Bank Lending is Profit Constrained not Equity (Capital) Constrained – A Small but Important MMT Error

MMT is almost but not quite correct.  Their sectoral balance view of the economy is correct but one area they continually stumble over is the treatment of time.

One such area where a mistake is made is over the ability of banks to lend.  Lets take a much repeated quote from the more vulgar end of MMT (Billy Mitchell et. al.) rather than the theoretically pioneering end (Wray, Fullweiller etc.)

So lets repeat: bank lending is capital constrained not reserve constrained. Fundamental point that comes out of an understanding of how the monetary system works.

Nearly but not quite true.  The fundamentally true point is that under modern monetary conditions bank reserves do not act as quantity constraint on credit growth, the reserve constraint view is simply the old ‘loanable funds’ view of credit capacity in new clothes.  It has endlessly been debunked by modern monetary thinkers of all persuasions and if their is a canonical statement  of the modern view it is this from Scott Fullweiller, which arose from the recent Keen/Krugman debate on endogenous money.  If then we have debunked the view that the constraint on bank lending is a not a quantity constraint but depends of future profits – a price constraint not a quantity constraint, then why do we again say that the constraint is a quantity constraint – but a different one equity not reserves?

At the root of this is a number of theoretical confusions:

1) That Schumpter’s rhetorical statement that bank credit is ‘created out of thin air’ means it can be created without limit, no it cannot it is constrained.

2) That because banks can create credit with zero reserves reserves play no part at all in the ability of banks to lend – they do though it is an indirect secondary effect

3) That it makes no difference if one bank creates money endogenously or if all banks do so at the same time – it does though the neoclassical view of this is flawed.

4) A confusion of stocks and flows in terms of anticipated bank revenue from loans.

5) A confusion in causation between profits and bank equity prices.

Ill set out the issues under 3) more fully on some future occasion as it depends on a treatment of the full monetary circuit and the relationship between horizontal and vertical money which is beyond the scope of this article.  Here I want to stick to one simple point – Q why do banks lend money A Because they find it profitable.  Q so then if banks find it more profitable to lend they will lend more A correct, and if they find it less profitable to lend they will lend less A correct.  Q So the anticipated changes in bank profitability from loans affects the changes in net bank credit creation A QED.

Dealing first with the much misused ‘out of thin air’ point.  I shall take as my witness here banker turned Journalist John Carney

The idea that banks are not constrained by reserve levels—because they can always borrow required reserves should they fall short—shouldn’t leave anyone with the impression that banks do not face serious constraints on their lending.

The biggest constraint on lending is the basic business model of banking.

When I was a banking lawyer, we usually referred to this as “cost plus lending.” The idea was that the bank’s source of profit was charging more for loans than it cost the bank to make the loan.

This sounds pretty simple until you start thinking about the source of the costs to make a loan. The first cost, of course, is what’s known as “the cost of funding”—the amount the bank must pay to borrow the reserves required to make loans.

But there are a host of non-funding costs as well. The bank must also somehow acquire the regulatory capital to back the loan, and capital can be expensive. The bank faces administrative costs in making the loan. Bankers must be paid. There may also be various taxes and government fees that apply.

Banks also face interest rate risk that they attempt to hedge by either borrowing funds at durations that match the term of the loan they are making—which raises the bank’s cost of funding—or making floating rate loans. But with very large loans, banks will often require floating rate borrowers to hedge against interest rate risk themselves—since you don’t want your borrower defaulting just because rates have increased. The price of interest rate hedging also has to be figured into the borrowers ability to pay the loan.

These aren’t trivial costs. The best estimate is that these add up to almost 300 basis points—the spread between the Fed Funds rate and the Prime Rate.

This means that even if banks can make loans out of thin air, there are plenty of loans that a bank cannot make. Or, to put it differently, if a bank makes too many of these loans it will lose money and eventually fail. A person or business whose free cash flow is too light to support the costs of the loan, for example, is not credit worthy—that is, not eligible to receive a loan under most circumstances.

This is one way that monetary policy works to encourage more lending. When monetary policy brings down the price of banks borrowing reserves—that is, reduces the Feds Funds rate—it brings down the cost of making loans. This means some borrowers will be credit worthy at lower rates that wouldn’t make the cut at higher rates.

The flip side of this is that potential borrowers must also see opportunities to put bank loans to work. If they view potential for future profit as weak, the demand for new loans shrinks. Of course, monetary policy plays a role here too. Low rates can make otherwise uneconomic projects work and, as the Austrians put it, may send a signal that there is more saving in the economy available for future spending….

Banks also must price in the “credit risk” of making a loan—the risk that the borrowers may default, either because the collateral for the loan becomes so undervalued that the borrower “walks away” or because cash flow fall short of what is needed to make the loan.

Credit risk depends both on the individual character of each borrower and the general economic prospects. Rising unemployment, fall in consumer demand, the plausibility of management’s business plans all feed into credit risk.

Evidence indicates that credit risk, interest rate risk and other costs of lending exercise very strong influences over lending. The stuff the Fed tends to influence most directly—the cost of funding—can be overwhelmed by the credit risk and cost of capital….

