Budgets are about much more than fiscal sustainability

 

Ben Zaranko of the IFS has just published a paper on fiscal rules. It is well researched and well written, but I disagree with its conclusions. For those interested in the subject I strongly recommend reading the paper. Here I will restate his main argument, and say why I disagree with his solution. I will also add what I think is the main problem with the current fiscal debate, and it has nothing to do with fiscal rules.

Problems with the current system

For clarity, let me focus on the current situation where the golden rule (taxes match current spending) is binding, and ignore the other fiscal rule (falling debt to GDP) which is terrible and should be abolished. This also happens to be the fiscal set-up that I prefer. Ben has a section (“Why not focus solely on the current budget balance?”) where he notes two potential problems with this set-up. I will add a third, which he mentions in his introduction.

1. The golden rule does not ensure ‘sustainability’ because it excludes public investment.

2. Volatility in the forecast will lead to volatility in policy.

3. In practice this rule is supplemented with a variable amount of headroom which leads to various problems.

I will deal with each problem in turn.

The first disappears when you recognise that there is nothing magic about zero for a target for the current budget deficit. It could equally be a target for a 1% surplus or deficit. Which it should be is governed by the society’s long term goals. Suppose, for the sake of argument, that everyone agrees government debt should be around 50% of GDP in fifty years time in the absence of major shocks. [1] The target for the current balance should be consistent with that long term objective.

Working out what current balance target corresponds with achieving that longer term objective involves various assumptions, like what the average real interest rate will be over the next fifty years. It will also include what the average level of public investment will be. It really is irrelevant in this respect whether that investment is partially ‘self-financing’ or not. Once those assumptions are made, working out what the current balance target should be becomes a calculation that can be done on a spreadsheet, but it would be good if the government could periodically (every few years) show that calculation. [2]

The second problem sounds more serious. To quote Ben:

“the framework and the way it has come to be operated mean that when forecasts move around – as they inevitably do – policy has to respond, often in a rush and with a spurious degree of precision. This does not make for good policymaking ….”

A core principle of fiscal policy is that taxes and spending should be smooth, letting changes in borrowing absorb short term volatility. If forecast volatility is leading to spending or tax volatility, we have a problem.

However I’m not at all sure forecasts are volatile in the sense of moving randomly either side of an underlying trend, particularly if they relate to the medium term that excludes cyclical variation. (Which is one reason why targets should be for quantities five years ahead, not three.) My impression is that often forecasts respond gradually to real shifts in the economy (like changes in productivity, for example), leading to years where the fiscal corrections due to the forecast are of the same sign. For this reason, I think adjusting the public finances once a year to achieve a fiscal rule seems about right to me. (Doing this every six months was silly, as I said at the time.)

Which brings us to the third problem, which is headroom. Everyone seems to agree that it is sensible for the government to go beyond its fiscal rule and add a certain amount of headroom: a buffer to absorb shocks. But if medium term forecasts are not erratic, why do you need a buffer? Why not set policy to hit the fiscal rule exactly? The answer has to be that tax rates are round numbers, so it is never possible to hit a fiscal rule exactly, and trying to do so by adjusting minor tax rates or allowances would lead to distortions.

My own preferred solution to the problem of headroom is to have none, and treat the fiscal rule as symmetrical rather than as some upper limit. As I explained, headroom is a concept that makes sense at the individual or company level but makes no sense for a government. However the political reality is that although central banks are seen as being responsible even when they forecast inflation two years out a bit above their target, we seem to live in a media climate where a government occasionally allowing itself to have a bit less than 50% chance of meeting its fiscal rule is deemed unacceptable. As a result, governments will allow themselves headroom.

As Ben explains, headroom has only become a problem when it started to be set at persistently very small levels (his Figure 3.1), an ‘innovation’ introduced by Jeremy Hunt. (Hunt was a terrible Chancellor for many reasons.) As Ben also notes, Reeves has now gone back to something more like pre-Hunt headroom levels, so hopefully the problem of headroom has also now gone away.

A traffic lights approach

Ben’s solution to these perceived problems is to move from a single fiscal rule to a multi-indicator approach, where a number of measures would be looked at and they would be assessed with a traffic light system. It’s like having multiple fiscal rules, but where none is binding, so policy can fail on some as long as it meets others.

One problem with this is that it makes the problem of fiscal sustainability sound like a complex issue when it is not. I don’t need lots of indicators to work out whether my bank balance will go into the red in a few months’ time, and sustainability is really just a similar matter of arithmetic, albeit over a much longer time period and without a bank manager or overdraft. We have fiscal rules because politicians are not very good at raising taxes or cutting spending, leading to deficit bias, and one simple rule should be sufficient to correct this bias.

In addition, if you in effect have lots of targets, then how do we know when the government is avoiding ‘difficult decisions’ by spending too much or not taxing enough? You need an expert to interpret all the traffic lights. Traffic lights are great for people like me or the IFS, who will be called on to interpret all the red, yellow and green, but it is not good for public transparency.

