Talk on where macroeconomics went wrong

I gave a short talk yesterday with this title, which takes some of the main points from my paper in the  OXREP 'Rebuilding Macro' volume  It is mainly of interest to economists, or those interested in economic methodology or the history of macroeconomic thought. When I talk about macroeconomics and macroeconomists below I mean mainstream academic economists.  

I want to talk today about where macroeconomics went wrong. Now it seems that this is a topic where everyone has a view. But most of those views have a common theme, and that is a dislike of DSGE models. Yet DSGE models are firmly entrenched in academic macroeconomics, and in pretty well every economist that has done a PhD, which is why the Bank of England’s core model is DSGE. To understand why DSGE is so entrenched, I need to tell the story of the New Classical Counter Revolution (NCCR).

If you had to pick a paper that epitomised the NCCR it would be “After Keynesian Macroeconomics” by Lucas and Sargent. Now from the title you would think this was an attack on Keynesian Economics, and in part it was. But we know that revolution failed. Very soon after the NCCR we had the birth of New Keynesian economics that recast key aspects of Keynesian economics within a microfoundations [1] framework, and is now the way nearly all central banks think about stabilisation policy. But if you read the text of Lucas and Sargent it is mainly a manifesto about how to do macroeconomics, or what I think we can reasonably call the methodology of macroeconomics.And on that their revolution was successful, and it is why nearly all academic macro is DSGE.

Before Lucas and Sargent complete macroeconomic models, of both a theoretical and empirical kind, had justified their aggregate equation using an eclectic mix of theory and econometrics. Microfoundations were used as a guide to aggregate equation specification, but if this equation fell apart in statistical terms when confronted with data in would not become part of an empirical model, and would be shunned for inclusion in theoretical models. Off course ‘falling apart ‘ is a very subjective criteria, and every effort would be made to try and make an equation consistent with microfoundations, but typically a lot of the dynamics in these models were what we would now call ad hoc, which in this case meant data based. .

Lucas famously showed that models of this kind were subject to what we call the Lucas critique [2], and that forms an important part of Lucas and Sargent paper. They argue that the only certain way to get round that critique is to build the model from internally consistent microfoundations. But they also ask why wouldn’t you want to build any macroeconomic model that way? Why wouldn’t you want a model where you could be sure that aggregate outcomes were the result of agents behaving in a consistent manner

If you want to crystallise why this was a methodological revolution, think about what we might call admissibility criteria for macro models. In pre-NCCR models equations were selected through an eclectic mixture of theory-fit and evidence-fit. In the RBC and later DSGE models internal theoretical consistency is an admissibility criteria. Or to put it another way, a DSGE model never got rejected because one of its equations didn’t fit the data, but if one equation had a theoretical foundation that was inconsistent with the others it would certainly not be published in the better journals.

Have a look at almost any macro paper in a top journal today, and compare it to a similar paper before the NCCR, and you can see we have been through a methodological revolution. Unfortunately many economists who have only been taught and who only known DSGE just think of this as progress. But it is not just progress, because DSGE models involve a shift away from the data. This is inevitable if you change the admissibility criteria away from the data. It inevitably means macroeconomists start focusing on models where it is easy to ensure internal theoretical consistency, and away from macroeconomic phenomenon that are clear in the data but more difficult to microfound.

If you are expecting me at this point to say that DSGE models where were macroeconomics went wrong, you will be disappointed. I spent the last 15 years of my research career building and analysing DSGE models, and I learnt a lot as a result. The mistake was the revolution part. In the US, DSGE models replaced traditional modelling within almost a decade [3]. In my view DSGE models should have coexisted with more traditional modelling, each tolerating the other.

To get a glimpse of how that can happen look at the UK, where a traditional macromodelling scene remained active until the end of the millenium. Traditional models didn’t stand sill, but changed by adopting many of the ideas from DSGE such as rational expectations. Here the account gets a little personal, because before I did DSGE I built one of those models, called COMPACT. There are not many macroeconomists who have built and operated both traditional and DSGE models, which I think gives me some insight of the merits of both.

COMPACT was a rational expectations New Keynesian model with explicit credit constraints in a Blanchard-Yaari type consumption function, a vintage production model, and variety effects on trade. So in terms of theoretical ideas it was far richer than any DSGE model I subsequently worked with. Most of COMPACT’s behavioural equations were econometrically estimated, but it was not an internally consistent model like DSGE.

COMPACT had an explicit but exogenous credit constraint variable in the model because in our view it was impossible to model consumption behaviour over time without it. Our work was based heavily on work in the UK by John Muellbauer, and Chris Carroll was coming to similar conclusions for the US. But DSGE models never faced that issue because they worked with de-trended data. Let me spell out why that was important. Empirical work was establishing that you could not begin to understand consumption behaviour over a 20/30 year time horizon without seeing how the financial sector had changed over time, and at least one traditional macroeconomic model was incorporating that finding before the end of the last millennium.. Extensive work on exactly that issue did not begin using DSGE models until after the financial crisis, where changes in the financial sector had a critical impact on the real economy. DSGE was behind the curve, but more traditional macroeconomics was not. .

