully reserved on the recipient bank’s books, and the loan is fully hedged (by the original state funding) on the lending bank’s books.
This description should make it clear once again that there must be an endogenous money process under the Chicago Plan, and that the true endogenous money propositions that prevail under today’s actual monetary system (loans create deposits; banks don’t lend reserves to non-banks) must still prevail under the Chicago system.
The important operational point here is that the private sector commercial banking system has created its own deposit. This deposit has not been “created by government money” as seems to be the descriptive language in the BK paper. In the example just noted, the new bank deposit liability may have been created at a bank other than the lending bank. But this is irrelevant. The loan has resulted in a deposit. And that deposit is in addition to the state funding that was designed to back the loan. Thus, a requirement for advance reserves provided by advance funding from the state certainly does not displace the creation of endogenous money.
Moreover, the idea of advance funding is problematic from both logical and operational perspectives. It is only sensible that the loan in question be negotiated with the customer in accordance with regulatory requirements for the type of credit risk that the Chicago plan envisages, with matching funding then specified in accordance with regulatory requirements for liquidity management. If the purpose of regulatory constraints on funding is to hedge the liquidity risk of specific loans, then the specification of funding must logically flow from the liquidity characteristics of the loan itself. That is the natural order of hedging risk in bank asset-liability management. Then, provided the loan fits within regulatory guidelines, there is no reason for the state to deny the funding that it is expected to provide in accordance with the Chicago plan. Thus, the precise operational pre-positioning of funding and reserves is not only unnecessary, but logically dysfunctional from the perspective of the natural ordering of operational commitment to lending followed by funding, even in a Chicago reformed system. Funding specification flows from asset specification in asset-liability management.
So let’s have a look at the alternative scenario, in which a private sector bank in the BK reformed system lends without having any associated funding or reserves already in place at the moment of lending. Again, there are two possible outcomes in terms of the inevitable endogenous money process.
The first is that the lending bank credits the borrower’s deposit account with funds. Assuming this qualifies as a reservable deposit, the lending bank is now short reserves in respect of that deposit and must go to the state to borrow reserves to cover that requirement. At the same time, it requests that the form of borrowing be designed so as to create a suitable hedge for the liquidity characteristic of the loan as required by the Chicago Plan.
The second possibility is that the lending bank presents the borrowing customer with a cheque/draft which the customer (for some reason) deposits with another bank. The payee bank clears that cheque back to the lending bank through the interbank settlement mechanism, and the lending bank now owes the payee bank reserves. In this case, the lending bank must borrow reserves from the state, not to reserve against a new deposit it has on its books, but to source reserves it owes as payment to the payee bank. When that happens, the form of borrowing must simultaneously be tailored to match and hedge the liquidity profile of the loan. After the state provides reserves to the lending bank, that bank can satisfy its interbank clearing obligation and the payee bank now has reserves it requires against its new deposit. Note that where the loan is advanced by a cheque/draft that eventually clears back from another bank, the only immediate purpose for reserves is that of operational settlement of the cheque through the interbank clearing system. The lending bank does not require reserves against deposits in this case – because it has not created the deposit. That happens at the payee’s bank.
So that’s how it works, whether required reserve sourcing occurs before or after the immediate creation of the loan. Thus, the fact of endogenous money creation is technically independent of the timing of state funding and reserve injection.
In practice, the matching up of the lending and funding and reserve injection would be timed to be as coincident as possible, so that risk hedging and pricing is on sound footing as between asset and liability risk matching. But these risk management considerations are quite separable from the fact of the dynamics of endogenous money creation and the timing of facilitating funding. And this is really no different than what occurs in today’s banking system when the risk characteristics of particularly large or complex transactions are locked in for hedging purposes in fairly precise timing sequences. And of course, this aspect of risk hedging holds a fortiori in fast paced trading operations.
Notwithstanding the rule of matching lending with state funding, it is not unreasonable to anticipate the odd situation where the lending bank may be short of reserves at an operational level, for whatever reason, at the end of the day. Suppose hypothetically, that the lending bank has not completed the state funding transaction that it needs to match against a new loan. This could result from some operational glitch in a normally smooth lending/funding matching operation. In such a case, the lending bank may be in a position where it is forced to borrow reserves temporarily from the state in the latter’s capacity as lender of last resort (LLR) – i.e. before the transaction that locks in permanent state funding has been entirely negotiated and/or settled (for whatever reason). The only way the state could deny such last resort liquidity at the operating level is if it shuts down the lending bank. And this is simply a non-starter as a matter of ongoing system reserve management. The state won’t declare a solvent bank to be insolvent just because it has experienced some atypical liquidity interruption in its ongoing operations. In the normal course, such LLR utilization should not be necessary, given the intention of the reformed system to match lending and funding in a closely controlled manner. But the fact that it is operationally possible and that that such a risk must be accommodated sensibly in operational terms is further evidence of the endogenous money dynamic. Thus, in such an exceptional case the CTRB’s supply of accommodating reserves is forced by double entry bookkeeping, in the light of rational standards for recognizing and acknowledging bank solvency. Again, this is no different than the operation of the monetary system we have today. The state LLR function is always faced with the implicit or explicit choice of providing liquidity to a solvent bank, or denying liquidity to a bank that is judged to be insolvent. Broadly speaking, the state LLR function, which as noted above must continue to exist, must regularly make decisions that reasonably distinguish between operational liquidity issues and fundamental solvency issues in evaluating the criterion for execution the LLR option in response to a particular bank’s need for it.
While the Chicago Plan reserve ratio of 100 per cent is much higher than exists in today’s systems – e.g. 10 per cent in the US and 0 per cent in Canada – it remains the case that the state will produce those reserves on demand for a solvent bank. The only effective constraint on lending volumes and the supply of matching reserves is the continuation of a monetary policy framework that in the most general sense already exists today – foundational pricing discipline in the form of an administered policy interest rate (today, the discount rate), and quantity discipline in the form of Basle type constraints on capital ratios etc., as noted in the paper.
