Inflation and pandemic recoveries in five major economies

My discussion about current inflation two weeks ago focused on the UK. Over a year ago I wrote a post called “Inflation and a potential recession in 4 major economies”, looking at the US, UK, France and Germany. I thought it was time to update that post for countries other than the UK, with the UK included for comparison and with Italy added for reasons that will become clear. I also want to discuss in general terms how central banks should deal with the problem of knowing when to stop raising interest rates, now that the Fed has paused its increases, at least for now.

How to set interest rates to control inflation

This section will be familiar to many and can be skipped.

If there were no lags between raising interest rates and their impact on inflation then inflation control would be just like driving a car, with two important exceptions. Changing interest rates is like changing the position of your foot on the accelerator (gas pedal), except that if the car’s speed is inflation then easing your foot off the pedal is like raising rates. So far so easy.

Exception number one is that, unlike nearly all drivers who have plenty of experience driving their car, the central banker is more like a novice who has only driven a car once or twice before. With inflation control, the lessons from the past are few and far between and are always approximate, and you cannot be sure the present is the same as the past. Exception number two is that the speedometer is faulty, and erratically wobbles around the correct speed. Inflation is always being hit by temporary factors, so it’s very difficult to know what the underlying trend is.

If driving was like this, the novice driver with a dodgy speedometer should drive very cautiously, and that is what central bankers do. Rapid and large increases in interest rates in response to increases in inflation might slow the economy uncomfortably quickly, and may turn out to be an inappropriate reaction to an erratic blip in inflation. So interest rate setters prefer to take things slowly by raising interest rates gradually. In this world with no lags our cautious central banker would steadily raise interest rates until inflation stopped increasing for a few quarters. Inflation would still be too high, so they might raise interest rates once or twice again to get inflation falling, and as it neared its target cut rates to get back to the interest rate that kept inflation steady. [1]

Lags make the whole exercise far more difficult. Imagine driving a car, where it took several minutes before moving your foot on the accelerator had a noticeable impact on the car’s speed. Furthermore when you did notice an impact, you had little idea whether that was the full impact or there was more to come from what you did several minutes ago. This is the problem faced by those who set interest rates. Not so easy.

With lags, together with little experience and erratic movements in inflation, just looking at inflation would be foolish. As interest rates largely influence inflation by influencing demand, an interest rate setter would want to look at what was happening to demand (for goods and labour). In addition, they would search for evidence that allowed them to distinguish between underlying and erratic movements in inflation, by looking at things like wage growth, commodity prices, mark-ups etc.

Understanding current inflation

There are essentially two stories you can tell about recent and current inflation in these countries, as Martin Sandbu notes. Both stories start with the commodity price inflation induced by both the pandemic recovery and, for Europe in particular, the war in Ukraine. In addition the recovery from the pandemic led to various supply shortages.

The first story notes that it was always wishful thinking that this initial burst of inflation would have no second round consequences. Most obviously, high energy prices would raise costs for most firms, and it would take time for this to feed through to prices. In addition nominal wages were bound to rise to some extent in an attempt to reduce the implied fall in real wages, and many firms were bound to take the opportunity presented by high inflation to raise their profit margins (copy cat inflation). But just as the commodity price inflation was temporary, so will be these second round effects. When headline inflation falls as commodity prices stabilise or fall, so will wage inflation and copy cat inflation. In this story, interest rate setters need to be patient.

The second story is rather different. For various (still uncertain) reasons, the pandemic recovery has created excess demand in the labour market, and perhaps also in the goods market. It is this, rather than or as well as higher energy and food prices, that is causing wage inflation and perhaps also higher profit margins. In this story underlying inflation will not come down as commodity prices stabilise or fall, but may go on increasing. Here interest rate setters need to keep raising rates until they are sure they have done enough to eliminate excess demand, and perhaps also to create a degree of excess supply to get inflation back down to target.

Of course reality could involve a combination of both stories. In last year’s post I put this collection of countries into two groups. The US and UK seemed to fit both the first and second story. The labour market was tight in the US because of a strong pandemic recovery helped by fiscal expansion, and in the UK because of a contraction in labour supply partly due to Brexit. In France and Germany the first story alone seemed more likely, because the pandemic recovery seemed fairly weak in terms of output (see below). 

Evidence

In my post two weeks ago I included a chart of actual inflation in these five countries. Here is a measure of core inflation from the OECD that excludes all energy and food, but does not exclude the impact of (say) higher energy prices on other parts of the index because energy is an important cost.

Core inflation is clearly falling in the US (green), and rising in the UK (red). In Germany (light blue) core inflation having risen seems to have stabilised, and the same may be true in France and Italy very recently. The same measure for the EU as a whole (not shown) also seems to have stabilised.

If there were no lags (see above) this might suggest that in the US there is no need to raise interest rates further (as inflation is falling), in the UK interest rates do need to rise (as they did last month), while in the Eurozone there might be a case for modest further tightening. However, once you allow for lags, then the impact of the increases in rates already seen has yet to come through, so the case for keeping US rates stable is stronger, the case for raising UK rates less clear (the latest MPC vote was split, with 2 out of 7 wanting to keep rates unchanged) , and the case for raising rates in the EZ significantly weaker. (The case against raising US rates increases further because of the contribution of housing, and falling wage inflation.)

