Structural deficit doubts

The OBR’s new forecasts remind me why I hate the concept of a structural budget deficit.

It has reduced its estimate for trend growth in recent years and thus reduced its estimate of the output gap. Because of this, it thinks more borrowing is “structural” and less “cyclical” than it previously thought.:

By 2015-16, we expect PSNB to have fallen to £53 billion or 2.9 per cent of GDP, compared to the £29 billion or 1.5 per cent of GDP that we forecast in March. The extra borrowing is primarily structural rather than cyclical, in other words it will not disappear as the economy recovers.

This in turn has put pressure on Osborne to tighten policy in order to meet his self-imposed target of balancing the structural current budget in five years’ time.

I have four problems with this:

1. It makes policy pro-cyclical. If estimates of trend growth fall when growth is weak, then the desire to balance the structural budget imposes fiscal tightening in bad times*.

And it’s not just fiscal policy that becomes pro-cyclical. So does monetary policy. A lower output gap estimate means - other things equal - a higher forecast for inflation and hence, under inflation targeting, less justification for loose monetary policy.

Luckily, it’s not clear that the Bank shares the OBR’s pessimistic view of spare capacity; it seems more optimistic about inflation than the OBR. But this is a happy accident. A policy framework that depends upon the monetary and fiscal authorities having different views is not sustainable.

2. The distinction between “cyclical” and “structural” borrowing does nothing to answer the key question, which is: will the gilt market remain content to finance borrowing? It might be that the gilt market is happy to finance even “structural” borrowing, insofar as this is the counterpart of weak long-run growth and hence poor returns on equities. Or maybe not. Merely labelling borrowing “structural” doesn’t help.

3. The concept of a structural deficit leaves unanswered the question: how, exactly, does a fiscal tightening reduce the deficit?

What I mean is that a structural deficit implies that, if the economy were at its trend level of output, the private sector would be saving more than it invests: this is the counterpart of the government’s deficit. To reduce the structural deficit thus requires that private sector investment rises and/or savings fall. But how does fiscal tightening achieve this?

4. Forecasts of the structural deficit rest upon two huge uncertainties - and it’s not clear that they offset each other: an uncertain estimate of the output gap (and sensitivities of spending and revenues thereto); and  the usual uncertainties surrounding any fiscal forecast. I’m with Simon Ward on this: “it is troubling that the fiscal framework pivots on a concept subject to huge empirical uncertainty.”

I suspect that the notion of a structural deficit is playing an ideological role - one that‘s a little analogous to the reasons why the Greeks and Italians have “technocratic“ governments. A pseudo-scientific claim to expertise is being used to disguise what is, in fact, a judgment, that borrowing is too high and can and should be reduced by fiscal policy.

* It’s not good enough to claim that Osborne’s new tightening is back-end loaded, with spending cuts in 2015-17. The anticipation of future tightening might depress activity now.

Comments

[John H]

I can't pretend to understand all the economical ins and outs of this, but to a layperson the concept of "structural deficit" and its elimination seems to bear more than a passing resemblance to Gordon Brown's fabled "Golden Rule"...

[Luis Enrique]

Isn't it a reasonably sensible question to ask: when things recover (for a given value of 'recover') how large a deficit do we expect the government to be running? It looks to me like the sort of question one might ask if considering lending money to a government.

If we revise our answer to that question in the light of negative developments, you object that "makes policy pro-cyclical". Doesn't that beg a few questions? Namely: whether we need to do something because long-run expectations of the state of public finances have worsened. If we do need to do something, then we need to do it! no point in complaining you don't like it because this is "pro-cyclical". If we don't need to do something, well then don't. And of course we don't know what we can do - if fiscal tightening doesn't actually improve the expected state of public finances, then there's no point in doing it just because the expected 'structural deficit' has increased. Hence no pro-cyclical policy.

[Jorjun]

Reading between the lines, because the language is very technical, I think you are saying that when we lowered sterling to fudge the stats, and sold off cash-cows such such as Cadbury's and Boots to foreign investors at a discount, this reduced our income not just in this financial period but for ever after. But calling this 'structural' is meaningless within a theoretical framework.

And that rather than reduce expenditure on items we that we desire but, cannot afford, and in actual fact place an activity-reducing burden on the masses (taxation) we should carry on chugging on the Keynesian train. There are still plenty more assets that we can get rid of to trigger a revolution when we finally hit the buffers & are run on much more efficient, since authoritarian, line.

