The Output Gap is no longer a sufficient statistic for inflationary pressure

One of the features of the latest OBR forecast is that they believe the economy is operating slightly above its sustainable level (a positive output gap), where the sustainable level is the level that would keep inflation constant. To see how startling that hypothesis is, here is the latest version of a chart I have probably posted more than any other since I started writing this blog.

It is UK GDP per head (source), which is a pretty good measure of average prosperity, and a trend line in red for 2.23% growth p.a. So from 1955 to 2007 prosperity grew at an average rate of almost two and a quarter percent each year. Since then it has increased at an annual rate of around 0.35%. And if the OBR are right, none of this is due to unutilised resources and lack of demand.

The shift in trend is just as clear if we look at output per worker. Some people try and rationalise this by saying that 2007 was a boom year, and so trend growth had really been falling long before the Global Financial Crisis (GFC). But the evidence does not support more than a slight downward shift in the growth trend before the GFC: the OBR estimate an output gap of 0.7% in 2006/7 and 1.8% in 2007/8.

I find it extraordinary that most economists still talk about the output gap after the GFC in the same way that it was talked about before the crisis: as a limit to how far and fast the economy can expand. To do that is in my view quite wrong. It ignores what I call the innovations gap: the difference between actual output and the level of output that firms could achieve if they started using the best technology available to them. Because there is currently a large innovations gap, firms are likely to meet additional demand not be raising prices but by investing in these more efficient techniques.

Before the GFC, we could ignore the innovation gap because it was relatively small. But since the crisis that gap for the UK and many other countries must have increased, because it is simply not plausible to assume that since the GFC technical progress has come to a virtual halt. Innovations may not have been increasing at the pre-GFC rate, but they cannot have almost stopped, which is the implicit assumption in the OBR’s analysis. Hence we have in the UK, and I suspect in many other countries, a subsrantial innovations gap which will prevent any excess demand leading to significant inflationary pressure. Some supporting evidence for this comes from the growing productivity divergence between leading firms and the rest.

Why have most firms not been investing in the most productive equipment and techniques since the GFC? I think the simple answer is fixed costs and demand. Investment projects almost always involve a large fixed cost element (disruption, retraining), and with static demand those fixed costs may exceed any efficiency gain. But in a normal recovery from a recession, where demand is growing rapidly, firms are happy to incur that fixed cost because they need to expand capacity anyway to meet growing demand. In a weak recovery, on the other hand, many firms may not need to expand capacity, with any modest increases in demand going to leading firms, firms that do invest in the latest technology. Hence the divergence noted above.

Exactly the same argument applies to the NAIRU: the level of unemployment at which inflation is constant. The NAIRU is almost certainly lower than most central banks think for a variety of reasons, but when it is approached I expected to see a pick up in investment and innovation more than a pick up in wage inflation. Investment and productivity growth go together, as a nice chart in the OBR’s latest forecast report shows (page 43).

A large innovation gap in the UK is being enhanced by Brexit. The more uncertain future demand is the more firms are likely to postpone productivity enhancing investment. It may be politically useful to delay creating a new customs union/SM for goods with the EU to try and keep the Conservative party together (as regular readers will know, I think this is inevitable because of the Irish border), but the uncertainty that delay creates just holds back UK growth. Just one more way in which both Brexit and more generally a Conservative government is an economically destructive project.

The existence of a large innovations gap, both in the UK and elsewhere, means that we need two things. First, we need a monetary policy that is very relaxed about raising interest rates. Second we need, in the UK and pretty well everywhere, a large increase in public sector investment. The first needs independent central banks to be less inflation averse and to stop treating the sustainable level of output as something which is independent of what they do. The second requires governments to stop being obsessed about deficits and instead to start investing in the future of all the people they govern.

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.

Productivity and monetary policy

The Bank are warning of imminent rises in interest rates. As Chris Giles points out, we have been here before, and before that, but that shouldn’t mean we should dismiss this talk, because one day it will happen. [1] They (the MPC) certainly sound serious. But why when current growth is so slow are they even contemplating it? Here is a clue from Mark Carney’s latest speech (my italics).