So if the BK paper places any emphasis on full reserves as a constraint on lending or as a constraint on the creation of associated endogenous money, that emphasis is misplaced. In this regard, the Chicago Plan version of 100 per cent reserves – including debt jubilee and onerous credit risk restrictions – is only one design possibility for a full reserve system. Numerous other proposals promote 100 per cent reserve requirements. And the fundamental analytical questions we’ve raised here probably apply to most of them.
With regard to the BK Chicago Plan, the advance securing of funding and reserves (if that is what the paper suggests) is not a substantive issue in the dynamics of endogenous money creation. There is no effective funding or reserve supply constraint when lending is subject to broader credit risk criteria as is the case here. The state manages the system according to a policy interest rate, risk criteria, and capital requirement supervision, just as it does now. Shuttering a bank is an option in cases of presumed insolvency, but cannot be an operating option in the case of solvent banks that happen to need more reserves to cover their immediate needs. Thus, any suggestion that a bank absolutely needs to source reserves prior to lending and/or deposit creation would be nonsense. The discipline on lending volumes comes from the imposition of credit criteria – not from the imposition of 100 per cent reserves – just as is the case today with reserve requirements other than 100 per cent. Furthermore, there are sound operational reasons why a bank should be able to source required reserves after the moment of lending, within regulatory guidelines. And this is the way in which actual reserve systems work today.
In summary, the paper appears to bypass the operational detail whereby private sector bank lending will still create private sector deposits (or alternative bank liability forms) under the Chicago Plan. This consistent dynamic cannot be avoided in a coherent system of double entry bookkeeping. If a private sector borrower receives funds, it will disburse those funds, and the ultimate recipient will deposit those funds back into the system. The recipient and the bank will negotiate the type of deposit, but a demand deposit form is the likely default in most cases. The state must provide the reserves that it has specified must be held according to 100 per cent reserve rule. To do otherwise would be to contradict other credit directed regulations that allow banks to lend under the new system. In short, there is no fundamental difference in lending and associated endogenous monetary effects in this new Chicago system. There is only a difference in the regulatory detail that specifies permissible forms of lending, the size of the required reserves, and the flexibility that the full reserve requirement provides to the state in the control of money supply. The knock-on endogenous money creation dynamic is something that regulation does not preclude, since it is the result of lending that falls within regulations, assumed institutional configurations, and the inevitable accounting entries that follow along in the process of money movement.
Macroeconomic Interest Rate Risk Management
The BK paper:
“The zero lower bound on the policy interest rate in the present monetary environment binds because at a less than zero nominal interest rate on bonds private agents can switch their investment to cash. This is not a relevant consideration for treasury credit, which is only accessible to banks for the specific purpose of funding investment loans. The policy rate is therefore not constrained by a zero lower bound.” (Page 38)
This is unclear. It amounts to the claim that there can be no arbitrage between (potentially) negative interest rates paid on state funding of banks and zero interest rates paid on deposits. But we’ve already demonstrated that the state must issue debt (or some non-monetary liability) in order to control money supply. And those non-banks who buy CTRB issued bonds must be attracted away from endogenously created deposits. And those who hold those deposits also presumably have the option of converting to zero yielding banknotes. (The BK paper assumes away the importance of banknotes; yet banknotes are a critical factor in the interpretation of zero bound monetary behavior). Therefore, bonds must at least yield a zero rate of interest in order to attract those money holders to switch out of their deposits. And if the state provides negative interest funding to banks, while issuing bonds that pay a non-negative rate of interest, it will incur a negative spread on that account. Thus, the nominal interest rate on state funding provided to banks can only pierce the lower bound and become negative through an effective state subsidization of what otherwise would be a zero rate when determined by market forces.
A “liquidity trap”, if it exists, manifests itself in the failure of deposit holders to use their money to create aggregate demand for economic output, effectively by failing to activate a higher velocity of deposit money flows directed toward such transactions. And if that happens, investment projects will lay fallow until aggregate demand materializes. In short, there is an extricable circuit of influence between the various categories of financial assets in the Chicago Plan, even though there is a bold attempt to restrict financial asset accumulation opportunities for households and businesses. But one cannot prevent the natural forces of interest rate arbitrage from pursuing their own effect – unless every last financial asset in the system is priced (and effectively issued) by the state – in other words unless monetary capitalism is entirely purged, in favor of a fully state controlled system.
The Questionable Notion of Bank Reserves as “Government Equity”
The BK paper establishes a wide conceptual framework in which bank reserves are viewed as “government equity”. A very different interpretation is conceivable, which may be more consistent with a post Keynesian framework for analysis – although I don’t know how credentialed post Keynesian economists might think about this area. In any event, what follows is my own analysis:
The Chicago plan uses bank reserves to fund the state’s purchase of the banking system’s legacy loan portfolio. As explained, this is an arrangement in which an initial infusion of reserves ultimately hedges the liquidity risk on legacy deposits. The state transfers the private sector loan portfolio onto its own books in return for cancelling the banks’ loan liability created by the original reserve injection. What remains is a reserve asset on the books of the banks, funded by deposits, and the legacy loan portfolio on the books of the state, funded by reserves.
We note again that the use of the term “funded” is appropriate here. “Funded” means a balance sheet construction for asset-liability management purposes whereby a specific category of assets is offset by a specific category of liabilities and/or equity funding. In this case, it is perfectly rational to say that the state funds the legacy loan portfolio with bank reserves that it issues, because that is the asset-liability configuration following the full sequence of transactions implemented by the Chicago Plan.
(This exclusive reserve funding arrangement pertains to the legacy loan portfolio only. From that point on, the state provides funding for new bank lending and funds its own balance sheet requirements in that regard with a mix of reserves and debt, depending on money supply control parameters, as described earlier. But the issue of “government equity” is most dominant with respect to the permanent reserve funding of the legacy loan portfolio, due to its massive size, and especially in the context of the subsequent debt jubilee adjustment and its effect on the equity position of the state balance sheet.)
