Monday, 27 February 2012

Andrew Haldane’s ideas on banks and risk can be improved.

Summary. Haldane says banks profit from the risks they take at the taxpayer’s expense. Agreed. He then argues that we need to take account of this “fake” contribution to GDP made by banks, but he is vague on exactly how to do this.

I argue below that the way to do it is to drastically curtail or even ban maturity transformation. Plus the other risks that banks take at the taxpayer’s expense should be banned.  Banks’ profit and loss accounts would then show a realistic profit or loss.


Haldane argues that the rise in bank profits over recent decades is due to the increased risks banks have taken, but that those risks are underwritten by the taxpayer. Thus to that extent, bank profits are illusory: that is, they don’t actually contribute to GDP even though they are counted as part of GDP. Agreed.

And his final sentence is, “Investors, regulators and statisticians now need to adjust their measuring rods to ensure they are not blind to risk when next evaluating the return to banking.”

Well that conclusion is right as far as it goes, but Haldane does not give us much idea as to exactly HOW regulators etc ought to do their “evaluating”. That is, exactly how much does one subtract from bank profits in order to arrive at the correct figure? Indeed, when it comes to the actual figures for the implicit too big to fail subsidy, Haldane himself is vague. See his paragraph starting, “Elsewhere, we have sought to estimate…”. (Though congratulations to him for at least ATTEMPTING to estimate the size of this subsidy.)

I suggest that part of the answer to this problem lies in the fact that maturity transformation (i.e. “borrow short and lend long”) is a complete nonsense and should be banned or drastically curtailed.

And having done that, the profit figure that banks produce would then be more realistic.

But that obviously calls for an explanation as to why maturity transformation (MT) is nonsense. So here goes.

Maturity transformation.

From the perspective of a micro economic entity, MT makes sense. E.g. from my own personal perspective, money is an asset, and it pays me to make the best use I can of that asset: it would not pay me to leave large amounts of money for months on end in a zero interest current account or “checking” account if I can get a significant rate of interest on a deposit account.

However, from the perspective of a country as a whole, money is simply numbers in computers: worthless. Or as Milton Friedman put it, “It need cost society essentially nothing in real resources to provide the individual with the current services of an additional dollar in cash balances. (Ch 3 of his book, “A Program for Monetary Stability”).

Thus it makes no sense for a country to allow or encourage the banking system to economise on the stock of money the banking system holds if in doing so any sort of risk is entailed.

Thus what is required is to ban MT, or at least to drastically curtail MT -  to the point where the risks deriving from MT are next to non-existent.

Now the effect of that curtailment would obviously be deflationary. But that is easily compensated for by increasing the total stock of money, which, to repeat, can be done at no cost.

Incidentally, the Vickers commission failed to get this point. That is they were very concerned about the deflationary effect of what they called “trapped deposits”: that is money sitting in bank accounts which cannot be used due to a tightening up in the rules that govern what banks can do. To repeat, any such deflationary effects can be countered by increasing the total stock of money.

Of course curtailing MT is difficult in practice: banks and shadow banks would continually try to indulge in more MT than they were supposed to, just as they currently try to take risks at the taxpayer’s expense. But regulating banks will always be a game of cat and mouse.

Having said that the deflationary effect of drastically curtailing MT can be countered by expanding the money supply,  that is not to suggest that there’d be no NET EFFECT on the size of the banking industry. One effect of curtailing MT and expanding the money supply would be that there’d be more money in the average citizen’s bank account, which would REDUCE the need for the average citizen to resort to bank loans. Given the VAST increase in bank assets, liabilities and turnover relative to GDP over the last two decades, it is hard to see that much harm would come from reducing the overall size of the bank industry: people were not impoverished prior to 1980 through reason of an inadequately sized banking industry as far as I know.

Another change that curtailing MT would bring would be that banks would rely for funding more on shareholders at the expense of depositors. But that would not do much harm. Indeed, regulators are currently trying to get banks to do just that: rely more on shareholders.

Other risks.

MT is not the only risk that banks can take while having the taxpayer stand behind the risk. Another risk is lending risky borrowers. That is, depositors can in a sense “have their cake and eat it”: they can reap some of the benefit of their bank lending to risky borrowers, while not paying any penalty when it all goes wrong. So how do we insulate the taxpayer from this risk? Well it’s easy, and as follows.

Depositors must be made to come clean and make a choice between two sorts of account, as advocated in this submission to the Vickers commission.  First, depositors can have 100% safe and taxpayer backed accounts. The relevant money would be lodged in as safe a way as possible: perhaps it could be deposited at the central bank. Plus the money would be instant access.  But the money would not be doing anything, so it would earn little or no interest.
Second, depositors wanting the bank to invest their money could have an “investment” account (for want of a better word). Those accounts WOULD earn a significant rate of interest – reflecting the fact that the relevant money was doing something. But this is COMMERCIAL activity, and there is no obligation on taxpayers to subsidise commercial activity. Thus if the relevant bank went bust, there’d be no taxpayer backed guarantee for depositors.

Hey presto: taxpayers are almost entirely insulated from risk.


Maturity transformation should be banned or drastically curtailed.

Plus we need to abolish taxpayer backing for commercial activity in the form of money deposited at banks which is then loaned on. Then bank profits can be taken at face value.



  1. Ralph, interest rate risk in the banking book, as a measure of maturity imbalance of the balance sheet, is limited under Basel 2 at 20% of equity (that is 200bps shift to the yield curve shall not exceed in NPV terms 20% of capital). This, of course, assumes that banks calculate this risk properly but lets take it at face value. Anyways the point is that such measures were introduced following the S&L fiasco ahead of which banks were running interest rate risks of well above 100% of equity. So on this measure the maturity transformation HAS already BEEN drastically reduced. And yet this did not help us.

  2. Игры рынка, The intracicies of bank regulation, Balse included, are not my strong point. But isn’t the quick answer to your point that “that’s why we have Balse III”? I.e. Basle II was inadequate, so it is being improved.

    Having said that, you raise a good point about interest rate risk – I didn’t mention that above. That is, I dealt with quantity but not price. I.e. under a 100% ban on maturity transformation, the MATURITY and volume of loans made by a bank would need to equal the maturity and volume of its funding, PLUS the commitment by the bank to pay interest to depositors etc would have to not exceed the commitment by those borrowing from the bank to pay interest.

  3. Basel 2 ideologically introduced internal models which was an invitation for a grand systemic failure. And it happened. Basel 3 per se does not correct this issue but instead focuses on two topics from the liabilities side: capital and liquidity. I am not sure I would follow the Mosler's optimism about liabilities being a wrong place of regulatory attention but Basel 3 is quite mistaken about the source of bank problems.

    Anyways I would be curious to see studies on interest rate risk on systemic level but my gut feeling is that it is NOT outrageously high meaning that banks do NOT do lots of maturity transformation and most of the interest rate income is derived from product margins.

    But again that assumes that banks calculate the risk properly and my gut feeling tells me that they plus regulators have no clue. But still I will not be convinced about either side until I see a proper study which I am afraid will never exist.


Post a comment.