Now I don’t think it is fanciful to think that if at least some macroeconomists had continued working with more traditional data-based models alongside those doing DSGE, at least one of those models would have thought to endogenise the financial sector which was determining those varying credit constraints.

So the claim I want to make is rather a big one. If DSGE models had continued alongside more traditional, data-based modelling, economists would have been much better prepared for the financial crisis when it came. If these two methodologies had learnt from each other, DSGE models might have started focusing on the financial sector before the crisis. Of course I would never suggest that macroeconomics could have predicted that crisis, but macroeconomists would have certainly had much more useful things to say about the impact on the economy when it happened.

Just being able to imagine this being true illustrates that moving to DSGE involved losses as well as gains. It inevitably made models less rich and moved them further away from the data in areas that were difficult but not impossible to model in a theoretically consistent way. The DSGE methodological revolution set out so clearly in Lucas and Sargent's paper changed the focus of macroeconomics away from things we now know were of critical importance.

I’ve been talking about this since I started writing a blog at the end of 2011, but recently we have seen similar messages from Paul Romer and Olivier Blanchard in this OxREP volume. What I have called here traditional models, and in the paper I call Structural Econometric Models, Blanchard calls provocatively policy models. It was provocative because most academic macroeconomists think DSGE models are the only models that can do policy analysis ‘properly’, but Blanchard suggests policymakers want models that are closer to the data more than they want a guarantee of internal consistency, and they want models that are quick and easy to adapt to unfolding problems. The US Fed, although it has a DSGE model, also has a more traditional model that has similarities to a traditional model like COMPACT, and guess which model plays the major role in the policy process?

[1] Microfoundations means deriving aggregate equations from microeconomic optimisation behaviour

[2] The Lucas critique argued that many equations of traditional macroeconomic models embodied beliefs about macro policy, and so if policy changed the equations would no longer be valid.

[3] The difficulty of identification in single equation estimation highlighted by Sims in 1980 probably also contributed.  . 

Did macroeconomics give up on explaining recent economic history?

The debate that continues about whether a Phillips curve still exists partly reflects the situation in various countries where unemployment has fallen to levels that had previously led to rising inflation but this time wage inflation seems pretty static. In all probability this reflects two things: the existence of hidden unemployment, and that the NAIRU has fallen. See Bell and Blanchflower on both for the UK.

The idea that the NAIRU can move gradually over time leads many to argue that the Phillips curve itself becomes suspect. In this post I tried to argue this is a mistake. It is also a mistake to think that estimating the position of the NAIRU is a mugs game. It is what central banks have to do if they take a structural approach to modelling inflation (and what other reasonable approaches are there?). Which raises the question as to why analysis of how the NAIRU moves is not a more prominent part of macro.

The following account may be way off, but I want to set it down because I am not aware of seeing it outlined elsewhere. I want to start with my account of why modern macro left the financial sector out of their models before the crisis. To cut a long story short, a focus on business cycle dynamics meant that medium term shifts in the relationship between consumption and income were largely ignored. Those who did study these shifts convincingly related them to changes in financial sector behaviour. Had more attention been paid to this, we might have seen much more analysis and more understanding of finance to real linkages.

Could the same story be told about the NAIRU? As with medium term trends in consumption, there is a literature on medium term movements in the NAIRU (or structural unemployment), but it does not tend to get into the top journals. One of the reasons, as with consumption, is that such analysis tends to be what modern macroeconomists would call ad hoc: it uses lots of theoretical ideas, none of which are carefully microfounded within the same paper. That is not a choice by those who do this kind of empirical work, but a necessity.

Much the same could apply to other key macro aggregates like investment. When economists ask whether investment is currently unusually high or low, they typically draw graphs and calculate trends and averages. We should be able to do much better than that. We should instead be looking at the equation that best captures the past 30 odd years of investment data, and asking whether it currently over or under predicts. The same is true for equilibrium exchange rates.

It was not just the New Classical Counter Revolution in macro that led to this downgrading of what we might call structural time series analysis of key macro relationships. Equally responsible was Sims famous paper 1980 ‘Macroeconomics and Reality’, that attacked the type of identification restrictions used in time series analysis and which proposed instead VAR methods. This perfect storm relegated the time series analysis that had been the bread and butter of macroeconomics to the minor journals.

I do not think it is too grandiose to claim that as a consequence macroeconomics gave up on trying to explain recent macroeconomic history: what could be called the medium term behaviour of macroeconomic aggregates, or why the economy did what it did over the last 30 or 40 years. Macro focused on the details of how business cycles worked, instead of how business cycles linked together.

Leading macroeconomists involved in policy see the same gaps, but express this dissatisfaction in a different way (with the important exception of Olivier Blanchard). For example John Williams, who has just been appointed to run the New York Fed, calls here for the next generation of DSGE models to focus on three areas. First they need to have a greater focus on modelling the labour market and the degree of slack, which I think amounts to the same thing as how the NAIRU changes over time. Second, he talks about a greater focus on medium- or long- run developments to both the ‘supply’ and ‘demand’ sides of the economy. The third of course involves incorporating the financial sector.