The BK paper:
“In this context it is critical to realize that the stock of reserves, or money, newly issued by the government is not a debt of the government. The reason is that fiat money is not redeemable, in that holders of money cannot claim repayment in something other than money. Money is therefore properly treated as government equity rather than government debt.” (Page 6)
The BK paper thus argues that “non-redeemability” is the essence of the attributed equity characteristic. More generally, it seems to suggest that what seems in structural appearance not to be debt must therefore be equity. It argues elsewhere that bank reserves issued by the state need not pay interest, and that this strengthens the argument for an equity interpretation. But even if we allow for that assumption about zero interest rates (questionable in my view), neither that nor the other factors noted above mean that bank reserves amount to equity or state equity funding in any sense. A counter argument follows.
Consider the following very simple example, which includes a bit of informal circuit analysis:
Assume a firm makes a new real investment I, where the production of I generates income ultimately paid to the household sector H. The firm uses temporary bank financing to pay for the investment, which provides monetary income to the factors of production which are attributable ultimately to H.
No consumption goods have been produced. Therefore, H saves the income – in the form of bank deposits. By construction, those deposits are the same by amount as the original loan to the firm, reflecting the flow of expenditure on I and resulting income to H.
The firm then issues securities to pay off the temporary bank loan. H purchases the securities with money drawn on its deposit accounts. This reverses the original process of loan and deposit creation. Loans and deposits are destroyed.
H ends up holding securities that are the new financing for I.
Saving S (conceptually equal to what is not spent on consumer goods – which is all income by assumption) has matched investment I.
Supposes those securities consist of 90 per cent debt and 10 per cent equity claims – i.e. bonds and common stock for example.
Government and foreign sectors are absent from this example as a matter of simplification. So the result is the same as a closed, balanced-budget economy.
Also, private sector saving is equivalent to household saving, because the model has assumed all income has been paid to H.
Therefore, S = I for this model in particular, where the saving of the private sector S is also the saving of the household sector H considered as a full subset of the private sector.
Now consider the household balance sheet, interpreted at the margin in correspondence to the transactions of the example.
The household assets are debt and equity financial claims.
Here is the critical point about the subject of “equity”:
The right hand side of the household balance sheet corresponds to the correct interpretation of “equity” in this example. It is the amount that has been saved from income during the accounting period, and it represents a net addition to household wealth. The period income and saving flow becomes the end of period stock, when considering the effect of these transactions at the margin.
A household balance sheet does not issue equity claims to agents beyond that balance sheet. This is very important to an understanding of the function of equity accounting in a stock flow consistent macro framework. One might attempt to argue that a household has a claim against its own book equity, but this is a stretch in semantics rather than a reflection of substance. The important point is that there is no equity security issued against this type of book equity, such as is often the case in the business sector of the economy.
Also, all rivers of income and saving lead to the household balance sheet (and this at the global level as well). The household sector is the mouth of the river in terms of the chain of financial claims that intermediate through the entire economy.
The only potential accounting interference to the ultimate flow of income and saving to households is the case of retained corporate earnings, where businesses rather than households are the direct recipients of income that is saved. But this element of financial wealth can be projected indirectly (optionally) to households via stock market valuation. This is the case with the Federal Reserve Flow of Funds report, where cumulative saving is reflected in a household wealth component that includes the (marked to market) stock market translation of corporate equity as it reflects both paid in capital and retained earnings.
Moreover, such equity saving booked on the balance sheets of business can be paid out optionally to households via current and/or future dividends, provided such equity endures and is available in liquid asset form for eventual distribution.
Apart from this particular retained earnings piece, the macro business sector balance sheet eventually intermediates all income and saving directly through to the household sector.
And this paradigm is stock flow consistent when adding government and foreign sectors. This is evident in the message of:
S = I + (S – I), as discussed here:
As a general proposition, that equation isolates the net financial asset wealth (S – I) that is produced by the combined effect of government deficit spending and foreign net saving and which is eventually projected onto private sector savings in stock (cumulative flow) form. Those elements of saving can be transformed and projected to household saving in full according to the government and business distribution of income along with the retained earnings adjustment just referred to above.
The term (S – I) reflects the comprehensive net financial asset (NFA) component of private sector (in this example household) saving, which when added to the component that funds investment I, sums to total saving S. This is normally a private sector quantity construct, but in this case (all income paid to households) it becomes equivalent to household saving. That household focus is the context for S = I + (S – I) in this example.
So with all that, the concept of saving and equity is inherent in the right hand side of the household balance sheet, allowing for a suitable adjustment for the projection of business retained earnings onto the measure of household income and saving, via the stock market.
The example illustrates a case in which 100 per cent of the right hand side of the household balance sheet consists of equity or net worth. This is because all expenditure was in the form of investment and therefore all income was saved. The example was considered at the margin so that the cumulative flow equalled the stock at the end of the period. The 100 per cent equity position is the household version of book equity, where no financial claims have been issued against it.
Here is the important observation:
The asset side of the household balance sheet in our example consists of financial claims which are 90 per cent debt and 10 per cent equity securities. That corresponds to a household equity position which is proportionately equivalent to 100 per cent in magnitude. Therefore, the measure of household sector equity has nothing directly to do with the role of either equity or debt financial claims appearing on either the asset or funding sides of the household balance sheet. We’ve demonstrated the appropriate distinction as far as the asset side is concerned, and such issued financial claims play no material role at all on the funding side.
This is the logical starting point for a stock flow consistent macro model in terms of the concept of equity. The idea of “equity” is larger than and separate from the identification of equity financial claims, or any other financial claims.
So we now move to the interpretation of “equity” in the context of sectors apart from households. This includes the non-household private sector (i.e. business) and the foreign and government sectors. The analysis therefore moves beyond the specific example noted above, and into the realm of a fully open and budget flexible economy.
In the case of the business sector, equity is reflected in the retained earnings account. New issues of common stock do not reflect equity directly, as they are merely asset swaps at the time of their creation – cash for stock. This is an equity financial claim, with no necessary direct relationship to a measure of cumulative saving as is the case when focusing on the right hand side of household balance sheets. To the degree that such financial stock acquisition reflects the deployment of savings already saved by households, it will be reflected as a household asset directly or indirectly but without changing the value of household equity.