[Luis Enrique]

also ... "The distinction between “cyclical” and “structural” borrowing does nothing to answer the key question, which is: will the gilt market remain content to finance borrowing?"

couldn't you say that the distinction between cyclical and structural is actually based on what bond investors will be content to finance? What I mean is, if you are content to finance a deficit that would be unsustainable if kept up over the long-run because you believe it will be finite and that soon enough it will shrink and stay shrunk (for long-enough), then you have made a distinction between the cyclical and the structural. You are saying that you are content to finance deficits whilst the government is coping with a recession in the belief that something called normality will return, when the deficit will shrink sufficiently. If conversely you thought what we were experiencing was not a cyclical (temporary) state of affairs, but the new normal, then you would not be willing to finance such a deficit (because you have no hope of it shrinking) hence you have just classified that deficit as structural not cyclical.

[Luis Enrique]

sorry, I may, as I often do, have missed your intention here - if you are arguing that thinking in this way encourages governments to attempt things they needn't attempt using policy instruments that don't achieve what they want them to, then quite possibly you are correct! Above I was arguing that I can't see much wrong with thinking in these terms per se.

[Gareth]

Good post.

"The anticipation of future tightening might depress activity now."

Yup. Now apply this to monetary policy too...

[Ralph Musgrave]

Re 1, the removal or “balancing” of a structural deficit, by definition, does not result in “fiscal tightening”. Reason is that INCURRING a structural deficit does not, by definition, impart stimulus. Therefor removing a structural deficit does not impart “anti-stimulus”. At least that is the case if one adopts the Reuters or Wiki definition of the phrase “structural deficit”. At a stretch same applies if one adopts the (rather poor) Financial Times Lexicon definition.

Re 2, I realise the conventional (and nonsensical) wisdom is that it is important that markets “remain content to finance borrowing”. The reality is that there is no need whatever for a monetarily sovereign country, like the UK, to borrow something (i.e. money) which it can produce at no cost and in limitless quantities any time. As both Keynes and Milton Friedman pointed out, stimulus deficits (as opposed to structural deficits) can be funded with EITHER printed OR borrowed money.

As to what to do if creditors don’t want to roll over structural debt – no problem. Just pay back the creditors with a combination of printed money and money obtained from increased taxes. As long as the inflationary effect of the former equals the deflationary effect of the latter, there is no net effect on GDP, numbers employed, etc.

For more details on this see: http://mpra.ub.uni-muenchen.de/34295/1/MPRA_paper_34295.pdf

Re 3, see the latter URL.

[Luis Enrique]

Ralph, I just don't understand why you think financing a deficit of the size we have for the duration we have it by printing money would not have an inflationary impact so large that if we wish to avoid it we do, in fact, have to borrow money.

[gastro george]

Luis, I'm but poorly read on MMT, but I think that the point is that bond sales are not *required*, but that the excess liquidity would have to be removed in some way to prevent inflation. However, there are other options available, and the threat of inflation is low while we have spare capacity (industrial and unemployment). Bill Mitchell's blog would describe this much better than I can.

[Philip Walker]

Why is everyone keen on Ricardian equivalence only when it suits their side of the argument?

[Keith]

It would be simpler to just say that Osbourne and every one else is talking rubbish: the concept of a structural deficit involves arbitrary assumptions and conventions none of which have any necessity about them. No one really knows what the long run or even medium term growth rate of the economy could be or how much inflation will be or indeed by how much or, if at all, inflation might rise if the state deficit were partly financed by creating monetary base. Savings have to go some where, and you can probably borrow if you want since some of the savings will come to you, and you can expand the money supply until you actually get inflation and worry about it then. The difference between today and 1964 or 1974 is not that it is impossible to boost demand and growth or borrow but that economic fashion has changed. Full employment and maximum growth were the absolute goal of economic policy during the social democratic hay day. Now fear of Inflation obsesses the formation of policy even when Inflation is low. It also seems that every one thinks the state should refuse to do useful things to avoid upsetting bankers and financiers; rather than control finance to achieve economic goals. That is a political choice and not an economic one. If you do get economic growth the Budget takes care of itself. It may even go into surplus as in 1945 -1950. In 1948 or 1952 no one worried about financial aggregates but they complained about rationing of bread. The production of more bread is an economic problem; finance is or should be a mere technical issue! You don't refuse to solve economic problems using finance as an excuse unless you are being deliberately obtuse.

[BlackRaven]

surely a much easier way for government to estimate what is structural and cyclical is to look at the nature of the spending, it is pretty obvious that a rise in jsa payments is cyclical, whereas an increase in the cost of an xray machine is not.

[ShodanAlexM]

The whole discussion reminds me of the natural rate of unemployment, which moved arbitarily and sharply up or down in line with actual unemployment depending on whether it looked like inflation might be developing. Suggesting it wasn't 'natural' after all.

That was guff. The idea of a structural deficit isn't much better.

[Account Deleted]

Hello,
The fast and convenient online store offering your favorite British Food right here in the USA.http://www.britishfooddepot.com/