Perhaps one day DSGE models will do all this, although I suspect the macroeconomy is so complex that there will always be important gaps in what can be microfounded. But if it does happen, it will not come anytime soon. It is time that macroeconomics revisited the decisions it made around 1980, and realise that the deficiencies with traditional time series analysis that it highlighted were not as great as future generations have subsequently imagined. Macroeconomics needs to start trying to explain recent macroeconomic history once again.

Why the microfoundations hegemony holds back macroeconomic progress

When David Vines asked me to contribute to a OXREP (Oxford Review of Economic Policy) issue on “Rebuilding Macroeconomic Theory”, I think what he hoped I would write on how the core macro model needed to change to reflect macro developments since the crisis with a particular eye to modelling the impact of fiscal policy. That would be an interesting paper to write, but I decided fairly quickly that I wanted to say something that I thought was much more important.

In my view the biggest obstacle to the advance of macroeconomics is the hegemony of microfoundations. I wanted at least one of the papers in the collection to question this hegemony. It turned out that I was not alone, and a few papers did the same. I was particularly encouraged when Olivier Blanchard, in blog posts reflecting his thoughts before writing his contribution, was thinking along the same lines.

I will talk about the other papers when more people have had a chance to read them. Here I will focus on my own contribution. I have been pushing a similar line in blog posts for some time, and that experience suggests to me that most macroeconomists working within the hegemony have a simple mental block when they think about alternative modelling approaches. Let me see if I can break that block here.

Imagine a DSGE model, ‘estimated’ by Baynesian techniques. To be specific, suppose it contains a standard intertemporal consumption function. Now suppose someone adds a term into the model, say unemployment into the consumption function, and thereby significantly improves the fit of the model. It is not hard to think why the fit significantly improves: unemployment could be a proxy for the uncertainty of labour income, for example. The key question becomes which is the better model with which to examine macroeconomic policy: the DSGE or the augmented model?

A microfoundations macroeconomist will tend to say without doubt the original DSGE model, because only that model is known to be theoretically consistent. (They might instead say that only that model satisfies the Lucas critique, but internal consistency is the more general concept.) But an equally valid response is to say that the original DSGE model will give incorrect policy responses because it misses an important link between unemployment and consumption, and so the augmented model is preferred.

There is absolutely nothing that says that internal consistency is more important than (relative) misspecification. In my experience, when confronted with this fact, some DSGE modellers resort to two diversionary tactics. The first, which is to say that all models are misspecified, is not worthy of discussion. The second is that neither model is satisfactory, and research is needed to incorporate the unemployment effect in a consistent way.

I have no problem with that response in itself, and for that reason I have no problem with the microfoundations project as one way to do macroeconomic modelling. But in this particular context it is a dodge. There will never be, at least in my lifetime, a DSGE model that cannot be improved by adding plausible but potentially inconsistent effects like unemployment influencing consumption. Which means that, if you think models that are significantly better at fitting the data are to be preferred to the DSGE models from whence they came, then these augmented models will always beat the DSGE model as a way of modelling policy.

What this question tells you is that there is an alternative methodology for building macroeconomic models that is not inferior to the microfoundations approach. This starts with some theoretical specification, which could be a DSGE model as in the example, and then extends it in ways that are theoretically plausible and which also significantly improve the model’s fit, but which are not formally derived from micofoundations. I call that an example within the Structural Econometric Model (SEM) class, and Blanchard calls it a Policy Model.

An important point I make in my paper is that these are not competing methodologies, but instead they are complementary. SEMs as I describe them here start from microfounded theory. (Of course SEMs can also start from non-microfounded theory, but the pros and cons of that is a different debate I want to avoid here.) As a finished product they provide many research agendas for microfoundation modelling. So DSGE modelling can provide the starting point for builders of SEMs or Policy Models, and these models when completed provide a research agenda for DSGE modellers.

Once you see this complementarity, you can see why I think macroeconomics would develop much more rapidly if academics were involved in building SEMs as well as building DSGE models. The mistake the New Classical Counter Revolution made was to dismiss previous ways of modelling the economy, instead of augmenting these ways with additional approaches. Each methodology on its own will develop much more slowly than the two combined. Another way of putting it is that research based on SEMs is more efficient than the puzzle resolution approach used today. 

In the paper, I try to imagine what would have happened if the microfoundations project had just augmented the macroeconomics of the time (which was SEM modelling), rather than dismissing it out of hand. I think we have good evidence that active complementarity between SEM and microfoundations modelling would have investigated in depth links between the financial and real sectors before the financial crisis. The microfoundations hegemony chose the wrong puzzles to look at, deflecting macroeconomics from the more important empirical issues. The same thing may happen again if the microfoundations hegemony continues.