However, as noted above, business retained earnings do reflect income that is saved by business – since by construction and accounting classification, such retained income cannot be spent on consumer goods. (If it were, such an expenditure would be accounted for as an expense on the business income statement, making profit that much smaller.) As noted above, it is possible to project this component of equity through to household balance sheets via the valuation of equity claims that appear directly or indirectly as household assets (the full scope of this projection also includes the book value of paid-in capital). For example, the retained earnings of a business in which a pension fund invests will be reflected first in the asset value of the stock held by the pension fund and subsequently in the value of the pension reserve that is an asset on the household balance sheet.
The next step is the treatment of the foreign sector. For example, the rest of the world (ROW) currently enjoys a net international investment position with the US. This appears as an asset on the ROW balance sheet. There is a corresponding equity position on the right hand side of that ROW balance sheet. (We interpret this in the context of the marginal position only, as the ROW balance sheet obviously includes a far larger equity position from all other sources.) That marginal equity position corresponds to the cumulative saved income attributable to the cumulative current account surplus that ROW has amassed against the US (with subsequent marked to market value). But similar to the US domestic household example, the size of that equity position has nothing directly to do with the particular asset composition of the ROW balance sheet. For example, we know that ROW owns $ trillions in US Treasury bonds. That is obviously not an equity financial claim, illustrating again that the appropriate identification and measurement of “equity” must be disassociated from any relationship in general to equity financial claims or any other types of financial claim.
Having dealt with core household, business, and foreign balance sheets, this leads to the primary discussion relevant to this section, which is the treatment of “equity” in the context of the government balance sheet – which the BK paper wants to identify as “government equity”.
If we are to have a stock flow framework that is globally consistent, the government sector must fit seamlessly within that paradigm. A high level version of the sector financial balance model partitions the world between state and non-state sectors, with the latter consisting of everything outside the state, including domestic private sector and the foreign sector. A popular interpretation for example is that the state delivers a “net financial asset” position (NFA) to the non-state sector via deficit spending.
The non-state NFA position in this construct is the marginal global wealth contributed by a specific sovereign state – which in this example we take to be the US. Thus, the global non-state here includes the US domestic private sector and the rest of the world in total. The result is a US centric measure of state “NFA production” and its contribution to the global non-state balance sheet. This “NFA asset” accumulated by the global non-US-state appears as a mixture of US bank reserves, currency, and treasury debt, matched by equity value on the right hand side of the consolidated non-state balance sheet. That change in non-state equity produced by US deficit spending will ultimately be reflected on the balance sheets of households somewhere in the world. The global household balance sheet is the “end of the line” in terms of the chain of balance sheets that constitute financial intermediation across the global economy. Again, household equity, or net worth, does not exist as a financial claim issued to some further agent, unless one views the head of the household as having an equity claim on his/her own balance sheet, which is unproductively abstract.
Such equity positions result when cumulative deficit spending creates income for the non-state sector – income saved by necessity at the macro sector level, since it cannot be used to acquire the same goods and services already purchased by the state. This non-state NFA/equity position (at the margin) is NOT a function of the particular liability mix issued by the state in conjunction with its deficit spending. Therefore, it is not affected by the proportion of bank reserves contained in that liability mix. Government debt contributes as much to the non-state equity effect as bank reserves. This is the only stock/flow consistent approach to accounting for the equity effects of state liabilities that are created in the process of deficit spending.
For this reason, the BK paper characterization of bank reserves as “government equity” is incongruent with a stock/flow consistent accounting framework for the monetary system as a whole.
We now explore more of the detail of deficit spending as an example of the equity characteristic. At this point, it is appropriate to refer directly to “Modern Monetary Theory” (MMT). It should be well known by now that MMT fashions the impact of state deficit spending in terms of its delivery of “net financial assets” or NFA to the non-state sector. The concept of non-state NFA is a particular construction in the more general category of “Sector Financial Balance” models used by MMT and in post Keynesian types of analyses of the flow of funds and the role of government deficits. This all seems consistent with how the concept of equity should be interpreted. The NFA effect is a marginal equity infusion to the non-state balance sheet. As described above, there is no direct functional relationship in general between the idea of equity and the relative incidence of debt and equity financial claims in the financial and monetary system. And there is certainly no correspondence in the case of NFA. Whether the NFA component of non-state equity is manifested entirely in Treasury bonds or entirely in currency or entirely in bank reserves makes no difference whatsoever to the conceptual quality of non-state NFA as equity. Therefore, there cannot be any unique logical connection between reserves as a non-state asset and the presence of non-state equity.
The NFA/equity contribution from deficit spending is only one component of total non-state equity, as explained in the S = I + (S – I) discussions. When one extends the observation about equity as it relates to NFA to the totality of either private sector or non-state saving, it becomes even more obvious that there is no necessary correspondence between non-state balance sheet equity and the equity financial claims that exist on non-state balance sheets as a result of financial intermediation in total.
Thus, the composition of both central banking and treasury liabilities, whether captured in actual bifurcated form or optional CTRB institutional fusion form, is irrelevant to the equity characteristic of the state balance sheet. This establishes why the BK concept of “government equity” is arguably wrong, in the broad context of post Keynesian type stock flow consisting financial analysis.
Yet, a greater irony remains.
And that is that the treatment of bank reserves as “government equity” is a case where, not only has an equity characteristic been misattributed to reserves, but the correct attribution of the equity characteristic is in fact directionally opposite to what the BK Chicago Plan describes.
This directional error in equity characteristic can be illustrated by an example from a wider scope than the banking treatment contained in the Chicago Plan – the (extreme) case of a cumulative government budget surplus:
Consider the unusual situation of a cumulative budget surplus spanning all prior accounting periods (I believe this has occurred once in US history). This is a situation in which the standard stock NFA profiles of state and non-state are reversed. The state has accumulated a positive NFA position on its balance sheet, as a result of a cumulative budget surplus over time.
Thus, the left hand side of the state balance sheet must include a portfolio of assets that represents the deployment of cumulative surpluses. This asset portfolio must consist of either financial claims on the non-state sector, or direct real assets acquired from it.
The right hand side of the state balance sheet consists of state equity. This latter position is not a specific financial claim issued to the non-state sector. It is a book equity position reflecting cumulative budget surpluses – i.e. taxes in excess of spending. Such a book equity position is analogous to corporate balance sheet book equity that accumulates through retained earnings. Retained earnings are undistributed profits that result from an excess of corporate revenues over expenses. In the same way, taxes represent an excess of state revenues over expenses.