To put it differently, banks are not free to create money willy-nilly. They are subject to restraints imposed by both the markets and regulators. But under current procedures, these restraints do not arise from a hard limit on the amount of reserves in the system. They arise from the costs of lending, which is conditioned by (a) the interest rate targeted by the Fed, (b) regulatory and market capital requirements and the market price for bank capital, (c) by back-office administrative and hedging costs of lending, and (d) the credit-worthiness and credit-hungriness of borrowers.

Or put simply the discounted to present risk adjusted cash flow from the asset created by the loan must exceed the costs in creating the liability of increased reserves.  That is a profitable loan.  And each bank must compete in a market for loans and equity for this funding against other banks.  These profits create a cash flow and it is this cash flow, and the anticipated cash flow which create the power for banks to create loans either from cash on hand or from the financial markets.

Neil Wilson has modelled in T-Accounts, commenting on Steve Keen’s models, the ‘potential loans’ from a bank – what a ‘credit licence’ the power to create loans to a certain value.

But what benefit does carrying the amount of ‘potential loans’ give us in the model? Well it helps to show how ‘hungry’ a bank is to lend. A bank with a high valuation on its credit licence has a lot of capacity to make loans, whereas one with a low valuation hasn’t. It is very likely that the first is going to be selling loans as hard as its can whereas the second is more likely to be putting its efforts into lobbying regulators to relax the cap on its lending capacity.

Steve Keen has adopted this accounting approach and it now appears in his models including his recent INET paper.

However this model shows how past profitability allows for creation of loans at any particular instant it does not show how an individual loan decision is made in terms of anticipation of future interest rates, debtor creditworthiness or bank creditworthiness, as well as how a bank might extend its ‘lending power’  in terms of its decisions to sell assets, rebalance its portfolio, extent its creditworthiness, acquire collateral etc. We require matching modifications to finance theory as well as macroeconomics.

My witness on the Second point is Scott Fullweiller; in his aforementioned piece

the business model of banking—..is to earn more on assets than is paid on liabilities, and to hold as little capital (equity) as possible (since that’s generally more expensive than assets). The most profitable way to do this is to make loans (that are paid back, obviously, so credit analysis is an important part of this) that are offset by deposits, since deposits are the cheapest liability; borrowings in money markets would be more expensive, generally. So, Bank A, if it is not able to acquire deposits is not operationally constrained in making the loan, but it will find that this loan is less profitable than if it could acquire deposits to replace the borrowings…. banks create loans without regard to the quantity of reserve balances they are holding; they obtain any reserve balances needed at the federal funds rate or roughly equal to it. Their ability to replace withdrawals with other deposits merely affects the profitability of lending, not the ability to do so…. the loanable funds model is wrong. Banks are not constrained by deposits whatsoever, but the quantity of deposits they can raise after making a loan to replace a withdrawal will affect the profitability of the loan. Again, the constraint is a price constraint, not a quantity constraint.

The third point serves a longer treatment but if every bank extends credit at once then between t0 and t1 extra currency will have needed to enter the monetary circuit because in the M-C-M’ accumulation an extra quantum of money is needed to pay for the interest/bank profit on the loan  If it isn’t created then both growth and credit creation will be price deflationary and will be hindered.  This is quite an old point raised by German Monetary thinkers like Schact before the war and in a different form by Hawtry.  Given in modern monetary conditions central banks have to be accommodative to monetary demands so it does not undermine the central thrust of the endogenous (credit) theory of money but it is an important plank of why certain narrow ‘chartelist’ views of MMT are flawed – an issue for another day.

On the fourth and fifth points the view seems to be expressed that the constraint is a quantity one the equity stock at any time.  Though capital will effect the ability of a bank to lend at start up it failed to explain why banks have sought to minimise their capital once they have a viable profits stream.  I hope I have demonstrated that the lending power of a bank is not simply the function of working capital – but at any point in continuous time is topped up by the flows of (income from loans minus flows from liabilities) : the net cash flow to the bank at that point in time.  This affects the short term constraint on lending through requiring overnight lending from banks and the cost of such lending.  Of course liabilities and income flows from principle + interest occur at different points in time so seen from the perspective of the loan the constraint is caused, over the period of the loan and risk adjusted – in terms of the profitability of the loan.  So we might say that banks are short run cash flow constrained and long run profit constrained.  Short run cash flows though will rise and fall if the decisions made in the past on loans about current conditions have lead to a greater or lesser achieved profit than was anticipated on the loan in question.

I think this approach has a number of advantages, in particular in helping understanding how banks have manipulated their appearance of credit worthiness in order to expand their lending power through the financialisation of risks on assets, something difficult to explain in a pure equity model of banks ability to extend credit.  It also helps understand some of the causal factors behind the credit accelerator/impulse. It also helps explain why banks that are profitable might not extend credit if the wider economy depresses future loan profitability and why unprofitable banks are forced to furiously de-lever.  Finally it helps explain the Minksy Ponzi type loans where short term income from loans is high even though long term profitability from the loan might be negative.

The final point is on causation, a simple one, equity prices don’t determine themselves, they are determined by the markets assessment of current and future profits and losses.