Sunday 2 November 2014

Maturity transformation is bunk.




Commercial banks borrow short and lend long and that is one of their basic activities, if not “the” basic activity performed by banks under the existing banking system. “Borrow short and lend long” is often referred to as “maturity transformation” (MT). And MT has plenty of adherents. The following are just some.
1. Mervyn King, recently retired governor of the Bank of England said in his "Bagehot to Basel" speech that "Maturity transformation brings economic benefits..” 
2. Paul Tucker (former BoE official).
3. Sir John Vickers: see his paper entitled “Some economics of bank reform” (section 3, p.5).
The alleged merit in MT is that it enables bank depositors to share in the relatively high interest that is earned from making long term loans or investments, while retaining liquidity. Plus those who supply wholesale money market loans to banks appear to gain in the same way: those wholesale money market loans are essentially just large deposits. I’ll refer to both those types of bank creditor as “depositors” henceforth.
In fact there are several flaws in that argument, as follows.

1. Shares offer liquidity.
If MT were banned, there would not be a total loss or absence of liquidity. Reasons are as follows.
If a bank were funded just by shares rather than deposits, those shares (like all shares) would have a degree of liquidity. Indeed my Penguin Dictionary of Economics starts its definition of the word liquidity thus. “The ease with which an asset can be exchanged for money. The liquidity of an asset is determined by the nature of the market on which it is traded. For example an ordinary share of a British company is liquid because there is a well-organised market on which shares can be bought and sold….”.
Having said that, MT does provide what might be called a more “precise” form liquidity than shares. That is, buying £A of shares does not guarantee you’ll get £A back when you sell. In contrast, when depositing £A at a bank, the bank promises to give you £A back at some stage.

2. Central banks can issue any amount of liquidity / money.
A second flaw in MT is thus. If commercial banks were barred from doing MT, i.e. if all depositors who wanted their money loaned on had to deposit their money for an extended period (or buy shares in the relevant bank), that would reduce the liquidity or money creating activities of commercial banks. And no doubt that that would reduce aggregate demand and raise unemployment. But that would be very easily countered simply by having the central bank create and spend whatever amount of money is needed to bring the economy back up to full employment. Alternatively, the state could use that extra money created to cut taxes rather than raise public spending)
Moreover, central banks can create money at zero real cost. As Milton Friedman put it, in his book “A Program for Monetary Stability” (1960), “It need cost society essentially nothing in real resources to provide the individual with the current services of an additional dollar in cash balances.”

3. What if central banks didn’t exist?
A third flaw in MT is thus. Even if there were no central banks, commercial banks would be able to supply any amount of liquidity / money without extending loans at the same time, i.e. without engaging in MT. Since central banks obviously do exist,  this little section is arguably arcane, so skip it if you like.
Anyway, in a "no central bank" scenario", anyone would be able to have their bank credit their account with any amount of money they like by depositing enough collateral.
Note that at the point when the collateral is deposited and the account is credited, no loan has been extended: that is the account holder has not consumed real goods or services supplied by anyone else.
In contrast, as soon as the account holder spends £X, that means they’ve been supplied with £X of goods or services: a loan has been extended.
However, if the account holder, and similar account holders around the country were simply looking for day to day transaction money (i.e. liquidity), rather than long term loans, and say everyone had £Y credited to their account (to take a simple example) then the balance on everyone’s account would rise above £Y as often as it fell below £Y. Thus essentially no long term loans would be involved.
In short, even if there were just private banks and no central banks, we could manage just fine without any MT.

So what’s the point of MT?
Well I’m darned if I know. It does give us liquidity / money, but liquidity / money can be supplied for free in whatever quantity we need by central banks. Moreover, central banks perform that function without the risk that is inevitably involved in MT.
That is, risk is absolutely inherent to MT: if you accept £B in deposits and lend on or invest the money, the claim that you can repay the depositor £B is essentially fraudulent because loans and investments can go wrong. Indeed the recent crisis was just the umpteenth example that we’d seen throughout history of MT going wrong. And it went wrong in spectacular style recently: trillions of dollars of public money was needed to rescue banks their farcical “borrow short and lend long” confidence trick.
The above point that risk is absolutely inherent to MT was eloquently made by Douglas Diamond. In his abstract and in reference to the liquidity producing characteristics of commercial banks he says, “We show the bank has to have a fragile capital structure, subject to bank runs, in order to perform these functions.”
Now perhaps I’ve missed something, but if central banks can provide us with a form of money without risks being involved, and if private banks can only provide us with a form of money by at the same time running the risk of bank runs, I don’t see the point in letting private banks do any money / liquidity creation.
Why not have taxpayers cover the risks?
Under the existing system, the risks thrown up by MT are covered to some extent by taxpayers. But that’s a subsidy of banks, and subsidies misallocate resources, i.e. reduce GDP as it explains in the introductory economics text books. Or as the Vickers commission put it, “The risks inevitably associated with banking have to sit somewhere, and it should not be with taxpayers”.
So that’s that idea smacked down.

If MT is so useless, why do banks offer it?
Assume a totally free market, in particular an economy where governments offer no sort of lender of last resort facility or bail outs to commercial banks. That’s the sort of scenario that existed prior to the 1929 crash.
Private banks in that scenario are in competition with each other, and they are induced to make depositors all sorts of tempting offers: e.g. that depositors can earn the interest that comes from long term loans and investments, while retaining liquidity. That offer is fraudulent because the long term loans can go wrong, which is plain incompatible with offering to return £X for every £X deposited. As Martin Wolf put it "If we were not so familiar with banking we would surely regard it as fraudulent".
But the fact that some fraud is involved there doesn’t bother private banks. Indeed almost every advert issued by every firm and corporation on planet Earth is fraudulent to some extent: almost all adverts over-state the merits of the product concerned.
Now if you can make some extra money by sharp practice / fraudulent / semi fraudulent activities, then why not?

Why not just raise capital requirements?
The risks involved in MT can of course be reduced by raising bank capital requirements. But strictly speaking, risks are never TOTALLY REMOVED until banks are funded JUST BY shares: i.e. the capital ratio is 100%.
As to the idea that funding banks via shares rather than deposits raises the cost of funding banks, that idea was demolished by Modigliani and Miller.
So . . . the risks that private banks run can be totally removed by funding them entirely by shares. And the costs of doing that are zero, as Modigliani and Miller showed.
As to money creation, that can be done at zero cost by central banks. And that all equals full reserve banking  - or at least one version of full reserve banking (there are actually several variations on the full reserve idea).
MT is in check mate.

No comments:

Post a Comment

Post a comment.