The accumulation of state equity through surplus taxes must result in some corresponding asset portfolio, as described above. Suppose this asset portfolio is composed of financial claims issued by the non-state (although in theory it could be real assets acquired by the state). Such financial claims represent a marginal reduction in the NFA position of the non-state. The asset portfolio directly reflects the equal and opposite negative NFA position of the non-state – i.e. with budget surpluses, the non-state holds an (NFA) position, reverse to the standard orientation in the case of budget deficits.
It is also the case that in addition to that core NFA profile, bank reserves and currency will continue to exist as state funding liabilities, reflecting a basic liquidity service provided by the state. Again, the CTRB construct comes in handy in visualizing this balance sheet configuration.
Consider these two different pieces of the CTRB balance sheet – an NFA/ equity section and a reserve/currency funding liability section. The NFA/equity section consists of a portfolio of claims on the private sector funded by true government equity. This is a book equity position that has been created by a cumulative budget surplus – i.e. by a cumulative tax surplus. The reserve/currency section must include a parallel portfolio of similar claims, in that CTRB asset acquisition is the prerequisite transaction in order to create such funding liabilities. However, unlike the equity section, this reserve/currency funding pieces has a neutral NFA effect – because financial assets acquired are offset by financial liabilities issued.
The acquisition of financial claims that create the desired reserve and currency funding liabilities may be viewed as analogous to the Chicago Plan second stage transaction noted earlier, in which the state takes the banking system legacy loan portfolio onto its books. Both transactions are essentially assets swaps that have no initial effect on sector NFA distribution.
However, in the case of the Chicago Plan, the subsequent debt jubilee transaction has a profound effect on sector NFA distribution. The plan uses bank reserves to purchase the legacy loan portfolio from the banking system. The question becomes – what is the equity effect of this transaction?
There are two answers, depending on the treatment of the legacy loan portfolio after it lands on state books.
If the legacy portfolio is retained as a performing asset, with the private sector continuing to service its liability, there would be no equity or NFA effect from this transaction. The principal value of an equal value asset swap creates no equity effect. (This is the basis for the more general argument that fiscal policy is superior to quantitative easing in stimulating additional aggregate demand in the current environment, for example.)
However, the debt jubilee adjustment actually prescribed by the Chicago Plan creates a very different result. If the state takes the legacy loan portfolio onto its balance sheet, and writes down the value of that portfolio to zero (which it does in effect in a debt jubilee), then the result is the immediate creation of negative state equity (and positive non-state equity) in the same amount.
In our template example, the BK paper would interpret the creation of $ 10 trillion in bank reserves as “government equity”. But the stock flow consistent treatment is a negative equity position of $ (10) trillion, and the role of reserves as the liability of choice has no bearing on this result. In fact, as noted earlier, the BK paper itself uses a much larger order of magnitude for such a reserve injection – two times GDP or $ 32 trillion. And there, the corresponding measure of government equity is $ (32) trillion.
The usual intuitive response to the notion of negative equity is the perception of financial weakness. And this is perfectly rational in the case of the private sector and the non-state balance sheet more generally. Private sector entities require positive equity as a confirmation of balance sheet strength and capital adequacy. This is true for banks, non-bank corporations, and households. Banking is the standard example of equity capital as insurance against risk – the capacity to absorb unexpected losses. And this is no less true for other private sector entities, including households.
By contrast, not only does the state not require positive equity, but positive equity as a balance sheet objective is arguably inappropriate. This reflects a number of factors. First, a sovereign fiat currency issuer cannot go involuntarily bankrupt, because of the capacity of the central bank to create money as required. Second, the state has the equally powerful capacity of taxation, which can be used to adjust the equity profile as desired. Therefore, there is no rationale for positive stock equity or a set measure for positive state equity that is required in order to protect the state’s financial position. Third, the state is in the unique position of issuing financial claims (reserves, banknotes, debt) that are effectively risk free in terms of involuntary default. And such risk free claims are in demand. The non-state sector will seek to acquire such risk free claims as a risk adjustment element in portfolios whose asset mix is otherwise risky. This demand may intensify during periods of economic or financial turmoil, as has been the case during the financial crisis. Because of these factors, there is in general a natural demand for state issued liabilities of all forms, and in a growing economy there is a natural demand for trend growth in the supply of such liabilities. Negative state equity is a natural balance sheet profile, as is stable trend growth in that profile, reflecting the direct transfer of positive equity value from the state to the non-state sector, as a feature of macroeconomic risk management. Because of this, a growing economy should incorporate trend growth in the stock of negative state equity, or state (NFA), and this should be interpreted as a sign of strength rather than weakness. It reflects the capacity of the state to absorb risk and to transfer surplus risk absorption capacity to the non-state sector. Negative state equity is a source of marginal capital infusion for the non-state sector, allowing the non-state sector to take on more risk in its own economic activity. This infusion is marginal to the equity position that is created by real investment growth, as expressed in the equation S = I + (S – I), where S is the flow of total private sector equity accumulation. This obviously doesn’t mean unlimited negative equity for the government. But it does mean that negative state equity and trend growth in negative state equity is the financial system norm, rather than a departure from the norm. The legitimate issue of focus is not the negative value reading of the appropriate state equity position, but the optimal size for such a negative equity position at a given point in time – just as the issue for the private sector is to be neither undercapitalized nor overcapitalized relative to its optimal positive equity capital position. Negative state equity is an extension of capital guidelines from the banking system to the state balance sheet. The first guideline is not to hoard an excessive amount of the economy’s overall risk absorption capacity directly on state books.