As we noted at the start, because of lags and temporary shocks to inflation it is important to look at other evidence. A standard measure of excess demand for the goods market is the output gap. According to the IMF, their estimate for the output gap in 2023 is about 1% for the US (positive implies excess demand, negative insufficient demand), zero for Italy, -0.5% for the UK (and the EU area as a whole), and -1% for Germany and France. In practice this output gap measure just tells you what has been happening to output relative to some measure of trend. Output compared to pre-pandemic levels is strong in the US, has been pretty strong in Italy, has been quite weak in France, even weaker in Germany and terrible in the UK (see below for more on this).

I must admit that a year ago this convinced me that interest rate increases were not required in the Eurozone. However if we look at the labour market today things are rather different. Ignoring the pandemic period, unemployment has been falling steadily since 2015 in both Italy and France, and for the Euro area as a whole it is lower than at any time since 2000. In Germany, the US and UK unemployment seems to have stabilised at historically low levels. This doesn’t suggest insufficient demand in the labour market in the EZ. Unemployment data is far from an ideal measure of excess demand in the labour market, so the chart below plots another: employment divided by population, taken from the latest IMF WEO (with 23/24 as forecasts).

Once again there is no suggestion of insufficient demand in any of these five countries. (The UK is the one exception, until you note how much the NHS crisis and Brexit have reduced the numbers available for work since the pandemic.)

This and other labour market data suggests our second inflation story outlined in the previous section may not just be true for the US and UK, but may apply more generally. It is why there is so much focus on wage inflation in trying to understand where inflation may be heading. Of course a tight labour market does not necessarily imply interest rates need to rise further. For example in the US both wage and price inflation seem to be falling despite a reasonably strong labour market, as our first inflation story suggested they might. The Eurozone is six months to a year behind the US in the behaviour of both price and wage inflation, but of course interest rates in the EZ have not risen by as much as they have in the US.

Good, bad and ugly pandemic recoveries

The chart below looks at GDP per capita in these five countries, using the latest IMF WEO for estimates for 2023.

Initially I will focus on the recovery since the pandemic, so I have normalised all series to 100 in that year. The US has had a good recovery, with GDP per capita in 2023 expected to be five percent above pre-pandemic levels. So too has Italy, which is forecast to do almost as well. This is particularly good news given that pre-pandemic levels of GDP per capita were below levels achieved 12 years earlier in Italy.

Germany and France have had poor recoveries, with GDP per capita in 2023 expected to be similar to 2019 levels. The UK is the ugly one of this group, with GDP per capita still well below pre-pandemic levels, something I noted in my post two weeks ago. Unlike a year ago, there is no reason to think these differences are largely caused by excess demand or supply, so it is the right time to raise the question of why there has been such a sharp difference in the extent of bounce back from Covid. To put the same point another way, why has technical progress apparently stopped in Germany, France and the UK since 2019.

Part of the answer may be that this reflects long standing differences between the US and Europe. Here is a table illustrating this.

Real GDP per capita growth, average annual rates	2000/1980	2007/2000	2019/2007	2023/2019
France	1.8	1.2	0.5	0.1
Germany	1.8	1.4	1.0	-0.1
Italy	1.9	0.7	-0.5	0.8
United Kingdom	2.2	1.8	0.6	-0.7
United States	2.3	1.5	0.9	1.1

Growth in GDP per capita in the US has been significantly above that in Germany, France or Italy since 1980. At least part of that is because Europeans have chosen to take more of the proceeds of growth in leisure. However this difference is nothing like the gap in growth that has opened up since 2019. (I make no apology in repeating that growth in the UK, unlike France or Germany, kept pace with the US until 2007, but something must have happened after that date to reverse that.)

I have no idea why growth in the US since 2019 has been so much stronger than France or Germany, but only a list of questions. Is the absence of a European type furlough scheme in the US significant? Italy suggests otherwise, but Italy may simply have been recovering from a terrible previous decade. Does the large increase in self-employment that occurred during the pandemic in the US have any relevance? [1] Or are these differences nothing to do with Covid, and instead do they just reflect the larger impact in Europe of higher energy prices and potential shortages due to the Ukraine war. If so, will falling energy prices reverse these differences?

[1] If wage and price setting was based on rational expectations the dynamics would be rather different.

[2] Before anti-lockdown nutters get too excited, the IMF expect GDP per capita in Sweden to be similar in 2023 to 2019.

Fiscal policy remains in the stone age

Or maybe the middle ages, but certainly not anything more recent than the 1920s. Keynes advocated using fiscal expansion in what he called a liquidity trap in the 1930s. Nowadays we use a different terminology, and talk about the need for fiscal expansion when nominal interest rates are stuck at the Zero Lower Bound or Effective Lower Bound. (I slightly prefer the latter terminology because it is up to central banks to decide at what point reducing nominal interest rates further would be risky or counterproductive.) The logic is the same today as it was in the 1930s. When monetary policy loses its reliable and effective instrument to manage the economy, you need to bring in the next best reliable and effective instrument: fiscal policy.

The Eurozone as a whole is currently at the effective lower bound. Rates are just below zero and the ECB is creating money for large scale purchases of assets: a monetary policy instrument whose impact is much more uncertain than interest rate changes or fiscal policy changes (but certainly better than nothing). The reason monetary policy is at maximum stimulus setting is that Eurozone core inflation seems stuck at 1% or below. Time, clearly, for fiscal policy to start lending a hand with some fiscal stimulus.