“On the supply side, the process of leaving the EU is beginning to be felt. Brexit-related uncertainties are causing some companies to delay decisions about building capacity and entering new markets. Prolonged low investment will restrain growth in the capital stock and increases in productivity. Indeed, if the MPC’s current forecast comes to pass, the level of investment in 2020 is expected to be 20% below the level which the MPC had projected just before the referendum. Net migration has also fallen by 25% since the Referendum.

As a result of these factors and the general weakness in UK productivity growth since the global financial crisis, the supply capacity of the UK economy is likely to expand at only modest rates in coming years.”

When people, like me, say how can the Bank be thinking of raising rates when demand is so weak, the response from the Bank would be that supply has been at least as weak.

This pessimism about the supply side comes straight from the data. If I hear people talking about the UK being a ‘strong economy’, I know they either have not seen this chart or are just lying.

UK Output per hour, whole economy (ONS)

The red line is a trend that pretty well matches the trend in the data until the end of 2007, with the amount you can produce with an hours worth of labour increasing by 2.2% a year. Since the global financial crisis (GFC) there has been almost no growth at all. If you want to know the main reason real wages have stopped increasing, this is it. [2]

I hear some people say this is just oil and financial services. It is not, as this table from a recent Andy Haldane speech shows.

Start at the bottom: total average growth has been non-existent since the crisis. The rest of the table looks at the contribution of each sector to that total. To see what productivity growth would be excluding financial services, just add that figure to the total: 1.8% 1998-2008, 0.4% 2009-2016. That table makes it clear that the productivity crisis is economy wide.

It is worth looking at aggregate productivity since the GFC period in more detail (same data). I often hear people say the productivity slowdown started before the GFC. From the chart below, it clearly did not. (We have just seen the tenth anniversary of Northern Rock going bust, and the UK productivity slowdown started shortly after that event.)

We could describe this data as five phases. 1) Productivity in the recession fell, as it often does in a recession for various reasons. 2) As the economy begins to grow again, so did productivity growth. 3) As it becomes clear, in 2011, that the ‘recovery’ is going to be very weak because of austerity, productivity growth stops growing. 4) By the end of 2013, with stronger growth under way (although still no catch up to previous trends, so not a true recovery) productivity starts growing again, although rather slowly. 5) Since the 2015 election, with the prospect and then the reality of Brexit, even that modest growth disappears. (My data does not include 2017Q2, which saw a very slight fall.) I could shorten the description as follows: recession, modest optimism, pessimism, even more modest optimism, uncertainty.

That is my gloss on the numbers, but I’ve done it to make a point. Productivity growth invariably requires an investment of some kind. It may not be physical investment, but just training someone up to be able to use some new software. Whether a firm incurs that cost will depend, in part, on their expectations about the future. There is a regrettable tendency in macro (I blame RBC theory) to treat productivity growth as manna from heaven. But the idea that potential improvements in technology stopped after the GFC, and just in the UK, is simply ridiculous. The problem is that firms are not investing in new technology. What I call the ‘innovations gap’ has emerged in the UK because of weak growth and the consequent pessimistic expectations of most firms. [3]

The idea that the economy could get itself in a low growth expectations trap is increasingly being put forward by economists: here is George Evans, for example. The UK has got itself into that trap because on the two occasions that a recovery of sorts appeared to be under way, the economy has been hit with terrible policy errors (austerity and Brexit). But the idea that UK firms are incapable of upgrading their production techniques is nonsense. They will do so initially if they can be confident that the demand for their products will increase, or subsequently when the innovation pays for itself even though demand is flat.

Which is why an increase in interest rates right now would be very bad news. It would confirm the pessimistic expectations of most firms that demand is not going to grow fast enough to make innovation worthwhile. Formally, the job of the MPC is not to worry about productivity but to control inflation. But elsewhere, where the same process may be happening to a lesser extent (the productivity slowdown is worldwide, just most acute in the UK), central banks are puzzled at why inflation just refuses to rise. 

