The Bank of England might resume quantitative easing as soon as next week. Would this work?
The Bank, naturally, thinks it will. An article in its latest Quarterly Bulletin claimed that QE in 2009-10 “may have raised the level of real GDP by 1.5 to 2 per cent”, and increased inflation by 0.75-1.5 percentage points, though it admitted that these estimates were “highly uncertain.”
Note that even if the upper end of these estimates are correct, £200bn of QE led to only a £50bn rise in money GDP. This means that most of the money printed by the Bank stayed in the financial system, rather than affects nominal GDP.
What’s more, I fear that the Bank’s estimates might be on the high side, especially for the next round of QE.
I say this because there are four possible channels through which QE might affect activity, and all are suspect.
The one which the Bank considers most important is the portfolio balance channel. The idea here is that when investors sell gilts to the Bank, they use the money to buy other financial assets such as corporate bonds or equities, and this raises their prices and so encourages firms to issue more of them, and use the proceeds to invest.
The Bank estimates that QE not only reduced gilt yields by a full percentage point, but also reduced corporate yields significantly, and this did lead to higher issuance.
However, the Bank’s research could find no direct effect of QE upon share prices; yes, shares rose sharply in 2009-10, but this was probably more due to a global recovery than to local QE. This suggests that there’s no reason why QE should boost capital spending via higher equity issuance.
And I’m not sure it will much affect bond issuance either. In recent months, non-financial firms have been net buyers of bonds and shares, despite low yields on the former; blue chip firms such as Prudential or Tesco can borrow long-term at less than five per cent. This suggests that the problem is a lack of investment opportunities, not – among large firms at least – a lack of finance for them.
Yes, some small firms are constrained by a lack of finance – twas ever thus! But these are too small to issue bonds anyway.
A second mechanism through which QE might work is a policy signalling effect. QE signals that the Bank intends to keep interest rates low for a long time. And faced with low returns on cash, people might go out and spend.
The problem with this is that – as we saw last week when the stock market fell after the Fed announced “operation twist” – is that signals are always ambiguous. Loose policy doesn’t just signal low future interest rates. It also signals that the economy is in a bad shape. The latter signal tends to reduce spending, whilst the former raises it. It’s not clear which effect dominates.
There is, though, a variant on this theme. Some Bank economists think that QE1 worked because it reduced tail risk. In 2009, people didn’t just fear recession, but a complete meltdown of the economy and financial markets. QE1 helped reduce this risk and so raise confidence.
But I’m not sure how relevant this is right now. Yes, there is tail risk. But it emerges from the euro area’s crisis, not the domestic economy.
A third possibility is that, insofar as the Bank buys gilts from commercial banks, it increases their cash reserves and hence their ability to lend.
However, the Bank itself doesn’t think this a significant channel. In terms of liquidity, gilts are a near-substitute for cash, so replacing one with the other doesn’t much improve banks’ balance sheets and lending capacity.
Finally, it’s possible that QE would reduce sterling – insofar as investors use the cash to buy overseas assets – and this would improve our net exports.
Again, though, the Bank thinks this mechanism is weak. It estimates that QE1 only reduced sterling’s trade-weighted index by around four per cent. Combined with the low price-elasticity of net exports, this implies only a marginal effect upon economic activity.
So, I doubt whether QE2 would do much to boost demand.
Worse still, even if it does, the effect might be to raise inflation more than output.
The reason for fearing this lies in a recent speech by Spencer Dale, an MPC member. He fears that the financial crisis, among other things, has reduced productivity growth and hence the UK’s supply potential. And if the economy doesn’t have the capacity to meet increased demand, QE – insofar as it works at all – might lead to higher prices rather than greatly increased output.
So, I’m sceptical that QE would work very much. What I’m not sure of, though, is whether this is an argument against QE per se, or merely against QE in the form of buying gilts.
There are some more radical forms QE might take. Adam Posen, another MPC member, has suggested that QE be used to provide capital for a government-run bank to lend to small firms, and to finance an entity which can securitize bank loans to firms – a kind of Freddie Mac for company lending*. Even more radically, some have proposed a direct “helicopter drop” of money to low income households.
I suspect these would do more than conventional QE to boost output – and in the latter case inflation!. The problem is not that monetary policy is impotent, but that policy-makers aren’t (yet?) ready to try its more extreme forms.
* He suggests this be called the Bennie, or British Enterprise Investment Entity. I fear he doesn’t remember Crossroads.