Yet the goal of the new German Finance minister, from the supposedly left wing Social Democrats, is to achieve a budget surplus of 1%. To achieve that he is cutting public investment from 37.9 billion euros in the coming year to 33.5 billion euros by 2020. Yet German infrastructure, once world renowned, is falling apart. Its broadband connectivity could be greatly improved.

The macroeconomic case for a more expansionary German fiscal policy is overwhelming. Germany has a current account surplus of around 8% of GDP. There are some structural reasons why you might expect some current account surplus in Germany, but the IMF estimates that these structural factors account for less than half of the current surplus. It estimates that a third of the excess surplus is a result of an overly tight fiscal policy. As Guntram Wolff points out, the main counterpart to the surplus is saving by the corporate sector. Perhaps more public investment might encourage additional private investment.

But this is not another article about how Germany needs to expand to help the rest of the Eurozone. The problem, as Matthew Klein points out, is that the whole of the Eurozone is doing the same. In the area as a whole, the fiscal position is as tight as it was in the pre-crisis boom. Unemployment in the Eurozone is still too high. And the reason fiscal policy is too tight is that key Eurozone policymakers think that is the right thing to do. “The right deficit is zero” says the French finance minister. He goes on: “ Since France is not in an economic crisis, we need to have a balanced budget, so that we can afford a deficit in tougher times.” You hear the same in Germany: the economy is booming so we must have budget surpluses.

A booming economy is not one that is growing fast, but is one where the level of output and employment is above the level compatible with staying at target inflation. Measures of the output gap are only estimates of what that level is: underlying inflation is the ultimate guide. Core inflation is well below target right now, which is why interest rates are at their effective lower bound. This is why the actions and rhetoric of most European (and UK) finance ministers are simply wrong.

You would think that causing a second recession after the one following the GFC would have been a wake up call for European finance ministers to learn some macroeconomics. (Yes, I know that the ECB raising rates in 2011 did not help, but I expect most macro models will tell you the collective fiscal contraction did most of the harm.) Yet what little learning there has been is not to make huge mistakes but only large ones: we should balance the budget when there is no crisis.

This is not a dispute between left and right as it is now in the UK, but a problem with the policy consensus in Europe. What we are seeing I suspect is a potent combination of two forces: a German obsession with balancing the budget which has it roots in currently dominant ordoliberal/neoliberal ideology, and Keynes famous practical men: advisers who learnt what economics they have in an era of the great moderation where the worst economic problem we had was relatively benign deficit bias. Fighting the last war and all that.

Some thoughts about the Job Guarantee

Mainly for economists

The idea of the state stepping in during a recession to offer some group of the unemployed a job was selectively adopted by the UK Labour government in 2009: see here by Paul Gregg. Richard Layard has proposed it for the long term unemployed. We can think of both schemes as providing a partial insurance policy against the failure of countercyclical stabilisation policy to completely do its job in a downturn/recession. But MMT (Modern Monetary Theory) economists go beyond that to suggest that it could be a permanent feature, which would eliminate involuntary unemployment (IU) without creating inflation.

Given our recent experience, the use of a Job Guarantee as an insurance policy for all unemployed people during a recession seems like a good idea. Unwanted leisure is replaced by labour producing useful output, and paying a wage greater than unemployment benefits would add to automatic stabilisers. The devil is of course in the detail: JG jobs need to be setup to allow job search (many jobs are created even in a recession) or (if necessary) retraining and the JG jobs would need to involve an output that was socially useful [1]. The main problem that Gregg discusses in his paper is the ‘Lock-in’ effect, where those in a JG job reduce their search activity. To combat that and for other reasons it also seems helpful to provide detailed individual advice along the lines of the Swedish scheme described here.

MMT economists have suggested extending such a scheme so that it operates at all times. As I understand it any worker without a job would be offered a JG job. It can be refused with no consequences in terms of unemployment benefit, so it that sense it is not workfare. Many people who were confident of getting another job quickly might want to focus all their spare time on job search, and so might decline a JG job. Anyone who wanted a job could receive one, so the scheme would largely eliminate involuntary unemployment (IU). [2] In the rest of this post I’m going to focus on this idea of JG as being a permanent feature of an economy, rather than just something put in place during an economic downturn.

A key issue is what the JG wage would be. In most MMT literature I have seen, the JG wage would be the minimum wage or better: see Mitchell here for example. There seems to be an obvious consequence of this. Unfortunately many private and public sector jobs are paid the minimum wage. These jobs are risky whereas JG jobs are by definition permanent. Minimum wage non-JG jobs may have compensating advantages like a career structure, but still it would seem probable that the existence of JG jobs paying the minimum wage would attract some workers from private sector minimum wage jobs.

The obvious response would be for private and public sector employers paying minimum wages to increase their pay sufficiently to stop this happening, which in turn would often lead private sector firms to raise prices. How far this ripples through the economy is not certain [3], but it is quite possible that the overall price level rises by a noticeable amount. This higher aggregate price level would reduce the real value of the JG wage. If this reduced wage differentials in the economy as a whole this process might be regarded by some as beneficial, but it is an implication that JG advocates need to acknowledge.

A perhaps more serious concern is the impact of the JG on inflation. What is conventionally believed to prevent policy makers expanding demand sufficiently to eliminate all IU is that to do so would embolden workers to ask for greater pay increases, generating an inflationary spiral. The existence of IU, and the possibility of joining their number. becomes a threat that keeps inflation stable. In a JG economy that threat is greatly reduced, both because an alternative job is always available and it will pay more than unemployment benefit. (JG and the lock-in effect will also reduce geographic mobility, although the other side of that coin is that joblessness would not be a feature of deindustrialisation.)