The concept of an innovations gap is one solution to that puzzle. Expanding demand allows firms to invest in more productive techniques, and so there is less incentive to choke of demand by raising prices. I suspect in an alternative world where Brexit had not happened the Bank of England would also be puzzling over why prices were not rising. As a result, if the MPC do finally raise interest rates this year, it would be one more mistake to add to the growing list under the heading Brexit.

[1] On each occasion I also wrote a post saying that they should not raise rates, starting I think at the beginning of 2014.

[2] I discussed in earlier posts why real wages are falling by even more than output per head.

[3] Or perhaps the pessimism of the bank manager lending money to those firms. The Haldane speech shows that productivity growth has remained strong among the top, frontier companies. Why? Because these companies, given their position, will be seeing growth relative to the average, and have got to the frontier through a culture of innovation.

Why recessions followed by austerity can have a persistent impact

Economics students are taught from an early age that in the short run aggregate demand matters, but in the long run output is determined from the supply side. A better way of putting it is that supply adjusts to demand in the short run, but demand adjusts to supply in the long run. A key part of that conceptualisation is that long run supply is independent of short run movements in demand (booms or recessions). It is a simple conceptualisation that has been extremely useful in the past. Just look at the UK data shown in this post: despite oil crises, monetarism and the ERM recessions, UK output per capita appeared to come back to an underlying 2.25% trend after WWII.

Except not any more: we are currently more than 15% below that trend and since Brexit that gap is growing larger every quarter. Across most advanced countries, it appears that the global financial crisis (GFC) has changed the trend in underlying growth. You will find plenty of stories and papers that try to explain this as a downturn in the growth of supply caused by slower technical progress that both predated the GFC and that is independent of the recession caused by it.

In a previous post I looked at recent empirical evidence that told a different story: that the recession that followed the GFC appears to be having a permanent impact on output. You can tell this story in two ways. The first is that, on this occasion for some reason, supply had adjusted to lower demand. The second is that we are still in a situation where demand is below supply.

The theoretical reasons why supply might adjust to demand are not difficult to find. (They are often described by economists under the jargon word 'hysteresis'.) Supply (in terms of output per capita) depends on labour force participation, the amount of productive capital in the economy, and finally technical progress, which is really just a catch all for how aggregate labour and capital combine to produce output. A long period of deficient demand can discourage workers. It can also hold back investment: a new project may be profitable but if there is no demand it will not get financed.

However the most obvious route to link a recession to longer term supply is through technical progress, which connects to the vast literature under the umbrella of ‘endogenous growth theory’. This can be done through a simple AK model (as Antonio Fatas does here), or using a more elaborate model of technical progress, as Gianluca Benigno and Luca Fornaro do in their paper entitled ‘Stagnation traps’. The basic idea is that in a recession innovation is less profitable, so firms do less of it, which leads to less growth in productivity and hence supply. Narayana Kocherlakota has promoted this idea: see here for example.

The second type of explanation is attractive, in part because the mechanism that is meant to get demand towards supply - monetary policy - has been ‘out of action’ for so long because of the Zero Lower Bound (ZLB). (The ZLB also plays an important role in the Benigno & Fornaro model.) However for some this type of explanation currently seems ruled out by the fact that unemployment is close to pre-crisis levels in the UK and US at least.

There are three quite different problems I have with the view that we no longer have a problem of deficient demand because unemployment is low. The first, and most obvious, is that the natural rate of unemployment might be, for various reasons, considerably lower than it was before the GFC. The second is that workers may have priced themselves into jobs. In particular, low real wages may have encouraged firms to use more labour intensive techniques. If that has happened, it does not mean that the demand deficiency problem has gone away, but just that it is more hidden. (For anyone who has a conceptual problem with that, just think about the simplest New Keynesian model, which assumes a perfectly clearing labour market but still has demand deficiency.)