Some final housekeeping items on the BK paper’s references to “government equity”:
“In terms of stock items, the government and the private sector share the benefits of replacing debt-based private money with debt-free government-issued money. The government swaps debt for equity, because the large new stock of irredeemable government-issued money represents government equity rather than government debt. This is not traded equity that pays dividends, but rather equity in the commonwealth, whose return comes in the form of lower government interest costs, lower distortionary taxes, and lower financial market monitoring costs. At the same time, treasury credit to banks dramatically reduces the government’s overall net debt. The non-bank private sector also swaps debt for equity, in this case due to government transfers that increase private sector wealth and reduce debt levels.” (Page 41)
The idea of “equity in the commonwealth” is somewhat vague. In the context of the Chicago Plan, the non-state equity position improves dramatically in the wake of its windfall debt jubilee. The ability to set and hold interest rates at zero is debatable, and the rate of interest on state liabilities has little to do with accurate equity characterization.
The BK paper:
“In this context it is critical to realize that the stock of reserves, or money, newly issued by the government is not a debt of the government. The reason is that fiat money is not redeemable, in that holders of money cannot claim repayment in something other than money.” (Page 6)
The notion that reserves are non-redeemable is not a persuasive argument in the context of stock/flow consistent equity recognition. Also, reserves are redeemable in exchange for the elimination of tax liabilities. They have option value as a tax asset or credit. And they are a contingent liability of the state, since they must be redeemed in exchange for the elimination of a non-state tax liability.
In summary, the measure of equity on balance sheets is independent of the mix of debt and equity financial claims. The state balance sheet equity position in particular has nothing to do with the composition of financial claims that it issues and is unrelated to the quantity of reserves in that mix.
Bank Reserves (yet again) – The Wild Card of Monetary Design
Bank reserves can be viewed from two perspectives – the banking system that holds them as assets, and the central bank (or CTRB) that issues them as liabilities.
As noted earlier, the primary purpose of reserves from the commercial bank perspective is as a medium of exchange in making payments through the interbank clearing and settlement system. The central bank of course has an interest in supplying sufficient reserves in order to accommodate this user objective. And in doing so, the central bank seeks to achieve its own policy interest rate objective.
The central bank is the issuer of reserves. The language of “liability” is occasionally contentious when applied to reserves from the issuer’s perspective. This semantic issue should be set aside in favor of more meaningful discussion. It is a harmless use of the word within a consistent balance sheet classification scheme. Nevertheless, the central bank’s “liability” in this case can be interpreted as the obligation to allow banks with reserve credit balances to use them in paying for net disbursements through the interbank clearing system. Liability also exists as a contingent commitment to provide central bank notes in exchange for reserves on demand. And it exists as an obligation to swap reserves in exchange for the elimination of a tax liability (equivalent to the taxpayer’s purchase of a tax asset). Thus, the currency doesn’t necessarily need to be “convertible” into something beyond this in order for the idea of liability to assume a reasonable meaning in discussions of monetary matters.
It is instructive to examine the role of bank reserves from the issuer’s perspective in four different structural environments where monetary design is altered in some significant way. Two of these include the existing monetary system as the anchor. The other two, including the Chicago Plan, assume more significant institutional change.
These monetary environments are:
b) Quantitative easing
c) (MMT) bond elimination
d) The Chicago Plan
In the case of a normally functioning fiat monetary system, reserve requirements meet the core user purpose as stated above. They pose no “ex ante” constraint to credit creation. This is because central banks always supply reserves needed to cover an individual bank’s “ex post” shortfall in securing its reserve requirement, provided that bank is still viewed as solvent by the monetary authorities. Reserve requirements are a form of self-insurance against inefficiencies in the interbank clearing and settlement process, such that individual banks can cope with unexpected swings in their reserve positions without being forced to go to “the window” of central bank last resort lending (LLR). Improvements in the technology of clearing have reduced the benefits of such a self-insurance constraint. This is why Canada has been able to progress to a system where the reserve requirement is zero. The issuer’s perspective in this mode of reserve management is simply to supply the level of reserves necessary to ensure smooth operation of the banking system while maintaining interest rate levels according to policy target (e.g. Fed funds).
The second approach of “quantitative easing” has been a fundamental characteristic of central banking policy since the start of the financial crisis in 2008. The Bernanke Fed has undertaken two different modes of extraordinary easing since the financial crisis began – credit easing and quantitative easing. The Fed acquired various forms of financial assets issued by the private sector in its early credit easing mode. It then proceeded to “standard” quantitative easing, with the acquisition of Treasury bonds. The various programs of purchasing MBS securities and related agency debt have typically been viewed as a category of quantitative easing, although a credit easing objective is evident in terms of the intended effect on mortgage market pricing.
The acquisition of all of these assets has resulted in the corresponding creation of bank reserves. These Fed asset portfolios can be viewed as “self-funded”, since the act of asset acquisition creates the reserves that are used as funding. This is an example of the general idea that “loans create deposits”, applied to central banking. But from an ongoing balance sheet perspective, reserves in place can be interpreted as funding these asset portfolios.
It was evident early on in both modes of Fed easing that bank reserve creation was more important to the central bank in its pursuit of unconventional monetary policy objectives than it was to the banks in terms of their demand for reserves. It might be interpreted otherwise if the Fed had been acquiring assets originally held in bank portfolios. Then one could argue that the banks chose to hold reserves rather than the assets they had sold, once having agreed to the transaction price. But this was not the case, because most of the assets acquired by the Fed were issued by or sold from portfolios of non-bank agents who only used the banks and bank dealers as the conduit for the transaction. The end result was that both reserves and bank deposit liability balances were created as a product of these credit and quantitative easing transactions.
Because the Fed injects reserves into the banking system as part of all such transactions, the banks in effect have no direct say in the creation of most of this reserve effect. That suggests that the banks’ “demand” for reserves should be interpreted with a grain of salt. The Fed’s creation of reserves was more accurately a by-product of its unconventional monetary policy objectives. In the case of credit easing, it sought to transfer credit risk from the private sector to the central bank balance sheet. In the case of quantitative easing, it aimed at producing yield effects in the markets where it acquired bonds. The efficacy of such programs has been debated of course. But that question is not the purpose of this essay. The point here is that these Fed actions corresponded to a use of reserves that for the most part did not reflect the conventional purpose of providing liquidity to the banks for interbank payments and settlements. That latter stage was surpassed long ago in the Fed’s journey from an excess issued reserve position of several billion dollars to one that is now on its way toward the $ 2 trillion mark. Reserves are now being used as a source of funding for the Fed’s asset portfolio expansion, and are conveniently created by the act of asset acquisition itself.