The day macroeconomics changed

It is of course ludicrous, but who cares. The day of the Boston Fed conference in 1978 is fast taking on a symbolic significance. It is the day that Lucas and Sargent changed how macroeconomics was done. Or, if you are Paul Romer, it is the day that the old guard spurned the ideas of the newcomers, and ensured we had a New Classical revolution in macro rather than a New Classical evolution. Or if you are Ray Fair (HT Mark Thoma), who was at the conference, it is the day that macroeconomics started to go wrong.

Ray Fair is a bit of a hero of mine. When I left the National Institute to become a formal academic, I had the goal (with the essential help of two excellent and courageous colleagues) of constructing a new econometric model of the UK economy, which would incorporate the latest theory: in essence, it would be New Keynesian, but with additional features like allowing variable credit conditions to influence consumption. Unlike a DSGE it would as far as possible involve econometric estimation. I had previously worked with the Treasury’s model, and then set up what is now NIGEM at the National Institute by adapting a global model used by the Treasury, and finally I had been in charge of developing the Institute’s domestic model. But creating a new model from scratch within two years was something else, and although the academics on the ESRC board gave me the money to do it, I could sense that some of them thought it could not be done. In believing (correctly) that it could, Ray Fair was one of the people who inspired me.

I agree with Ray Fair that what he calls Cowles Commission (CC) type models, and I call Structural Econometric Model (SEM) type models, together with the single equation econometric estimation that lies behind them, still have a lot to offer, and that academic macro should not have turned its back on them. Having spent the last fifteen years working with DSGE models, I am more positive about their role than Fair is. Unlike Fair, I want “more bells and whistles on DSGE models”. I also disagree about rational expectations: the UK model I built had rational expectations in all the key relationships.

Three years ago, when Andy Haldane suggested that DSGE models were partly to blame for the financial crisis, I wrote a post that was critical of Haldane. What I thought then, and continue to believe, is that the Bank had the information and resources to know what was happening to bank leverage, and it should not be using DSGE models as an excuse for not being more public about their concerns at the time.

However, if we broaden this out from the Bank to the wider academic community, I think he has a legitimate point. I have talked before about the work that Carroll and Muellbauer have done which shows that you have to think about credit conditions if you want to explain the pre-crisis time series for UK or US consumption. DSGE models could avoid this problem, but more traditional structural econometric (aka CC) models would find it harder to do so. So perhaps if academic macro had given greater priority to explaining these time series, it would have been better prepared for understanding the impact of the financial crisis.

What about the claim that only internally consistent DSGE models can give reliable policy advice? For another project, I have been rereading an AEJ Macro paper written in 2008 by Chari et al, where they argue that New Keynesian models are not yet useful for policy analysis because they are not properly microfounded. They write “One tradition, which we prefer, is to keep the model very simple, keep the number of parameters small and well-motivated by micro facts, and put up with the reality that such a model neither can nor should fit most aspects of the data. Such a model can still be very useful in clarifying how to think about policy.” That is where you end up if you take a purist view about internal consistency, the Lucas critique and all that. It in essence amounts to the following approach: if I cannot understand something, it is best to assume it does not exist.

Reform and revolution in macroeconomics

Mainly for economists

Paul Romer has a few recent posts (start here, most recent here) where he tries to examine why the saltwater/freshwater divide in macroeconomics happened. A theme is that this cannot all be put down to New Classical economists wanting a revolution, and that a defensive/dismissive attitude from the traditional Keynesian status quo also had a lot to do with it.

I will leave others to discuss what Solow said or intended (see for example Robert Waldmann). However I have no doubt that many among the then Keynesian status quo did react in a defensive and dismissive way. They were, after all, on incredibly weak ground. That ground was not large econometric macromodels, but one single equation: the traditional Phillips curve. This had inflation at time t depending on expectations of inflation at time t, and the deviation of unemployment/output from its natural rate. Add rational expectations to that and you show that deviations from the natural rate are random, and Keynesian economics becomes irrelevant. As a result, too many Keynesian macroeconomists saw rational expectations (and therefore all things New Classical) as an existential threat, and reacted to that threat by attempting to rubbish rational expectations, rather than questioning the traditional Phillips curve. As a result, the status quo lost. [1]

We now know this defeat was temporary, because New Keynesians came along with their version of the Phillips curve and we got a new ‘synthesis’. But that took time, and you can describe what happened in the time in between in two ways. You could say that the New Classicals always had the goal of overthrowing (rather than improving) Keynesian economics, thought that they had succeeded, and simply ignored New Keynesian economics as a result. Or you could say that the initially unyielding reaction of traditional Keynesians created an adversarial way of doing things whose persistence Paul both deplores and is trying to explain. (I have no particular expertise on which story is nearer the truth. I went with the first in this post, but I’m happy to be persuaded by Paul and others that I was wrong.) In either case the idea is that if there had been more reform rather than revolution, things might have gone better for macroeconomics.

The point I want to discuss here is not about Keynesian economics, but about even more fundamental things: how evidence is treated in macroeconomics. You can think of the New Classical counter revolution as having two strands. The first involves Keynesian economics, and is the one everyone likes to talk about. But the second was perhaps even more important, at least to how academic macroeconomics is done. This was the microfoundations revolution, that brought us first RBC models and then DSGE models. As Paul writes:

“Lucas and Sargent were right in 1978 when they said that there was something wrong, fatally wrong, with large macro simulation models. Academic work on these models collapsed.”

The question I want to raise is whether for this strand as well, reform rather than revolution might have been better for macroeconomics.

First two points on the quote above from Paul. Of course not many academics worked directly on large macro simulation models at the time, but what a large number did do was either time series econometric work on individual equations that could be fed into these models, or analyse small aggregate models whose equations were not microfounded, but instead justified by an eclectic mix of theory and empirics. That work within academia did largely come to a halt, and was replaced by microfounded modelling.

Second, Lucas and Sargent’s critique was fatal in the sense of what academics subsequently did (and how they regarded these econometric simulation models), although they got a lot of help from Sims (1980). But it was not fatal in a more general sense. As Brad DeLong points out, these econometric simulation models survived both in the private and public sectors (in the US Fed, for example, or the UK OBR). In the UK they survived within the academic sector until the latter 1990s when academics helped kill them off.

I am not suggesting for one minute that these models are an adequate substitute for DSGE modelling. There is no doubt in my mind that DSGE modelling is a good way of doing macro theory, and I have learnt a lot from doing it myself. It is also obvious that there was a lot wrong with large econometric models in the 1970s. My question is whether it was right for academics to reject them completely, and much more importantly avoid the econometric work that academics once did that fed into them.

It is hard to get academic macroeconomists trained since the 1980s to address this question, because they have been taught that these models and techniques are fatally flawed because of the Lucas critique and identification problems. But DSGE models as a guide for policy are also fatally flawed because they are too simple. The unique property that DSGE models have is internal consistency. Take a DSGE model, and alter a few equations so that they fit the data much better, and you have what could be called a structural econometric model. It is internally inconsistent, but because it fits the data better it may be a better guide for policy.

What happened in the UK in the 1980s and 1990s is that structural econometric models evolved to minimise Lucas critique problems by incorporating rational expectations (and other New Classical ideas as well), and time series econometrics improved to deal with identification issues. If you like, you can say that structural econometric models became more like DSGE models, but where internal consistency was sacrificed when it proved clearly incompatible with the data.

These points are very difficult to get across to those brought up to believe that structural econometric models of the old fashioned kind are obsolete, and fatally flawed in a more fundamental sense. You will often be told that to forecast you can either use a DSGE model or some kind of (virtually) atheoretical VAR, or that policymakers have no alternative when doing policy analysis than to use a DSGE model. Both statements are simply wrong.

There is a deep irony here. At a time when academics doing other kinds of economics have done less theory and become more empirical, macroeconomics has gone in the opposite direction, adopting wholesale a methodology that prioritised the internal theoretical consistency of models above their ability to track the data. An alternative - where DSGE modelling informed and was informed by more traditional ways of doing macroeconomics - was possible, but the New Classical and microfoundations revolution cast that possibility aside.

Did this matter? Were there costs to this strand of the New Classical revolution?

Here is one answer. While it is nonsense to suggest that DSGE models cannot incorporate the financial sector or a financial crisis, academics tend to avoid addressing why some of the multitude of work now going on did not occur before the financial crisis. It is sometimes suggested that before the crisis there was no cause to do so. This is not true. Take consumption for example. Looking at the (non-filtered) time series for UK and US consumption, it is difficult to avoid attaching significant importance to the gradual evolution of credit conditions over the last two or three decades (see the references to work by Carroll and Muellbauer I give in this post). If this kind of work had received greater attention (which structural econometric modellers would almost certainly have done), that would have focused minds on why credit conditions changed, which in turn would have addressed issues involving the interaction between the real and financial sectors. If that had been done, macroeconomics might have been better prepared to examine the impact of the financial crisis.

It is not just Keynesian economics where reform rather than revolution might have been more productive as a consequence of Lucas and Sargent, 1979.

[1] The point is not whether expectations are generally rational or not. It is that any business cycle theory that depends on irrational inflation expectations appears improbable. Do we really believe business cycles would disappear if only inflation expectations were rational? PhDs of the 1970s and 1980s understood that, which is why most of them rejected the traditional Keynesian position. Also, as Paul Krugman points out, many Keynesian economists were happy to incorporate New Classical ideas. 

Where macroeconomics went wrong

In my view, the answer is in the 1970/80s with the New Classical revolution (NCR). However I also think the new ideas that came with that revolution were progressive. I have defended rational expectations, I think intertemporal theory is the right place to start in thinking about consumption, and exploring the implications of time inconsistency is very important to macro policy, as well as many other areas of economics. I also think, along with nearly all macroeconomists, that the microfoundations approach to macro (DSGE models) is a progressive research strategy.

That is why discussion about these issues can become so confused. New Classical economics made academic macroeconomics take a number of big steps forward, but a couple of big steps backward at the same time. The clue to the backward steps comes from the name NCR. The research programme was anti-Keynesian (hence New Classical), and it did not want microfounded macro to be an alternative to the then dominant existing methodology, it wanted to replace it (hence revolution). Because the revolution succeeded (although the victory over Keynesian ideas was temporary), generations of students were taught that Keynesian economics was out of date. They were not taught about the pros and cons of the old and new methodologies, but were taught that the old methodology was simply wrong. And that teaching was/is a problem because it itself is wrong.

There had always been opposition to Keynesian ideas, and much (though not all) was ideological, such as attempts to remove Keynesian textbooks from US universities. However the NCR gave what should more precisely be called ‘aggregate demand denial’ an intellectual respectability that it never deserved. The reasons for believing that shifts in demand move output and employment in the short run and prices are sticky are overwhelming, so to deny both was a seemingly impossible task. It was achieved by adopting a methodological position which could ignore inconvenient evidence.

I do not think it had to be like this. Mainstream macroeconomics did not need a revolution in the 1970s and 1980s. Ideas like rational expectations could have been assimilated into the mainstream methodology, and microfounded models could have been developed alongside more eclectic econometric models (SEMs, not VARs), or aggregate theoretical models that Blanchard and Fischer rightly called ‘useful models’. Microfounded models could have shown the kind of errors that can arise in more empirically based models when theory is ignored or only applied piecemeal, and these empirical models could have highlighted the key areas where additional microfoundations were needed.

I think if this had happened, macroeconomics would have been better prepared when the financial crisis hit. Take just one issue: the role of credit conditions in influencing consumption. This is clearly crucial in understanding how consumption might respond to a credit crunch, yet any mechanism of this kind was absent from most DSGE models in 2008. However a more empirically based model of consumption would have had to address this issue well before 2008, as I argue here. If these types of models had continued to be developed within academia, rather than confined to the dustbin by the microfoundations revolution, then at least policymakers would have had something to work with. If there had been interaction between empirical and microfounded models some of the financial frictions literature that has flourished since 2008 might have appeared earlier.

So why didn’t this happen? Why did we have a revolution which overturned an existing methodology and temporarily banished Keynesian theory, rather than an adaptation and augmentation of what was then mainstream? Was the attraction of overturning orthodoxy too strong, as it is for a minority of heterodox economists today? Did an ideological imperative of dismissing Keynesian ideas play a role? To what extent was the hostile reaction of many in the macroeconomic establishment to eminently sensible ideas like rational expectations responsible? Was the attraction of a methodology where at least you could be sure you were consistent too enticing, perhaps encouraged by increasing segmentation between theoretical and empirical macro? I would love to know the answer to these questions. 

  

Methodological seduction

Mainly for macroeconomists or those interested in economic methodology. I first summarise my discussion in two earlier posts (here and here), and then address why this matters.

If there is such a thing as the standard account of scientific revolutions, it goes like this:

1) Theory A explains body of evidence X

2) Important additional evidence Y comes to light (or just happens)

3) Theory A cannot explain Y, or can only explain it by means which seem contrived or ‘degenerate’. (All swans are white, and the black swans you saw in New Zealand are just white swans after a mud bath.)

4) Theory B can explain X and Y

5) After a struggle, theory B replaces A.

For a more detailed schema due to Lakatos, which talks about a theory’s ‘core’ and ‘protective belt’ and tries to distinguish between theoretical evolution and revolution, see this paper by Zinn which also considers the New Classical counterrevolution.

The Keynesian revolution fits this standard account: ‘A’ is classical theory, Y is the Great Depression, ‘B’ is Keynesian theory. Does the New Classical counterrevolution (NCCR) also fit, with Y being stagflation?

My argument is that it does not. Arnold Kling makes the point clearly. In his stage one, Keynesian/Monetarist theory adapts to stagflation, using the Friedman/Phelps accelerationist Phillips curve. Stage two involves rational expectations, the Lucas supply curve and other New Classical ideas. As Kling says, “there was no empirical event that drove the stage two conversion.” I think from this that Paul Krugman also agrees, although perhaps with an odd quibble.

Now of course the counter revolutionaries do talk about the stagflation failure, and there is no dispute that stagflation left the Keynesian/Monetarist framework vulnerable. The key question, however, is whether points (3) and (4) are correct. On (3) Zinn argues that changes to Keynesian theory to account for stagflation were progressive rather than contrived, and I agree. I also agree with John Cochrane that this adaptation was still empirically inadequate, and that further progress needed rational expectations (see this separate thread), but as I note below the old methodology could (and did) incorporate this particular New Classical innovation.

More critically, (4) did not happen: New Classical models were not able to explain the behaviour of output and inflation in the 1970s and 1980s, or in my view the Great Depression either. Yet the NCCR was successful. So why did (5) happen, without (3) and (4)?