Suppose we start with an economy with stable inflation, implying unemployment was at the NAIRU, and introduce JG.. As this puts upward pressure on inflation because the costs of losing a job are reduced. the only way of keeping inflation stable is to deflate demand, which of course would reduce output, labour demand and therefore increase the number of people on JG jobs. So if we were to compare two economies where inflation was stable, one with IU and one with JG, the number of JG jobs would exceed IU in the other economy.

That does not mean that output would necessarily be lower in the JG economy, because JG workers are producing some kind of socially useful output while the IU workers are not. In welfare terms you have also eliminated any non-pecuniary costs associated with spells of unemployment, and the distribution of income in the JG economy is more equal than in the IU economy. However in practice the productivity of JG workers will be pretty low, as they need to be allowed time for intensive job search and the turnover in JG jobs is likely to be high. We have a trade-off, and if anyone can point me to any analysis of this particular trade-off I would be very grateful.

Can I end with a personal plea. When I write things like this it is often assumed by MMTers that I am being critical for the sake of it. In other words they think all I want to do is attack the JG or MMT. I don’t. I have far better things to do with my time. I actually find the idea of JG appealing at an intuitive level. More generally I agree with MMT on many things, although not all. But I am also fed up with policy makers implementing bad policies just because they sound good to those policy makers, so I want to subject any policy I intuitively like to rigorous analysis.

[1] JG jobs could be as assistants in police stations, schools and other public sector institutions. Or they could be jobs in social enterprises.

[2] Keynes defined involuntary unemployment as those seeking a job at the going real wage. As JG jobs would be at the minimum wage it would not eliminate all involuntary unemployment defined in this way: as I noted those looking for a higher than minimum wage job who chose to stay unemployed would still technically be classed as involuntarily unemployed. But this is being a little pedantic.

[3] Mosler and Silipo (section 7) talk about the JG wage as a nominal anchor. This captures the idea that movements in the JG wage would influence other wages. However nominal anchors, like the money supply or the exchange rate, are often talked about as being able to control the aggregate price level in the longer term on their own. The JG wage would not be able to do this. As the authors note, active stabilisation policy would still be required to do this, although the number of JG jobs could be a useful indicator of what action was required, just as the unemployment rate is now. Another way of saying the same thing is that the JG does not supplant the need for active macroeconomic stabilisation.

The Brexit interest rate increases and misunderstanding inflation

Last week’s rise in UK rates has been extensively analysed (see for example Tony Yates here) so I will be very selective. First, the justification for the title of this post is provided by an extract from the inflation report:

“The overshoot of inflation throughout the forecast predominantly reflects the effects on import prices of the referendum-related fall in sterling. Uncertainties associated with Brexit are weighing on domestic activity, which has slowed even as global growth has risen significantly. And Brexit-related constraints on investment and labour supply appear to be reinforcing the marked slowdown that has been increasingly evident in recent years in the rate at which the economy can grow without generating inflationary pressures.”

The last sentence is particularly important: in plain language it is saying that Brexit is contributing to lower trend productivity growth, which the Bank now put at 1.5% compared to a pre-recession level of 2.25%. The wording is chosen carefully: they are not talking about uncertainty effects, but permanent effects from a likely deal. So last year worries about the demand side effects of Brexit led the Bank to reduce rates, and now concerns about the supply side effects of Brexit are contributing to higher rates.

Whether these modest increases in interest rates continue, as the Bank are signalling, should largely depend on whether the pickup in earnings growth they anticipate actually happens. As Torsten Bell from the Resolution Foundation argues here, the set of information that might justify the Bank’s expectations of an imminent recovery in earnings growth is not empty, but nevertheless many economists regard it as a brave forecast.

However the labour market is not the only reason the Bank is raising rates. Putting labour market issues aside, they think that because firms are operating with little ‘spare capacity’, any large increases in demand will be met by firms raising prices. Ergo the Bank’s job is to use higher interest rates to stop demand rising too fast. I think this is conceptually wrong, because it underestimates the role that demand and expectations about demand play in determining investment decisions.

A firm can meet rising demand in three ways: by investing in more productive processes, by raising prices or by using more of its spare capacity. In a traditional economic upswing firms first use spare capacity, then invest, and when capacity utilisation is at a peak and there are no profitable investments to make it raises prices. At that point it is right for a central bank to step in to moderate demand growth.

This has not been a typical recovery from a recession. Firms have used up spare capacity, but have not invested in more efficient processes. This is what measures of capacity utilisation suggest (taken from an earlier Bank of England Inflation Report).

If you just take these surveys as measuring the state of the cycle (and if we ignore the Bank Agents) since 2013 the economy has been experiencing an economic boom. Yet from 2013 core inflation has been below target and falling. You can resolve this paradox by thinking about firms acting unusually, by failing to invest and meeting additional demand by utilising capacity as if they are in a boom The result of that is stagnant productivity growth.

The conceptual error is to read these capacity utilisation numbers as indicating that there are no profitable investments to make. We know these profitable investments exist, because leading firms are improving their productivity.* What we have is an innovations gap, where lagging firms are not copying leading firms and are instead holding back on investing. We do not know why they are holding back, but one obvious reason is they have expectations of low future growth and/or high uncertainty about this growth. Empirical evidence shows the strongest determinant of investment is output growth, and the obvious rationalisation for that ‘accelerator effect’ is that current growth influences expectations about future growth.