The third case of bond elimination reflects the specific proposal of several principal MMT adherents. For example, Warren Mosler’s proposal is here (along with several others):
The essence of this idea is that Treasury no longer issues debt in order to finance its deficit. Instead, deficit spending (spending net of taxes) results in net credits to bank deposits, mirrored by an expansion of bank reserve balances held at the Fed. The end result of such a methodology is that, in the context of the existing bifurcated institutional arrangement, the Fed balance sheet accumulates a massive reserve funding position as a mirror image of cumulative deficit spending. The banking system balance sheet expands by the amount of reserve assets and matching banking system deposits.
The Mosler plan also contemplates the elimination of the interbank trading market. This essentially places the Fed in the role of interbank dealer. It is likely that the importance of balance sheet risk adjustments now conducted through direct interbank trading would decline as a proportion of the overall banking system balance sheet. The interbank market sums to zero on consolidation of all interbank assets and liabilities (in a closed system context). Therefore, this proposed change in itself would not necessarily have a direct effect on the endogenous money process at the macro level. However, the Mosler plan also proposes unlimited Fed funding of bank balance sheets, as a policy that deliberately deemphasizes the importance of liability management “discipline” in banking. It is difficult to envisage the total effect of this on the system balance sheet. Intra-system redistribution of reserve balances is not so much the issue, as individual banks needing reserves would go to the Fed more often to get them. The marginal effect is that of an increase in system reserves, so that this in itself does not affect the endogenous money distribution. However, depending on the extent of this activity, there could be an effect on the fate of endogenous money if the Fed and or Treasury began to issue non-monetary liabilities in order to drain reserves and deposits from the banking system.
(We suggested earlier that Chicago Plan would require a similar mechanism to control money supply growth. In both cases, the state must take the initiative to “destroy” and replace endogenous money that has been created and otherwise trapped within the banking system.)
The final case is that of the BK Chicago Plan. The Chicago Plan contrasts interestingly with the first three cases of monetary design.
For example, the Federal Reserve used credit easing policies to transfer credit risk from the private sector to the Fed’s balance sheet, as an assist to markets that had seized up in the early stages of the financial crisis. The Chicago Plan is a dramatic extension of such credit easing. The Plan transfers the entire legacy loan portfolio of the banking system (including shadow banks in the BK paper) onto the books of the state (CTRB or Treasury/Fed). Massive reserves are created in the process. This huge amount of credit risk is extracted from the private banking system as a matter of institutional reform. The banking sector holds risk free reserves as replacement assets, instead of the previous risky loan portfolios.
The BK Chicago Plan further specifies that these credit assets are written down to a value of zero as a policy of debt jubilee. This has a significant meaning in fiscal policy terms. The write-down itself is a one-time fiscal deficit and a corresponding expansion of negative state equity and (NFA), as defined in the previous section. It is an extreme example of the idea that the state can absorb risk in great quantities, while transferring the same level of risk protection to the non-state balance sheet as an equity or NFA capital infusion. In this case, the effect of the write down is to improve dramatically the financial position of the non-state balance sheet.
However, the state loses all future income from those assets. And it funds its position with reserves. If the CTRB/Fed ever needs to pay a positive interest rate on reserves in order to tighten monetary policy, interest margins will become negative and add to future deficits. This is a contingency that the BK paper downplays and seems to dismiss outright. The question of zero rate sustainability is an interesting one, and one that is also contemplated in the Mosler plan as a matter of intended monetary and fiscal policy integration, with taxation as the primary lever of policy adjustment.
The common theme across all of these unconventional approaches to monetary and fiscal policy – quantitative easing, the elimination of bonds, and the Chicago Plan – is that the state is using reserves as a source of funding for the state balance sheet rather than as a supply of liquidity needed for interbank payment and settlement. The Chicago Plan does provide liquidity, but the primary purpose is to ensure that the value of deposits is unencumbered by credit or liquidity risk in relation to corresponding asset cash flows. Thus, the purpose of the Chicago Plan funding configuration, as the others, is balance sheet “risk transformation”. This is very different from the use of reserves in conventional central bank operations.
In summary, the credit easing component in the Fed’s original response to the financial crisis is somewhat analogous to the initial stage in the implementation of the Chicago plan. But the Chicago plan proceeds to a significant fiscal effect via the debt jubilee component, with a consequential boost to current and potential future deficits. As is the case with all deficits, the issue becomes the risk of future interest costs, which in this case would be payable on a massive system reserve position. Such costs might be ameliorated by a policy of permanently zero interest rates, but this is a contentious idea to say the least.
Final Comments on Risk Management
The core risks in banking are credit, liquidity and interest rate risk. The Chicago Plan includes hedging mechanisms for these risk types. But the overarching theme of the Chicago approach is the relationship between the state and the non-state in the proposed monetary architecture.
Bank reserves are a defining element in that relationship. The theme of any of the approaches listed in the previous section is that bank reserves can be used in a process of massive risk transformation – as opposed to their normal core role in clearing and settlement of interbank payments (and providing banknotes on demand).
The core purpose of reserves is to facilitate competitive operations in a banking system that is intended to be an integral component of capitalism – not to distort that competition by interposing huge quantities of forced state financial intermediation. This is why the Federal Reserve is focused on an eventual “exit plan” in respect of its unconventional operations during the financial crisis.
More permanent policies that use reserve funding in such extraordinary fashion are consistent with slow or fast nationalization of the banking system. The flirtation of the Chicago Plan with a deeper nationalization motivation is highlighted by Frances Coppola here:
“They seem to have completely forgotten about the profit & loss account and, indeed, the purpose of private banks. The sole aim of private banks is to make money, principally through lending. And the sole point of taking deposits is to fund lending. If deposits couldn’t be used to fund lending, why on earth would banks want them?”
