The Chancellor has announced that he wants to legislate to force future governments to run surpluses in ‘normal times’. Two reasons make it hard to comment definitively on this proposal. First, the devil will be in the detail of what constitutes ‘normal’. Second, there are many unanswered questions in the theory and empirics of fiscal policy. Nonetheless, a few points as a footnote to what others, particularly Simon Wren Lewis, have written in reaction.
First, it’s interesting to note that Robert Chote, head of the Office for Budget Responsibility, has suggested that they don’t want to be responsible for defining ‘normal’. Monitoring whether given circumstances meet a government definition is fine, but coming up with a definition is not.
There is a logic to this that parrallels how our central bank is treated. The Government designs the inflation target – the number, the price index, and so on – and the BoE decides whether policy is appropriate to meet it. Chote is urging the same division of responsibility. Goals are set by government, and delegated agencies operate the instruments to achieve them. In the case of fiscal policy, governments retain also the instruments, but the OBR would have independence over its advice, which plays the role of an instrument here.
Whatever they do, if it’s true, as I’m told by someone who has seen, that they only have 4 full time economists at working level, they are going to need more resources to do it. 4! The Bank of England probably employ about 250 economists. For sure, the OBR no doubt do much of their work through the capable hands of cooperative civil servants in the relevant departments. But still. 4?
On the concept of ‘normality’. The notion that the Chancellor – and Chote in his comments – are heading towards is whether or not there is an ‘output gap’. This is a term in quite wide usage, referring to the gap between what the economy is producing, and ‘potential’, if all resources were put to use.
But it is actually an extremely fuzzy idea, one that only really becomes concrete in a particular theoretical or empirical model. We could think of the gap between actual output and a trend line. And there are many concepts of trend lines. Straight line trends. Trends that assume that the level of otuput is mean-reverting, and trend lines that assme that the growth rate of output is mean-reverting. Moving averages, and other filters.
We could think of ‘normal’ as being, in the New Keynesian macro model, the level of output that would obtain if, counterfactually, all prices and wages were flexible. Or further, as the level of output we would get if some of the other frictions that slow the adjustment of capital for example, were not operating. Some business cycle theorists would see ALL movements in output as ‘normal’. ‘Normal’ is highly controversial in macro.
Exactly the same issue, of course, is grappled with by the Bank of England’s Monetary Policy Committee. In order to produce a forecast for inflation, and inform a judgement about the trajectory of their instrument settings, they need to take a punt on ‘normal’.
There is something to be said for the idea that, for the sake of credibility, this fuzziness should be resolved by making a choice – even if not warranted by the diffuse evidence – in favour of some definition. But we should not pretend that any Chancellor defined concept is more than this.
I’m in favour of the principle that we should aim at paying down the debt slowly, so that we can run it back up again with vigorous fiscal stimulus when needed in a crisis. But the unknown is wether fixing on an overall surplus is too conservative or not. Simon Wren Lewis worries that this would mean the same generation paying for the crisis it had experienced. That worry is founded if crises happen less than once in a generation. My take so far from the crisis is i) they are more likely than we previously thought, notwithstanding the financial reforms enacted subsequently; ii) in the absence of good knowledge about the distribution of future crises, it’s better to err a little on the side of caution.
Some have scoffed at the suggestion this be built into legislation. They point out correctly that a government cannot bind future parliaments. I’m not so sceptical. The Bank of England Act could be repealed easily enough, undoing central bank independence, which binds governments under ‘normal’ circumstances not to interfere in monetary policy. But it so far has not been. And I conjecture that’s because legislating has raised the political costs of changing tack. So I think legislation might be a positive step. And since it confers a certain amount of technocratic independence over fiscal policy [on the OBR] that’s a good thing too.
A particular concern I have – oft stated here and by others – is that this long-term fiscal musing is being done in the context of a determination to make aggressive cuts to spending while interest rates are trapped at the zero bound, there being now limited scope for the BoE to counter any ensuing contraction that might prevent them from returning inflation to target. Part of the reason for doing this, I suspect, is the perceived need to earn credibility for sound public finance management, so that the ‘job’ can be seen to have been ‘finished’. If we are in the business of mandating the OBR, or some other third-party, to comment on fiscal policy, a priority for me – more pressing, and less controversial than the Chancellor’s mooted surplus rule – is for such a body to be able give the government cover for vigorous deficit-stimulus in today’s circumstances.