If this is right, increases in demand will be met by firms finally coming off the fence and investing, rather than raising prices. But if the central bank starts raising interest rates to choke off demand, even when it is growing slowly by historical standards, it will validate the pessimism that has been holding back investment and productivity will continue to stagnate. There is a very real danger that the Bank may be playing its part in a self-fulfilling low growth recovery.

*Postscript (8/11/17) Discussion here by Berlingieri et al shows this growing divergence between leading and lagging firms is a global phenomenon.

The F story about the Great Inflation

Here F could stand for folk. The story that is often told by economists to their students goes as follows. After Phillips discovered his curve, which relates inflation to unemployment, Samuelson and Solow in 1960 suggested this implied a trade-off that policymakers could use. They could permanently have a bit less unemployment at the cost of a bit more inflation. Policymakers took up that option, but then could not understand why inflation didn’t just go up a bit, but kept on going up and up. Along came Milton Friedman to the rescue, who in a 1968 presidential address argued that inflation also depended on inflation expectations, which meant the long run Phillips curve was vertical and there was no permanent inflation unemployment trade-off. Policymakers then saw the light, and the steady rise in inflation seen in the 1960s and 1970s came to an end.

This is a neat little story, particularly if you like the idea that all great macroeconomic disasters stem from errors in mainstream macroeconomics. However even a half awake student should spot one small difficulty with this tale. Why did it take over 10 years for Friedman’s wisdom to be adopted by policymakers, while Samuelson and Solow’s alleged mistake seems to have been adopted quickly? Even if you think that the inflation problem only really started in the 1970s that imparts a 10 year lag into the knowledge transmission mechanism, which is a little strange.

However none of that matters, because this folk story is simply untrue. There has been some discussion of this in blogs (by Robert Waldmann in particular - see Mark Thoma here), and the best source on this is another F: James Forder. There are papers (e.g. here), but the most comprehensive source is now his book, which presents an exhaustive study of this folk story. It is, he argues, untrue in every respect. Not only did Samuelson and Solow not argue that there was a permanent inflation unemployment trade-off that policymakers could exploit, policymakers never believed there was such a trade-off. So how did this folk story arise? Quite simply from another F: Friedman himself, in his Nobel Prize lecture in 1977.

Forder discusses much else in his book, including the extent to which Friedman’s 1968 emphasis on the importance of expectations was particularly original (it wasn’t). He also describes how and why he thinks Friedman’s story became so embedded that it became folklore. The reason I write about this now is that I’m in the process of finishing a paper on the knowledge transmission mechanism and the 2010 switch to austerity, and I wanted to look back at previous macroeconomic crises.

If it wasn’t a belief in a long run inflation unemployment trade-off, what was it that allowed inflation to gradually rise during those two decades? Forder has a lot to say on this, but the following is my own take. I think two things were critical: the idea that demand management was primarily designed to achieve full employment, and that full employment had primacy over the objective of price stability. Although more and more economists over that period began to see the policy problem within a Phillips curve framework, many still hoped that other measures like prices and incomes policies (in the UK in particular but also in the US) could override the Phillips curve logic. The primacy of the full employment objective meant the problem was often described as ‘cost-push inflation’ rather than a rise in the natural rate of unemployment.

If you find this hard to imagine, think about historians discussing the current period in a possible future in 2050. By then nonlinearities in the Phillips curve and the power the inflation target had in anchoring inflation expectations were firmly entrenched in mainstream thinking. Imagine that partly as a result in 2050 the inflation target has been replaced by a level of nominal income target. With the benefit of hindsight these historians were amazed to calculate the extent to which resources were lost decades earlier because policy had become fixated by a 2% inflation target and budget deficits. They will recount with amusement at the number of economists and policymakers who thought that the way to deal with deficient demand was by ‘structural reform’. Rather than construct folk tales, they will observe that even when most economists realised what was required to avoid being misled again policymakers were extremely reluctant to change the inflation target.

Zero UK Inflation

Today it was announced that UK consumer price inflation hit zero in February. The ONS estimate that we have to go back to the 1960s for the last time this happened. More importantly, core inflation fell back 0.2% to 1.2%, after 0.1% increases in the previous two months. As Geoff Tily points out, if you take out the 0.2% contribution from the decision to raise student fees (which are hardly an indicator of excess demand), then the Governor could be writing a letter to the Chancellor based on core inflation, and not just the actual inflation rate.

The Chancellor is in election mode and so does not care: in fact he says zero inflation is good news, and he just hopes no one asks him why he chose to reaffirm a symmetrical 2% target. For the Bank of England it means the key question is now should they cut rates? As I noted here, optimal control exercises on the Bank’s model and forecast say they should, and I discussed here why there is an additional strong prudential case for doing so.

The point I want to make now is about survey evidence on capacity utilisation. I had a number of memorable meetings with various economists and officials at the Treasury, Bank and elsewhere when the Great Recession was at its height, but the one that left me most puzzled was with one of the more academic economists at the Bank of England. It was at about the time that core inflation started rising to above 2%, despite unemployment being very high and very little signs of a recovery. At much the same time survey measures of capacity utilisation were suggesting a strong recovery, completely at variance with the actual output data. The gist of our discussion was: what the hell is going on!? It was particularly puzzling for me, because I had many years before done a lot of work with survey data of this kind, and back then it seemed pretty reliable.