Of course, the answer to Frances’ question is that banks would not want them if they had that choice. But they would take them as a result of government imposition and prescription in the Chicago Plan. A similar explanation applies in the case of the MMT plan to eliminate bond financing, which feeds deficit spending directly into the expansion of bank balance sheets by using the operational capacity of the reserve system. The banks wouldn’t want that structure either – they’d be forced by the state to take it. And if the state offers the more “palatable” version of short dated Treasury bill sales instead of forced reserve injection, that would be a minor improvement, since a lack of natural enthusiasm for mass offerings of concentrated short date securities might well result in residual funds trapped by default as bank deposits and reserves instead.
The intended purpose of bank reserves is to facilitate banking competition. Reserves provide the interbank liquidity that allows independent, competitive asset-liability management processes and risk taking across the banking system. Commercial banks partition their asset and liability portfolios by risk measurement and management, according to core dimensions of credit, liquidity, interest rate, and other risk types. Within this approach, for example, they partition the liquidity and interest rate risk characteristics of deposit funding, using scenario modelling for potential liquidity and interest rate behaviors.
The pivotal position for the Mosler/MMT approach to monetary operations may well be the assumption of zero interest rates. The rationale for the elimination of bonds seems to depend on this assumption. The logic of eliminating bond financing is driven by the observation that governments don’t need to issue debt to replace currency (i.e. reserves) they otherwise issue. This is an issuer’s liquidity argument, in essence. It overlooks the dimension of interest rate risk, but can afford to do so if there is an accompanying assumption of permanently zero rates of interest. But this liquidity argument for the elimination of bonds ceases to be adequate if one allows for the possibility of non-zero interest rates – or more generally, if one allows for the likelihood of a central bank that pursues an active interest rate policy in monetary operations. If this is the case, then the state has a vested interest in pursuing policies of prudent interest rate risk management.
The Mosler/MMT plan contemplates a banking system liability position that can absorb the cumulative deficit – currently $ 16 trillion. The Chicago Plan similarly allows for a banking system liability position that can absorb, not the deficit, but an asset liability position consisting of $ 32 trillion in reserves and corresponding deposits – where reserves presumably pay no interest. The issue in both cases is what happens to the cost of reserves in either case if the central bank does need to pursue a policy of active interest rate management. Both plans assume explicitly or implicitly that there is no risk of future monetary policy regime change or strategy change. If that is wrong, and there is such a risk, and if the CTRB does in fact choose to increase interest rates at some point, then the fiscal cost of interest paid on reserves held by the banking system is subject to what is described as interest rate “gap” risk in banking – in this case on the position of the CTRB as a bank – on a rather colossal scale. The risk in this case is that the CTRB will have to increase rates paid on $ 48 trillion ($ 32 trillion Chicago; $ 16 trillion MMT) in conjunction with an increase in policy interest rates.
This interest rate gap risk is technically separate from short funding risk in the liquidity sense. Such risks are often combined in a single liability form – but not always. A tendency exists to conflate the two in more general discussions about banking risk. “Gap risk” is the one that came much to the fore in the banking environment of the 70’s and the 80’s, due to the severity of monetary policy tightening.
There may be a tendency to downplay the aspect of interest rate risk on the basis that the current interest rate environment is benign in the extreme, and that the state balance sheet can absorb such risk in any event, along the lines of our earlier discussion. But these factors are not an endorsement for indifferent risk management techniques by the state. So the question regarding bond financing is not merely one of a monopoly currency issuer position or the fact that a sovereign with full control over this capacity can’t go bankrupt involuntarily. It becomes a question of sensible risk management in total. It would be reckless for a government treasury function or a CTRB to run a liability structure whose interest rate sensitivity was exclusively short term. In the (likely) future absence of a permanently zero interest rate policy, the bond vigilantes will return some day, and when they do, it will be because they are front running the Fed with their own expectations for future interest rate policy. It will have nothing to do with bond trader debates about solvency or insolvency. The non-issue of government solvency is not terribly relevant to the issue of the bond vigilantes. But reasonable interest rate risk management is. It is a fairly obvious point that central banks can create reserves for the purpose of deficit financing, one way or another. The idea of a CTRB institutional structure was constructed as a clean demonstration of this capability. The CTRB put option on reserve liabilities can always be exercised operationally if private sector demand for government securities dries up. This includes the hypothetical case (very extreme) in which Treasury dealers exercise their own form of put option – to revoke their dealerships in the event of unacceptable interest rate risk bearing on their capital positions. In such an environment, the regular LLR function extends to an ILR option (issuer of last resort). This is the option that fundamentally distinguishes the nature of the US monetary architecture from that of Greece and the rest of the Eurozone.
The argument that the government “can’t go bankrupt” or is “not like a household” is not the most relevant issue. It is a somewhat trivial observation in itself. Anybody who doesn’t know that central banks can swap reserves for debt isn’t paying close attention. The central bank is the operational currency issuer. A government that controls policy for its central bank (yes for the Fed; no for the ECB) “can’t go bankrupt” – at least not involuntarily. And this particular observation has no consequence in itself for the question of how big deficits or debt should be. That question is answered by other factors. Therefore, while the fact that the government “can’t go bankrupt” correctly dismisses one non-constraint on deficit financing, it says nothing about how big the deficit should be. So it is insufficient in providing an actual answer to that question and it is insufficient in any argument about a general strategy for deficit financing. (E.g. MMT considers inflation risk in responding to this question.)
The Chicago Plan and Mosler/MMT both prescribe massive reserve funding of the negative state equity position. The difference is that the Chicago Plan focuses on the negative equity that has been created by the debt jubilee. The MMT plan focuses on the more typical balance sheet component created by deficit spending. Those two pieces are complementary and additive, in the sense that the result of a Chicago/MMT Plan fusion in that particular dimension would be a $ 48 trillion CTRB negative equity position funded by bank reserves. (The Mosler plan has other ideas for banking system reform.) Again, the management of interest rate risk on that position would be of importance in the event that monetary policy resumed with an actively managed non-zero interest rate approach, although neither plan contemplates this.
One is left wondering if the ultimate end game of such proposals may be a hyper-dense ‘banking singularity’ – a single CTRB institution that absorbs private sector banking in its entirety, with banking competition eliminated, CTRB reserve funding converted to demand deposits held by the public, and the full absorption of lending under state auspices – until nothing else remains that resembles private sector banking or finance.
A Few Links: