Monday, 6 June 2011

To lend or not to lend?

A couple of days ago I wrote a post showing how higher reserve requirements increased the cost of borrowing and depressed interest rates on savings.  Today, Tim Worstall came out with a similar argument in his blog.  However, he lost the plot when he started talking about the money supply and the way in which bank lending contributes to its expansion.  That's because he seems to believe that bank lending decisions are driven by the availability of reserves - this is known as the money multiplier theory.  Nothing could be further from the reality.

Any bank credit controller will tell you that availability of reserves doesn't come into the equation when lending decisions are being made.  Customer credit scoring, relationship with bank, credibility, affordability, provision of collateral or guarantees - yes, all of those are considered in depth.  But whether the bank has sufficient reserves to support that lending? Nah.

Bank lending decisions are driven ENTIRELY by commercial considerations.  The effect of higher reserves, therefore, is NOT to restrict the amount of lending that banks can do. What higher reserves do is increase the cost of borrowing, not reduce the availability of credit. (I should make it plain here that I am talking about cash reserves, not equity capital reserves - which protect customers against the possibility of loss due to insolvency. Capital reserves may also increase the cost of borrowing but not for the reasons given in this blog.)
It works like this.  Say that at close of business on Monday, Bank A has a total loan balance outstanding, including new loans, of £10m - and to keep it simple let's assume that all of that is made up of unsecured lending to individuals of average creditworthiness, so the amount "at risk" is also £10m.  Suppose on that day it also has £1m of liquid assets (such as retail deposits) in its reserve account at the Bank of England. If reserve requirements are sayy 7%, the amount it HAS to hold in that account at the end of the day is 7% of £10m, i.e. £700,000.  But it actually has £1m in the account. So it has a surplus of £300,000 in its reserve account.

Meanwhile, Bank B has also lent out £10m at 100% risk, but it only has £400,000 in its reserve account.  It should have £700,000.  So it borrows £300,000 from Bank A to cover its deficit.  This is interbank lending and it is usually done at LIBOR, which is currently about 30 basis points (that's finance speak for 0.3%) above the Bank of England's base lending rate.  The result is that both Bank A and Bank B meet but do not exceed the reserve requirement on that day.

This is, of course, a ridiculously simplistic example. But on any given day banks are supposed to maintain in their Bank of England reserve accounts an amount in liquid assets equal to the reserve requirement for the total "at risk" amount of their lending.  Any shortfall is made up by borrowing either from other banks (if funding is available) or as a "last resort" directly from the Bank of England.

So reserve allocation is done RETROSPECTIVELY. There is no limit on the amount of lending banks can do during the day.  But they must, at the end of the day, balance their books.  The more they have to borrow to support their lending, the higher the rates they will charge to their borrowers.  An increase in reserve requirements usually has to be met by borrowing more - by attracting savers or by borrowing from other banks.  Interbank lending is an exceedingly cheap form of financing and the cost can be fully passed on to borrowers, whereas attracting savers would mean increasing the interest rate paid on savings accounts - a direct, unrecoverable cost.  It's a no-brainer, really. Clearly the banks are going to borrow, aren't they? 

Now, it is a fact that credit is not as widely available as it used to be and lending criteria are much tighter. But that's not because of higher reserve requirements. Banks have been roundly (and rightly) criticised for taking excessive amounts of risk particularly in corporate and mortgage lending.  Northern Rock and Bradford & Bingley banks failed because of highly risky mortgage lending.  Excessively risky mortgage and corporate lending caused the failure of HBOS and the subsequent bailout of Lloyds TSB. And excessively risky corporate lending contributed to the collapse of RBS.  As Lord Levene said on the BBC's HardTalk programme the other day, it's a bit much to criticise banks for taking excessive risks in lending and then moan when they start being more careful about how much they lend and to whom!

In Tim's defence, I would have to say that the money multiplier (or deposit multiplier) theory is mainstream economic theory and widely promoted by all sorts of people. It just happens to be wrong.


  1. "What higher capital reserves do is increase the cost of borrowing, not reduce the availability of credit."

    Which means that your point makes no difference. Higher prices equals lower demand: we still get a contraction of credit/money supply out of higher capital requirements for banks.

    Please note that my snarl was at those who demand both an expansion of credit and higher capital reserves: the two are not compatible.

  2. Not sure I'm quite following you here.

    I understand the bit about reserves not being a consideration when banks make lending decisions but are you also saying that bank lending doesn't increase the amount of money in the economy?

  3. Rick, I would be the last person on this earth to suggest that bank lending doesn't increase the money supply. It does. Massively. Far more than it should. My point is that capital reserve requirements do NOT restrict bank lending. They raise the cost, which - as Worstall points out in his comment, above - does mean that bank lending should fall due to reduced demand.

    Banks do not lend against available reserves. They lend, and then obtain the reserves to support the lending they have ALREADY DONE. Intra-day - and, I suspect, intra-regulatory reporting periods - banks FAR exceed their nominal capital reserve limit.

    The risk they take is that having lent so much they are then unable to fund their reserve accounts. Prior to Northern Rock collapse reserve account funding in the UK was always done through interbank lending, with the BoE only providing emergency funds. However, the BoE now has a permanent discount window facility (similar to the Fed's facility, which has existed for a LONG time), which "tops up" reserve deficits if banks don't obtain funding through interbank borrowing.

    I hope this is clear?

  4. Before lending a book do you note how friends or loved ones handle books, weighing whether they are trust worthy to borrow a work? Or are you a throw caution to wind lender, not worried at all about the state of the book upon return.