The problem with the 2010 period was that it was exceptional, and so in that sense it was not that surprising to see surprising things going on. My own pet theory at the time was that the financial crisis had made firms much more risk averse, which made them less likely to cut prices in an attempt to gain market share. Move on to today, and things are perhaps a bit less exceptional. What today’s figures emphasise is that the inflation ‘puzzle’ of 2010/11 has gone away. Levels of inflation are now much more consistent with substantial spare capacity in the economy. However the bizarre behaviour of the survey data has not disappeared.

Here is a nice chart from the ONS, comparing various different measures of spare capacity.

  

What it shows is that labour market indicators suggest an amount of spare capacity well outside the ‘normal’ range from the pre-recession years. In contrast, both hours worked and survey based indicators (the first six measures) suggest the output gap is quite small, and in one case actually positive. As this OBR paper shows, survey measures of capacity utilisation were suggesting a positive output gap as early as 2012.

Faced with this combination of spare capacity in the labour market and firms reporting its absence, a macroeconomist would suggest it could be a consequence of high real wages, encouraging substitution from labour to capital. Firms were fully utilising their capital – hence no spare capacity – but had hired less labour as a result. However a notable characteristic of this recession in the UK has been the high degree of labour market flexibility, with large falls in real wages. One of the more persuasive theories for the UK productivity puzzle, which I outlined here, is that we have seen factor substitution going in the other direction.

Back in 2010, I was reluctant to suggest the survey data were simply wrong, partly because of their past reliability but mainly because inflation was telling a similar story. On the day that inflation hits zero, I think the argument that these survey measures are not measuring what we used to think they measured has become much stronger. But that still leaves an unanswered question - why have they gone wrong, when they worked well in the past? 

Simplistic theories of inflation

After I wrote this I saw that Frances Coppola has a post that covers some of the same ground, but the point I want to make is different.

One of the things that made monetarism so popular until governments actually tried it was its simplicity. You can express this simplicity in many ways, but most involve the idea that there is a stable demand for the real value of money (M/p), so if you can control M you must control p. Never mind that the immediate influences on inflation were much more complicated: if you knew what M was, you would know what p would be. If you controlled M you would eventually control p.

There are lots of problems with this idea. I talked about the difficulty in explaining prices by just using money in this post. The difficulty of finding the ‘right’ definition of money is not a technical problem but a feature: because money can be saved as well as buy goods (the medium of exchange is also a store of value) focusing on its role in buying goods (‘hot potatoes’) is misleading. But even if there was a stable long run demand for money for some definition, the usefulness of this becomes questionable if we cannot say what the future quantity of money will be.

This becomes blindingly obvious if money is base money and we think about Quantitative Easing. Printing base money under quantitative easing does not imply hyperinflation because the expansion in the monetary base will be reversed once the recession is over. Knowing what base money is becomes useless as a tool for saying what future prices will be. (For those more technically minded who still think there is a Pigou effect, I discussed why the Pigou effect has disappeared from modern macro here. It is based on the same point.)

The Fiscal Theory of the Price Level is potentially another simplistic theory of inflation. This works from the identity that the real value of government debt must equal the discounted value of primary surpluses (taxes less government spending). It also can be used in a naive way: treat future primary surpluses as fixed, and any increase in nominal government debt must lead to higher prices. But, as Chris Sims explains in this nice exposition at Lindau, future primary surpluses are not fixed. If debt increases, future primary surpluses can increase to pay the interest on that additional debt, and more.

There may be some that say that we cannot trust politicians to do that. To which I say which planet have you been on for the last five years? As Brad DeLong reminds us for the US, this recession has been unusual in the zeal that governments have shown in rapidly reducing primary deficits, and of course in the Eurozone this zeal - embodied in the fiscal compact - has led to a second recession. Chris Sims raised the possibility that so great has this zeal been that even though nominal debt has risen, the price level might fall to make the identity hold.

One lesson I would draw from this is that the Fiscal Theory of the Price Level, like monetarism, is not a terribly helpful way of thinking about future inflation. The idea that we can take one variable, or one equation, and distil from that the future price level is a fantasy. What is surprising is that this fantasy has been, and still remains, so attractive for some economists.

Has the Great Recession killed the traditional Phillips Curve?

Before the New Classical revolution there was the Friedman/Phelps Phillips Curve (FPPC), which said that current inflation depended on some measure of the output/unemployment gap and the expected value of current inflation (with a unit coefficient). Expectations of inflation were modelled as some function of past inflation (e.g. adaptive expectations) - at its simplest just one lag in inflation. Therefore in practice inflation depended on lagged inflation and the output gap.

After the New Classical revolution came the New Keynesian Phillips Curve (NKPC), which had current inflation depending on some measure of the output/unemployment gap and the expected value of inflation in the next period. If this was combined with adaptive expectations, it would amount to much the same thing as the FPPC, but instead it was normally combined with rational expectations, where agents made their best guess at what inflation would be next period using all relevant information. This would include past inflation, but it would include other things as well, like prospects for output and any official inflation target.

