Fiscal policy with a flat Phillips curve

It’s widely agreed that the Phillips curve is flat, that low unemployment is not stoking up wage inflation – though perhaps this has been true for longer than thought. This poses the question: what are the policy implications of this?

Simon says there’s a danger of the Bank raising interest rates too soon – because looking at unemployment and the output gap points to a risk of inflation which isn’t in fact there.

But I suspect it also has implications for fiscal policy.

I’m thinking here of the concept of the structural budget deficit. This is an estimate of how much the government would borrow if the output gap were zero. Conventional wisdom says that the bigger is this structural deficit, the more likely we are to need to tighten fiscal policy. One reason for this is that if the output gap is zero the economy is unlikely to grow quickly, so we can’t rely upon growth to narrow the deficit. Also, at a zero output gap interest rates might need to rise soon to prevent inflation rising, which would raise borrowing costs. Tighter fiscal policy would both obviate the need for much higher rates, and would make debt more sustainable if borrowing costs do rise.

This matters because the OBR believes that the deficit now is largely structural. Its latest Economic and Fiscal Outlook says:

With little sign of either spare capacity or overheating in the economy, we judge that the structural deficit (which excludes the effects of the economic cycle) is close to the headline deficit at 2.6 per cent of GDP. (Par 1.21)

But what if the output gap tells us nothing about future growth or inflation? (It doesn’t matter whether this is because the concept has always been wrong or because the link between “full employment” and inflation has changed). The need for a fiscal tightening then diminishes. If inflation doesn’t rise, government borrowing costs needn’t rise. And the economy is as likely to continue growing as it ever was, thus perhaps generating tax revenues.

The idea of a structural budget deficit makes sense to the extent that it’s sensible to think of inflation being caused by a negative output gap and its correlate full employment. If, however, the output gap and unemployment rate are uninformative for future growth or inflation, then the idea of a structural budget deficit doesn’t mean much. It should not then be used to justify fiscal tightening.

Of course, there might be other justifications – though I suspect these are weak at the zero bound and when real gilt yields are negative. All I’m saying is that if we’re throwing away the idea of the output gap, and if the Phillips curve stays flat, then we should also chuck out the notion of a structural deficit.

Comments

[Blissex]

«low unemployment is not stoking up wage inflation»

In not-so-recent recent "theories" of monopoly, some exceedingly well rewarded Economists of monopoly Economics have popularized the theory of contestability: that actual competition in a market is not necessary, because as along as a market is contestable, that is barriers to entry are low, monopolists won't raise prices to avoid attracting competition. Therefore the enormous monpoly profits of Microsoft or the City are not due to monopoly power, but to their fantastic productivity.

It is amazing for me that the theory of contestability has not been applied to the labour market. Except perhaps by a certain bearded Karl, who called it the "reserve army of labour" theory. But that is not equally popular. :-)

[Blissex]

«low unemployment is not stoking up wage inflation»

More seriously, a recent paper makes the point that given that the "methodologies" used to compute the reported unemployment index have been relentlessly "improved" over decades, it is more relevant to redo the Phillips curve with reference to the percentage of the population that is employed full time, and then things work rather better, "surprisingly".

[Ralph Musgrave]

“I’m thinking here of the concept of the structural budget deficit. This is an estimate of how much the government would borrow if the output gap were zero. Conventional wisdom says that the bigger is this structural deficit, the more likely we are to need to tighten fiscal policy.”

If that’s what “conventional wisdom” thinks, then CW needs its head looking into. If the output gap is zero, i.e. if we’re at full employment and inflation is not excessive, then whatever deficit obtains at that point must be the right deficit.

Thus the idea that in those circumstances we might need to “tighten fiscal policy” makes no sense: if the deficit is right, then it’s right. Obvious innit?

Chris then tries to justify cutting that structural deficit in his next sentence. He says “One reason for this is that if the output gap is zero the economy is unlikely to grow quickly, so we can’t rely upon growth to narrow the deficit.” But whence the assumption that the deficit needs to be “narrowed”? To repeat, if the deficit is right, then it’s right.

Chris does not seem to have got the point made by Keynes (which has been fully taken on board by MMTers and Positive Money) which is that the deficit needs to be whatever brings full employment (without exacerbating inflation). If that means a deficit of 2% of GDP, so be it. If it’s 4% or 6% so be it. As Keynes said:

“Look after unemployment, and the budget (i.e. the deficit) will look after itself”.

[Mark]

I believe Bill Phillips made a small but forgivable error of logic in describing his eponymous curve: he believed that inflation was inversely proportional to unemployment, whereas in fact the determining factor is excess capacity in the labour market - a subtle, but increasingly important difference.

As a result of EU freedom of movement, the UK labour market has huge spare capacity, probably in excess of the total number employed in the UK. There are a great many people in Eastern and Southern Europe who are unemployed, under-employed or, even if in full-time work, would find the wages on offer in the UK significantly more attractive than those attainable in the lower-waged economies in which they currently live.

Throw in a few genuine and many fraudulent asylum seekers (of 56,000 who landed in Italy between January and May, fewer than 1000 were actually from Syria, and some 23,000 were Nigerians) and you have a labour market that, despite headline unemployment at or below NAIRU, enjoys ample spare capacity. Thus employers are not competing among themselves for existing workers, raising rates to tempt the early-retired and stay at home moms back into the workforce or outbidding the benefits system to coax the unemployed back into the workforce.

While views on Brexit differ, I think commentators should reflect on the fact that Leave enjoyed support from the provincial working and lower-middle class, people whose wage rates might be expected to gain most from rationing the supply of migrant labour. They should also note how heavily employer interest groups such as the CBI are lobbying to retain Freedom of Movement...

[Gulliver Foyle]

This hinges on whether or not you actually believe the current unemployment figures: -

http://uk.businessinsider.com/unemployment-in-the-uk-is-now-so-low-its-in-danger-of-exposing-the-lie-used-to-create-the-numbers-2017-7

[Egmont Kakarot-Handtke]

Putting economic policy on scientific foundations
Comment on Chris Dillow on ‘Fiscal policy with a flat Phillips curve’

Chris Dillow says: “It’s widely agreed that the Phillips curve is flat, that low unemployment is not stoking up wage inflation ― though perhaps this has been true for longer than thought.” Perhaps it never has been true.

The fact that the Phillips curve now seems to be flat only tells one that it has been misspecified all along. Thanks to the scientific incompetence of economists this remained undetected since Samuelson/Solow messed things up. The methodological blunder consists in interpreting the Phillips curve as a behavioral relationship. What has to be done is to formulate the Phillips curve as a structural-systemic relationship.

To make matters short here, the elementary version of the correct systemic Phillips curve is shown on Wikimedia.#1, #2 This relationship consists alone of measurable variables and is therefore identical with the observed Phillips curve.

The correct relationship covers the familiar arguments about how effective demand affects employment. Secondly, the ratio rhoF embodies the macroeconomic price mechanism. It works such that overall employment L INCREASES if the average wage rate W INCREASES relative to average price P and productivity R and vice versa. This is the opposite of what microfounded employment theory teaches.

The correct systemic Phillips curve tells one that in the given situation the most urgent policy measure is to set an increase of the average wage rate in motion. Otherwise, the economy gets trapped a spiral of rising unemployment and deflation.

Egmont Kakarot-Handtke

#1 Structural-systemic Phillips curve
https://commons.wikimedia.org/wiki/File:AXEC36.png

#2 For the derivation see Sections 5 to 7 of the working paper ‘Keynes’s Employment Function and the Gratuitous Phillips Curve Disaster’.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2130421

[derrida derider]

"It’s widely agreed that the Phillips curve is flat"
Lo, how the mighty have fallen. It's not so long ago that the utter orthodoxy was that the thing was vertical in the long run and with precious little slope in the short run. In fact monetarist and RBC types held it to be vertical even in the very short run (neutrality and - for RBC - superneutrality of money and all that).

It turns out that the anti-austerity economists were right - who coulda known?

[Blissex]

«If the output gap is zero, i.e. if we’re at full employment and inflation is not excessive, then whatever deficit obtains at that point must be the right deficit.»

Technicality here: in "mainstream" Economics that only holds in a perfectly undistorted set of markets. If the markets are distorted by government intervention or any other limit to perfect competition, then nothing at all can be said about equilibrium, "optimality" or not. "because microfoundations!" :-)

[Arthur Murray]

At PMQs on 19 July (I think) May quoted the following statistic: “the top 1% now pay 27% of all the income tax paid in the UK”.

This must mean that the 99% pay the remaining 63%.

Whilst we were obviously meant to be impressed by how large 27% was, there is a way to increase this to a really impressive 100%.

Cut the pay of the 99% so that its mean (average) value before tax is £11500, the threshold above which anyone pays tax. Of course with £28000 as the median pre-tax pay rate for full-time workers, it might mean a bit of belt-tightening for them.

That might not be as unpopular as you think. A large percentage of Leave voters aged 65+ believe a big hit to prosperity is a price worth paying for Brexit.

[Blissex]

«A large percentage of Leave voters aged 65+ believe a big hit to prosperity»

Not that big, about as bad of the 2008 depression, over 10-15 years.

«is a price worth paying for Brexit»

To be paid by other people: people 65+ typically are on fixed incomes from pensions or assets. A blogger reported this conversation between a young friend and her father:

«Friend: How did you vote then, Dad?
Dad: I voted Out.
Friend: Dad! Why did you do that? The economy will crash! It’ll cause chaos!
Dad: That won’t bother me hen, I’m retired.
Friend: But it’ll affect me! What about me?
Dad: (Long silence).»

[Blissex]

«"It’s widely agreed that the Phillips curve is flat"
Lo, how the mighty have fallen.»

Very good point...

There is the other point that Phillips Curve/NAIRU have been shown by a paper to hold/be predictive in some periods but not in others, and it is not possible to figure out at the time whether they are going to hold or not, only retrospectively.

Translation in english: random bollocks.

[Luis Enrique]

"If inflation doesn’t rise, government borrowing costs needn’t rise. "

I don't get this: aren't we worried about real interest rates?

[rumplestiltskin]

https://www.bloomberg.com/view/articles/2017-08-03/japan-buries-our-most-cherished-economic-ideas

[The Arthurian]

Blissex: "There is the other point that Phillips Curve/NAIRU have been shown by a paper to hold/be predictive in some periods but not in others, and it is not possible to figure out at the time whether they are going to hold or not, only retrospectively."

Interesting. What paper?

(But maybe it's not random. Maybe it is part of an economic cycle other than the one we're thinking of...