The new theoretical ideas New Classical economists brought to the table were impressive, particularly to those just schooled in graduate micro. Rational expectations is the clearest example. Ironically the innovation that had allowed conventional macro to explain stagflation, the accelerationist Phillips curve, also made it appear unable to adapt to rational expectations. But if that was all, then you need to ask why New Classical ideas could have been gradually assimilated into the mainstream. Many of the counter revolutionaries did not want this (as this note from Judy Klein via Mark Thoma makes clear), because they had an (ideological?) agenda which required the destruction of Keynesian ideas. However, once the basics of New Keynesian theory had been established, it was quite possible to incorporate concepts like rational expectations or Ricardian Eqivalence into a traditional structural econometric model (SEM), which is what I spent a lot of time in the 1990s doing.

The real problem with any attempt at synthesis is that a SEM is always going to be vulnerable to the key criticism in Lucas and Sargent, 1979: without a completely consistent microfounded theoretical base, there was the near certainty of inconsistency brought about by inappropriate identification restrictions. How serious this problem was, relative to the alternative of being theoretically consistent but empirically wide of the mark, was seldom asked.   

So why does this matter? For those who are critical of the total dominance of current macro microfoundations methodology, it is important to understand its appeal. I do not think this comes from macroeconomics being dominated by a ‘self-perpetuating clique that cared very little about evidence and regarded the assumption of perfect rationality as sacrosanct’, although I do think that the ideological preoccupations of many New Classical economists has an impact on what is regarded as de rigueur in model building even today. Nor do I think most macroeconomists are ‘seduced by the vision of a perfect, frictionless market system.’ As with economics more generally, the game is to explore imperfections rather than ignore them. The more critical question is whether the starting point of a ‘frictionless’ world constrains realistic model building in practice.

If mainstream academic macroeconomists were seduced by anything, it was a methodology - a way of doing the subject which appeared closer to what at least some of their microeconomic colleagues were doing at the time, and which was very different to the methodology of macroeconomics before the NCCR. The old methodology was eclectic and messy, juggling the competing claims of data and theory. The new methodology was rigorous! 

Noah Smith, who does believe stagflation was important in the NCCR, says at the end of his post: “this raises the question of how the 2008 crisis and Great Recession are going to affect the field”. However, if you think as I do that stagflation was not critical to the success of the NCCR, the question you might ask instead is whether there is anything in the Great Recession that challenges the methodology established by that revolution. The answer that I, and most academics, would give is absolutely not – instead it has provided the motivation for a burgeoning literature on financial frictions. To speak in the language of Lakatos, the paradigm is far from degenerate.  

Is there a chance of the older methodology making a comeback? I suspect the place to look is not in academia but in central banks. John Cochrane says that after the New Classical revolution there was a split, with the old style way of doing things surviving among policymakers. I think this was initially true, but over the last decade or so DSGE models have become standard in many central banks. At the Bank of England, their main model used to be a SEM, was replaced by a hybrid DSGE/SEM, and was replaced in turn by a DSGE model. The Fed operates both a DSGE model and a more old-fashioned SEM. It is in central banks that the limitations of DSGE analysis may be felt most acutely, as I suggested here. But central bank economists are trained by academics. Perhaps those that are seduced are bound to remain smitten.

Rereading Lucas and Sargent 1979

Mainly for macroeconomists and those interested in macroeconomic thought

Following this little interchange (me, Mark Thoma, Paul Krugman, Noah Smith, Robert Waldman, Arnold Kling), I reread what could be regarded as the New Classical manifesto: Lucas and Sargent’s ‘After Keynesian Economics’ (hereafter LS). It deserves to be cited as a classic, both for the quality of ideas and the persuasiveness of the writing. It does not seem like something written 35 ago, which is perhaps an indication of how influential its ideas still are.

What I want to explore is whether this manifesto for the New Classical counter revolution was mainly about stagflation, or whether it was mainly about methodology. LS kick off their article with references to stagflation and the failure of Keynesian theory. A fundamental rethink is required. What follows next is I think crucial. If the counter revolution is all about stagflation, we might expect an account of why conventional theory failed to predict stagflation - the equivalent, perhaps, to the discussion of classical theory in the General Theory. Instead we get something much more general - a discussion of why identification restrictions typically imposed in the structural econometric models (SEMs) of the time are incredible from a theoretical point of view, and an outline of the Lucas critique.

In other words, the essential criticism in LS is methodological: the way empirical macroeconomics has been done since Keynes is flawed. SEMs cannot be trusted as a guide for policy. In only one paragraph do LS try to link this general critique to stagflation:

“Though not, of course, designed as such by anyone, macroeconometric models were subjected to a decisive test in the 1970s. A key element in all Keynesian models is a trade-off between inflation and real output: the higher is the inflation rate, the higher is output (or equivalently, the lower is the rate of unemployment). For example, the models of the late 1960s predicted a sustained U.S. unemployment rate of 4% as consistent with a 4% annual rate of inflation. Based on this prediction, many economists at that time urged a deliberate policy of inflation. Certainly the erratic ‘fits and starts’ character of actual U.S. policy in the 1970s cannot be attributed to recommendations based on Keynesian models, but the inflationary bias on average of monetary and fiscal policy in this period should, according to all of these models, have produced the lowest unemployment rates for any decade since the 1940s. In fact, as we know, they produced the highest unemployment rates since the 1930s. This was econometric failure on a grand scale.”