Which better describes the data? The great attraction of the FPPC is that it can describe stagflation. We have a boom, which while it lasts steadily raises inflation. When the boom comes to an end, inflation stabilises, but at a much higher level than it began. So policy has to engineer a recession to get inflation back down again: a period in which above average unemployment is accompanied by above average inflation, which we call stagflation. If over this period we had had credible independent central banks setting inflation targets, the NKPC would not give us stagflation: when the boom came to an end, inflation would return to target. (For more explanation, see this post.) The trouble is we did not have inflation targeting during this period, so it is difficult to tell whether stagflation is evidence against the NKPC. (As an example of this ambiguity, see this survey of the empirical evidence by Nason and Smith. This enabled Robert Gordon to be quite supportive of the FPPC in 2009.)

The Great Recession could provide a much better test. In some countries output fell sharply in 2009, but has since seen a slow but steady recovery, such that the output gap today is less than it was in 2009. With the FPPC, inflation should have been steadily falling over this period, reaching its lowest level today. So if we plotted the output gap (x axis) and inflation (y axis) together, we should see a line pointing South East. With the NKPC, we can consider two polar cases. In the first, agents fully anticipate that the recovery will be slow, so we will get a sharp immediate fall in inflation, but subsequently inflation will rise towards the target. That will give us a line pointing North East. In the second, agents keep thinking inflation will return to target next year. That also gives us a line pointing North East, but it is flatter. [Postscript - I should have added that this last gives us what Krugman calls the Neo-paleo-Keynesian Phillips curve.]

Here is this plot for four countries, using OECD estimates for the output gap on the horizontal axis, consumer price inflation less 2% on the vertical axis, and OECD forecasts for 2014 and 2015. I’ve chosen these countries simply because in most of Europe we had a double dip recession, which is a more complicated experiment. If you do not like the idea of including forecasts, just ignore the last two points for each country.

The lines point North East, not South East. This gives more support to the rational expectations NKPC than the adaptive expectations FPPC. To take just one example, US inflation in 2013 is higher than it was in 2009, which is consistent with the NKPC. The traditional FPPC, on the other hand, would suggest that after a string of negative output gaps, US inflation should be a lot lower in 2013 than it was in 2009.

OK, now the caveats. Commodity prices will have an important influence on the CPI, and these are not part of either simple Phillips curve story. They may help explain the blip in inflation around 2011 in some countries, but they also helped depress inflation in 2009. Exchange rate changes will also matter. The simple Phillips curve also takes no account of non-linearity caused by a reluctance to cut nominal wages. And of course estimates of the output gap may be wrong.

In the case of Japan, we also had a recent increase in the inflation target. This may explain the forecast upward shift in inflation in 2014/5. If it does, that is clear evidence in favour of rational rather than adaptive expectations. 

All these caveats point to the need to do more empirical analysis. Nevertheless we can see why some more elaborate studies (like this for the US) can claim that recent inflation experience is consistent with the NKPC. It seems much more difficult to square this experience with the traditional adaptive expectations Phillips curve. As I suggested at the beginning, this is really a test of whether rational expectations is a better description of reality than adaptive expectations. But I know the conclusion I draw from the data will upset some people, so I look forward to a more sophisticated empirical analysis showing why I’m wrong.

Inflation risks

When it comes to the issue of when interest rates should start rising, one of the points I and others have often made is that the risks are not symmetric. If inflation starts rising faster than we expect monetary policy can quickly respond. Alternatively if we actually have more ‘spare capacity’ than we currently believe, it may take some time for this to become apparent (inflation is more sticky when it gets low), and the zero lower bound limits what monetary policy can do.

In this light, the following table from the Bank of England’s inflation report issued today is rather interesting. 

In scenario 1, there is more slack in the labour market than the Bank currently thinks. In scenario 2, firms are currently working at a higher rate of capacity utilisation than the Bank estimates. In both scenarios monetary policy is endogenous. Inflation is higher in scenario 2, but monetary policy succeeds in getting it back to target by the middle of 2017. In scenario 1, inflation remains below target throughout the period.

Now as a macroeconomist I really want to know more. These are different shocks, so they are not a pure test of asymmetry. The path of interest rates is not shown, so we do not know how much of a constraint the zero lower bound is (if at all) in scenario 1. In scenario 1 unemployment falls quite a lot more than in the central projection, but the additional GDP growth seems small by comparison. The opposite is true in scenario 2. This is undoubtedly a result of different shocks being applied, but all the report tells us is that judgement was used in deciding how to shock the model to best capture each scenario. As I am sure Tony Yates would say, it would be nice for those who are interested in this detail to know a bit more.

However, putting all these qualifications aside, these simulations are welcome given all the discussion underway. They suggest that the upside risks to inflation are small, because monetary policy is capable of responding quickly. Overestimating the degree of current slack does not lead to inflation ‘taking off’. Perhaps this helps to explain why the Bank appears to some to be rather relaxed about the need to raise rates.     

Hawkery, or is the Bank biased

An interesting contrast in my evening reading yesterday. In the US, Matt O’Brien in the Washington Post’s Wonkblog making fun of reporting that inflation is just around the corner. There is one particularly nice line: “Well, there's always demand for pieces about why we need to raise rates — mostly from 60-year-olds who think it's always 1979 …” The contrast is with the Financial Times’s Chris Giles, who in yesterdays FT accuses the Bank of England of ‘institutional dovishness’, which he compares to institutional racism. The Bank is “institutionally biased against higher interest rates.”

