How to get banks lending to SMEs

David Cameron enjoyed his swipe at the banks today, using his speech at the party conference to lambast them for not lending to small and medium sized enterprises. He wants the economy to get moving despite huge public spending cuts and said he’s fed up with banks not doing their bit. Sadly for us, lending targets and prime ministerial taunts won’t help, even if they are fun.

How about some other options; couldn't we harness some of the billions the Bank of England is creating through quantitative easing to lend to SMEs? And what about the new Basel 3 rules, they must have something to offer?

First QE, or the Bank of England's creation of £200bn to buy gilts. It's the official response to our flagging economy but doesn't seem to be refreshing the economic parts that need it. Why not?

I'm no banker but from what I've read, the cash injected by QE is meant to boost the reserves of commercial banks so they can lend more and then lend again against the newly created deposits. This “deposit multiplier” acts like Red Bull on the economy, with each new deposit backing a new loan and each new loan creating another deposit. But when banks are short of capital, as now, they’ll either hoard money or lend it to higher rated borrowers, such as ABS-merchants. Which means the money’s not being lent to productive borrowers and doesn't stimulate many SMEs.

QE's other effect is to lower bond yields, which helps investment by making it cheaper to raise capital. But the effect is always limited by investor appetite and that's the other problem: selling low-yield securities to fund growth in a near-recession is a thankless task.

The sickness of banks and the economy are not the only limits on QE’s power as a stimulant. Like Red Bull, QE gets less effective the more you use it. It distorts the gilt markets, causing the Bank of England to start buying 3 year gilts after its 5 year purchases started to wear thin (para 15). And they both store up nasty headaches, in QE's case in the form of inflation or contraction of the money supply when the Bank’s growing pile of gilts is sold back to the market.

If all the newly-created QE money were lent directly to small businesses instead, things would be much simpler. The cash wouldn’t get trapped inside banks or credit securitisations, and instead of a deposit multiplier we’d have a demand multiplier, where money lent to SMEs would be spent on their salaries and suppliers, who would spend it on salaries and suppliers, who would spend it on... and so on until the economy starts breathing again.

Unfortunately, central banks are not allowed to print money to fund deficits, which is how this would look to outsiders. They are allowed to create money to chase deflation, which leaves us stuck with QE and, at the other end of the clogged plumbing, a SME sector still starved of credit.

Basel 3

A different approach would be to change how we implement risk-weightings in the Basel capital adequacy rules. Under the draft Basel 3, banks must increase the amount of regulatory capital they set aside to 7% of risk-adjusted assets. While the headline is good, it's only half the equation; risk-weightings are just as big an influence on how banks lend.

To illustrate, AAA-rated asset backed securities, like some of the toxic rubbish that caused the crisis, have a 20% risk-weighting, meaning a bank has to hold 7% of 20% in capital, or 1.4% of the value of the asset. This means a bank can gear up 71.4 times on these assets and earn 36% return on equity (assuming 0.5% margin).

SMEs, on the other hand, carry a 100% risk weighting, meaning the bank has to hold 7% of the full amount. This allows it to lend only 14.2 times its capital and earn 7% return on equity.

Lending to SMEs may well be five times riskier than buying AAA-rated asset backed securities (or not), but banks already know how to handle that through interest and other terms. The quantity of each they can hold, however, and the return on equity, is set by Basel.

Former World Bank director Per Kurowski has been arguing for years that the Basel system is flawed for this reason, since it encourages banks to load up on debt they think is safe, and when it turns out not to be we have a crisis (interestingly, sovereign debt is still in the safe category).

Kurowski has been promoting his message doggedly with letters to the FT and his entertainingly blunt blog. His message is slowly being heard (see here for Felix Salmon's latest) but so far not in Basel.

Kurowski proposes gradually equalising the risk weightings until there is no longer a regulatory incentive for banks to favour one asset class over another, and leaving banks to manage their exposure through interest and lending criteria, which they do already.

If David Cameron is serious about SME lending, he could do a lot worse than take Kurowski’s idea a step further and push for risk weightings to be set with the real economy in mind, instead of an out-of-date statistical view of which assets are risky.

By turning risk weightings into a policy tool, Cameron could make it cheaper and more profitable for banks to lend to productive businesses and less profitable to stuff their balance sheets with CDOs. He would boost SME lending more than any amount of QE or bank lending targets.

Market fundamentalists won’t like this but it’s not as radical as it sounds. We already use price interventions to change market outcomes on alcohol, tobacco and carbon fuels. We subsidise venture funding and R&D with tax breaks and corporate investment with capital allowances. We accept that pollution should be taxed, and the Bank of England’s Andy Haldane has asked openly whether finance’s “negative externalities” should be dealt with the same way.

In a recent white paper, the Treasury listed variable risk-weights as one of the macro-prudential tools it expects the new Financial Stability Committee to use to control the economy:

"Variable risk weights: This would involve raising capital requirements against specific types of lending. If the authorities felt financial institutions’ exposure to a certain asset class was too great, they could try to discourage it in this way, (page 16)"

You might question how a regulator would ever know the correct risk-weighting for SME lending but this misses the point: banks could lend as much or as little as they like, on their own judgement and on their own terms. All the regulator would be doing is adjusting the reward they end up with, like any government tax or subsidy.

There's a risk of arbitrage if risk-weights differ between countries, but we seem to manage other cross-border differences OK and there'a substantial upside if it can be made to work. Risk-weightings could be lowered in downturns and lifted in boom years. They could become a lever of industrial policy to boost lending in sectors with high-growth or structural change, and dampen it in sectors that are oversupplied or in the grip of a bubble.

In a covert way the existing Basel rules do this already by favouring investment banks over commercial banks. Worse, they do it by referring to ratings agency instead of deliberate, democratically chosen economic policy. Time for a re-think?