The third involves the nature of any productivity slowdown caused by lack of innovation. A key question, which the papers noted above do not directly address, is whether we are talking about frontier research, or more the implementation of innovation (for example, copying what frontier firms are doing). There is some empirical evidence to suggest that the productivity slowdown may reflect the absence of the latter. This is very important, because it implies the slowdown is reversible. I have argued that central banks should pay much more attention to what I call the innovation gap (the gap between best practice techniques, and those that firms actually employ) and its link to investment and aggregate demand.

All this shows that there is no absence of ideas about how a great recession and a slow recovery could have lasting effects. If there is a problem, it is more that the simple conceptualisation that I talked about at the beginning of this post has too great a grip on the way many people think. If any of the mechanisms I have talked about are important, then it means that the folly of austerity has had an impact that could last for at least a decade rather than just a few years.

Underestimating the impact of austerity

There have been many ideas put forward to explain the low growth in UK productivity, but among mainstream accounts the impact of austerity is not usually high up on the list of possibilities. I have talked before about what I call an ‘innovations gap’, and how the UK is currently suffering a particularly large innovations gap. The idea of an innovations gap can link inadequate demand, like austerity, to low productivity growth.

Let me use a very simple example to explain how an innovations gap can arise after a deep recession followed by austerity. Assume that improvements in technology and production techniques are constantly taking place, or being learnt from other firms/countries, but they need new investment to put them in place. This is what economists call embodied technical progress.

Imagine a firm where demand is not increasing. Will that firm invest to become more productive? Only if the additional profits it can make as a result of investing (suitably discounted) is greater than the cost of the investment. For this firm we therefore need quite a big innovation gap before it is worth its while to undertake investment and before its productivity increases.

Now imagine that demand increases. The firm now has to undertake some additional investment to increase its output. It makes sense to invest in techniques that embody the latest technology. The increase in demand leads to both higher investment (what economists call the ‘accelerator’) and higher productivity.

In a normal recovery from a recession, demand recovers rapidly (growth easily exceeds past trends), leading firms to invest in the latest technology. Any innovations gap that might have opened up in the recession is quickly closed. In contrast, a very slow recovery caused by austerity will reduce the need for investment, allowing a large innovations gap to open up.

This idea fits in with some recent work which suggests that productivity growth in ‘frontier’ firms (firms that already have relatively high productivity) has not slowed, and a gap has opened up between these frontier firms and laggards. (See also here.) This would make sense if the frontier firms are growing (because they are the most productive) but the laggard firms are not. Growing firms need to invest to expand, but stagnant firms are not expanding.

The same model could also suggest how wage led productivity growth could occur (see Ben Chu here for example). As most innovations are likely to be labour saving, then higher wages can increase the profits that come from any particular investment, without necessarily increasing the cost of that investment. So an increase in wages caused by an increase in the minimum wage, for example, could increase investment and therefore increase productivity. The other side of that coin is that the period of stagnant wage growth we have had since the recession provided no incentive for firms to invest in higher productivity techniques.

I doubt that this story explains more than a part of the UK’s productivity gap. In the UK investment in plant and machinery fell sharply in the recession, and has not yet recovered to pre-recession levels, but its decline is unlikely to be enough to explain a productivity standstill. (For an account of some other key factors that could explain the UK productivity puzzle, see here.) But if it explains even a little, it makes austerity a lot more costly.

One way of describing what I’m saying is that austerity influences supply as well as demand. You could say austerity ignores the accelerator as well as the multiplier, which is cute but does not capture the idea of embodied technical progress which is crucial to this argument. It is why it is always best to run a high pressure economy (see Martin Sandbu here who links to an interview between Jared Bernstein and Josh Bivens).

As I explain here, I do not use austerity as just another name for any fiscal consolidation, or fiscal consolidation involving cuts to spending. Fiscal consolidation need not reduce output for the aggregate economy if monetary policy is able to offset its impact. But if interest rates are at their lower bound, as they are once again in the UK [1], fiscal consolidation will reduce output and lead to another needless waste of resources. Since Brexit we have a second period of UK austerity. In assessing how costly this will be, we need to look not just at whether it creates a negative output gap, but also how it creates an innovations gap that reduces productivity.