Now, lest I be misunderstood, let me say three things before addressing Chris Giles’s charge. First, I’m pretty sure Chris is well short of 60. Second, Chris is no fool who blindly follows some party line: this piece on the Treasury’s exercise in dynamic scoring is as good as economic journalism gets. Third, central banks can suffer from what I and others prefer to describe as ‘groupthink’. Laurance Ball argued that this happened at the Fed when it came to not trying what I call forward commitment (promising higher inflation and a positive output gap in the future to combat the zero lower bound).

Having said that, Chris can occasionally pursue a line that, while popular in some quarters, makes little macroeconomic sense. The idea that UK austerity did not matter much had him clash with not just the usual suspects (including me), but also US academics Alan Taylor and Oscar Jorda. (I discussed an earlier version of their paper here: their latest version is here). In a similar way, over the last few months Chris has relentlessly pursued the idea that UK interest rates should rise very soon.

Chris’s charge against the Bank is that they keep moving the goalposts. For example, they say they will think about raising interest rates when unemployment dips below 7%, but when unemployment does go below 7% they decide that there is no reason to raise rates, and so on. But for the Bank the goalposts are the inflation target, and inflation is below target.

In the past I have made the point that, given uncertainties about the size of the output gap, it is best to err on the side of expansionary policy. This is because the Bank can easily deal with inflation if it does begin to rise, but because of the lower bound the opposite is not true. Chris responds that “no one should expect that an overheating economy will quickly set prices and wages on the climb”. “As the pre-crisis period showed, economies can overheat and develop dangerous imbalances without displaying the usual warning sign of inflation.” He is of course talking about house prices. But raising rates is a very inefficient way of dealing with a housing boom, which is why we now have the Financial Policy Committee at the Bank with its macroprudential tools. It is also very inefficient for the Bank to be trying to undo effects caused by the Chancellor’s policies (Help to Buy).

To see what can happen in this situation, we just need to look at Sweden. Sweden raised interest rates from almost zero to 2% beginning in 2010, because they were worried about overheating in the housing market. They now have deflation: inflation was -0.6% in March. As a result, the central bank has had to bring interest rates back down again. Lars Svensson, one of the world’s leading macroeconomists who resigned from their equivalent of the MPC while this happened, can only say I told you so.

While we are on the subject of premature interest rate increases, let us not forget the ECB raising rates just before the second Eurozone recession. And let us not also forget that the MPC almost followed their lead - not much evidence of institutional dovishness there.

I suspect and hope that the Bank and MPC have their eyes on the big picture. UK GDP per capita is currently around 15% below the level we might have expected it to be at if it had followed pre-recession trends. At no time since WWII has the economy not come back to this trend. We have no even half decent theories about why this trend should have dramatically changed. In these circumstances, starting to put on the brakes when we have only just begun to catch up lost ground, and when inflation is below target, just seems dumb and dangerous. 

New versus traditional Phillips curves and the Great Recession

For economists

One of the questions I like asking students is whether inflation following the Great Recession has tended to favour the New Keynesian (NK) Phillips curve or its more traditional counterpart (TK). I like it because it allows me to draw a nice diagram, and also because it shows students how difficult it is to discriminate between theories in macro.

So first the theory. The two competing models are

• NK: Inflation at t = expected inflation at t+1 together with a term in the output gap
• TK: inflation at t = inflation at t-1 together with a term in the output gap

I’m ignoring discounting in the NK Phillips curve for simplicity. Assume expectations about inflation are rational, and suppose the economy is hit by an unexpected recession of known size and duration. The two models predict the following:

With the traditional model, inflation gradually falls as the recession continues, and once it comes to an end, inflation remains lower. In the New Keynesian model, assuming that the inflation target is credible, inflation jumps down when the unexpected recession occurs, and then inflation gradually rises towards its target as the recession progresses. (We assume here that the output gap is constant while the recession lasts, again for simplicity.) For the NK model, it is critical in drawing this diagram that the extent of the recession is known – more on this below. The patterns implied by the two models are distinct, and this difference is likely to persist even if each curve becomes flatter as we approach zero inflation because of nominal wage rigidities.

To see what has actually happened, see this nice post from Gavyn Davies. The immediate aftermath of the recession looked more like the NK model: a sharp fall followed by a gradual rise. Furthermore I would argue that – once the recession hit – most people expected it to be large and persistent, so my diagram is not totally unrealistic. But if we look at what has been happening in the last two years, it looks much more like the TK model, with inflation gradually falling below target.

That is probably as far as we should go without doing some econometrics, and also taking account of some of the complexities discussed here. We could probably get any pattern to fit the NK model by imagining a suitable sequence of expectations errors. In addition if we are looking at consumer price inflation we should account for commodity price changes, which neither model does. (If we look at GDP deflators, you could tell a story where agents were initially expecting a recession lasting three or four years, and have been surprised that the recession has persisted ever since.) That is why some proper econometrics is required, preferably looking at both price and wage inflation together with expectations data. (If such studies have been done, please let me know.)

However perhaps I can suggest two possible conclusions that such studies could test more rigorously. First, the traditional Phillips curve, where expectations are implicitly naive and backward looking, does not look like a promising basis for explaining inflation following the recession. Either the New Keynesian model, or some combination of the two models, looks more like providing an adequate foundation for a reasonable explanation. Second, an explanation based on the NK model that treats the size and extent of the recession (whatever that turns out to be) as one initially unexpected but then completely anticipated shock is also going to struggle to fit the data.