With many targets you have a whole new set of problems. Are the targets consistent with one another, or are the assumptions that make them consistent plausible? In addition many indicators allow politicians to cherry pick the targets that happen to look OK even though sustainability is not being achieved, or alternatively it can allow the media to insist that most of the targets are hit, resulting in too tight a fiscal stance.

As I noted above, all you need to ensure governments are trying to ensure sustainability and avoiding deficit bias is one simple rule, and periodically the spreadsheet calculations that lie behind it. I think a good deal of the dissatisfaction that Ben and many others have about fiscal policy at present is that this issue of sustainability gets far too much focus, from the media and (therefore?) politicians. Ironically having a traffic light system rather than one simple rule will mean more, not less, focus on sustainability.

The real problem with fiscal policy is the focus on sustainability

But I do have some sympathy with Ben in that having a single target does make it appear as if the forecast is determining fiscal policy. But the fault here is with the media and politicians and not the target. Fiscal sustainability is just one aspect of fiscal policy, and in many senses the least interesting aspect. It may appear far more important than it really is because politicians have ignored or downplayed other key aspects of fiscal policy.

The most important microeconomic objective of fiscal policy changes is to achieve a coherent and fair system that provides appropriate incentives in the simplest possible way. We should be asking the Chancellor each budget what have they done to achieve this, rather than obsessing with how much headroom they are allowing for fiscal rules. Unfortunately the media typically only touches on the issue of fairness, by talking about winners and losers, and largely ignores issues of incentiives, efficiency and simplicity.

Just as there is a tendency for politicians to avoid difficult political decisions which then lead to deficit bias, there is also a tendency for them to go for tax changes that are popular in the short term but make the tax system both complex and inefficient and which provide the wrong incentives. For example Labour in opposition pledged to not raise taxes on working people, so it put up taxes on employment thereby reducing output. Not raising petrol duty gets favourable headlines, but as a result over the last decade the cost of driving a car has risen by less than the cost of rail travel and by much less than travelling by bus, which hardly provides an incentive for greener travel.

Creating fiscal rules and fiscal institutions like the OBR has helped reduce the problem of deficit bias. I don’t know how you incentivise politicians to focus on creating a more efficient tax system, but making achieving sustainability more complex is not the way to achieve that.

In addition, sustainability is not the only macroeconomic goal for fiscal policy. In the last budget the Chancellor had a clear aim to reduce short term inflation using various fiscal measures, with the goal of reducing both short and longer term interest rates. While the media picked this up, I saw very little discussion of the macroeconomic merits or otherwise of this strategy. Furthermore in a recession, of course, fiscal policy should be expansionary to support or replace monetary policy, and the fiscal rule should be suspended.

More generally, the stance of fiscal policy has implications for aggregate demand and therefore short term interest rates. While it is true that there is a broad relationship between financial fiscal rules and aggregate demand, it is not an exact equivalence. Hitting the golden fiscal rule will often imply a neutral medium term fiscal stance in aggregate demand terms, but not always. [3] In addition, there may be scope for influencing (in either direction) aggregate demand in the short run while keeping to fiscal rules in the medium term. By thinking about the impact of fiscal policy on aggregate demand the government is not trying to duplicate the Bank’s role in controlling inflation, but instead is trying to influence what short term interest rates the central bank will end up setting to control inflation over the forecast period.

Summary

The idea that the OBR forecast determines what a Chancellor does is understandable given much of the media’s coverage of this event. But it is also a big misunderstanding of what is going on. That the Chancellor has recently had to raise taxes to meet her spending plans and fiscal rule shows that the goal of ensuring fiscal sustainability is being achieved. In that sense it represents a success story. Replacing that fiscal rule with a set of traffic light indicators could put that success at risk by adding considerable confusion. However a Budget involves much moe important microeconomic and macroeconomic issues than fiscal sustainability, and if that doesn’t receive enough attention that is the fault of politicians and the media, not a simple fiscal rule.

[1] Whether that long term objective is reasonable would require far too much discussion for this post. Note, however, that a major justification for such a goal is that it allows space for debt to rise after major shocks, like a pandemic or a major recession.

Another point is if reducing debt over fifty years is a reasonable objective, what is wrong with the falling debt to GDP rule. The answer is simple, if you ask what this rule adds to the current deficit rule? What it adds is public investment, thereby undoing the good that is done by excluding it in the current balance rule. 

[2] Such a calculation is quite different from the 50 year projection done by the OBR. The OBR’s projection assumes current policies for spending and taxes, and calculates the implications for the government’s current balance. The calculation the government would do would assume the current balance target was met in each year beyond the five year forecast.

[3] In this sense there is no great contradiction between a focus on a medium term fiscal rule based on the government finances, and setting fiscal policy to eliminate inflationary pressure in the medium term. The financial reflects the real. MMT prefers setting fiscal policy to control inflation in the short term rather than using changes to interest rates to do this job.