[1] We know this, because the Bank is increasing the amount of QE.

The Output Gap and the Innovations Gap

A major objection to my suggestion that the UK is in an self-fulfilling expectations led recession is that measures of the output gap suggest that gap is near zero. What I want to argue here is that measures of the output gap ignore what I will call the innovations gap, and the innovations gap could indicate that demand expansion would not be inflationary.

The output gap is the difference between actual output and trend output, where trend output is the level of output at which inflation is stable. The OBR are the output gap kings. They have a composite measure, which Ben Chu shows here, but this is derived from many different measures, shown in the OBR’s latest forecast on page 36. The OBR also show (p38) that estimates produced by other organisations vary widely. Most measures of the output gap can be categorised into four kinds.

1. Time series filters. These, at their most simple, just smooth the data on actual output to produce the measure of trend output that is one half of the output gap. These have no economic content and therefore tell us almost nothing.
   
   
2. Production function estimates. These combine measures of the labour force with the capital stock to give potential output: the level of output that could be produced if all factors of production were fully utilised. The major problem with these measures is that they have no measure of technology: how much can be produced by capital and labour. What is generally done is to use time series methods to estimate this, which takes us back to the smoothing idea. As a result, measures produced by the IMF and OECD suggest the years before the recession were a huge boom, which is implausible given other evidence we have.
   
   
3. Labour market measures, like unemployment or participation. There are of course many problems in knowing what the non-inflationary level of these variables are.
   
   
4. Firm surveys. These ask questions like are you producing at normal levels of capacity utilisation. The answer you get right now is that firms are indeed working close to normal capacity.
   
   

With these definitions in mind, what we have to ask is do any of these measures tell us what we really want to know, which is would firms react to increases in demand by raising prices and wages. I want to argue that they may not after an economy has grown at rates well below previous trends for a while. The reason is that, in these circumstances, firms may know that their current production methods are outdated, too labour intensive and inefficient, but at current levels of demand it is not worth them investing in new techniques. However if demand did increase, rather than raise prices to choke off that new demand, it would be more profitable to investment in new equipment to meet that additional demand. An expansion in demand would not be inflationary because firms would not raise their prices. In addition, because these new techniques were labour saving, there would be no inflationary pressure in the labour market (although real wages would rise because productivity increased).

We can tell the following story about the UK economy. At the peak of the recession, unemployment was high and firms had spare capacity. All the output gap measures said the output gap was large. What would normally happen next is that output would start recovering rapidly at above trend rates of growth, leading to a pickup in investment and new techniques being embodied in new production. But that didn’t happen in the UK, mainly because austerity held back demand and interest rates couldn’t go negative. 

When demand did finally begin to expand at a modest rate in 2013, cautious firms decided to meet that additional demand not through new investment but by using existing spare capacity. During 2013 and 2014 employment increased and the output gap fell, but productivity was stagnant because most firms were not investing in new techniques. (The market leaders were, because being market leaders they were expanding more rapidly and investing. So as Martin Sandbu has discussed, the UK productivity puzzle is associated with the average firm, not leading firms.)

This meant that by 2015, unemployment had fallen to more normal levels, and firms no longer had spare capacity. All this had been achieved with stagnant productivity growth, because most firms had stopped investing in new innovations. It was not because those innovations had stopped being made. So the output gap had been replaced by an innovations gap, with most firms using out of date production techniques that are too labour intensive.

If this story is right, we have become locked in a self-fulfilling low growth trap. Firms will not invest because they see recent slow growth continuing. They are right, because policymakers, looking at the output gap rather than the innovation gap, are doing nothing to expand demand for fear of inflation, or worse still because of mistaken worries about government debt. I do not know if this story is right, but it seems to me that the cost in lost output if it is right is so great that it is foolish to ignore this possibility.