There is no attempt to link this stagflation failure to the identification problems discussed earlier. Indeed, they go on to say that they recognise that particular empirical failures (by inference, like stagflation) might be solved by changes to particular equations within SEMs. Of course that is exactly what mainstream macroeconomics was doing at the time, with the expectations augmented Phillips curve.

In the schema due to Lakatos, a failing mainstream theory may still be able to explain previously anomalous results, but only in such a contrived way that it makes the programme degenerate. Yet, as Jesse Zinn argues in this paper, the changes to the Phillips curve suggested by Friedman and Phelps appear progressive rather than degenerate. True, this innovation came from thinking about microeconomic theory, but innovations in SEMs had always come from a mixture of microeconomic theory and evidence. 

This is why LS go on to say: “We have couched our criticisms in such general terms precisely to emphasise their generic character and hence the futility of pursuing minor variations within this general framework.” The rest of the article is about how, given additions like a Lucas supply curve, classical ‘equilibrium’ analysis may be able to explain the ‘facts’ about output and unemployment that Keynes thought classical economics was incapable of doing. It is not about how these models are, or even might be, better able to explain the particular problem of stagflation than SEMs.

In their conclusion, LS summarise their argument. They say:

“First, and most important, existing Keynesian macroeconometric models are incapable of providing reliable guidance in formulating monetary, fiscal and other types of policy. This conclusion is based in part on the spectacular recent failures of these models, and in part on their lack of a sound theoretical or econometric basis.”

Reading the paper as a whole, I think it would be fair to say that these two parts were not equal. The focus of the paper is about the lack of a sound theoretical or econometric basis for SEMs, rather than the failure to predict or explain stagflation. As I will argue in a subsequent post, it was this methodological critique, rather than any superior empirical ability, that led to the success of this manifesto.

Understanding the New Classical revolution

In the account of the history of macroeconomic thought I gave here, the New Classical counter revolution was both methodological and ideological in nature. It was successful, I suggested, because too many economists were unhappy with the gulf between the methodology used in much of microeconomics, and the methodology of macroeconomics at the time.

There is a much simpler reading. Just as the original Keynesian revolution was caused by massive empirical failure (the Great Depression), the New Classical revolution was caused by the Keynesian failure of the 1970s: stagflation. An example of this reading is in this piece by the philosopher Alex Rosenberg (HT Diane Coyle). He writes: “Back then it was the New Classical macrotheory that gave the right answers and explained what the matter with the Keynesian models was.”

I just do not think that is right. Stagflation is very easily explained: you just need an ‘accelerationist’ Phillips curve (i.e. where the coefficient on expected inflation is one), plus a period in which monetary policymakers systematically underestimate the natural rate of unemployment. You do not need rational expectations, or any of the other innovations introduced by New Classical economists.

No doubt the inflation of the 1970s made the macroeconomic status quo unattractive. But I do not think the basic appeal of New Classical ideas lay in their better predictive ability. The attraction of rational expectations was not that it explained actual expectations data better than some form of adaptive scheme. Instead it just seemed more consistent with the general idea of rationality that economists used all the time. Ricardian Equivalence was not successful because the data revealed that tax cuts had no impact on consumption - in fact study after study have shown that tax cuts do have a significant impact on consumption.

Stagflation did not kill IS-LM. In fact, because empirical validity was so central to the methodology of macroeconomics at the time, it adapted to stagflation very quickly. This gave a boost to the policy of monetarism, but this used the same IS-LM framework. If you want to find the decisive event that led to New Classical economists winning their counterrevolution, it was the theoretical realisation that if expectations were rational, but inflation was described by an accelerationist Phillips curve with expectations about current inflation on the right hand side, then deviations from the natural rate had to be random. The fatal flaw in the Keynesian/Monetarist theory of the 1970s was theoretical rather than empirical.

Was the neoclassical synthesis unstable?

This post presents a very simple story of the development of macroeconomic thought from Keynes until today. It is related to a recent post from Brad DeLong on ‘economic theology’ and the neoclassical synthesis. (See also a response from Robert Waldmann.) 

Economics as a science that studies markets is ideologically neutral. Economic theory can be used to support ‘unfettered’ markets, or it can be used to justify interventions to avoid various kinds of market failure. The former means that it will inevitably be used by some to support a laissez-faire ideological position. There are two checks against this one-sided presentation of economic theory: economists presenting alternative theories that embody imperfections, and the use of evidence to show that a particular theory works, either in terms of its assumptions or results.

Before considering macroeconomics, take an example from labour economics: the minimum wage. Standard competitive theory suggests a minimum wage will reduce employment and raise unemployment. Card and Krueger undertook a famous study suggesting that in one particular example where the minimum wage was increased there was no reduction in employment. That led to a substantial amount of additional research, much (but by no means all) backing up the result that the impact of moderate increases in the minimum wage on employment was either non-existent or very small. For similar developments in the UK, see this account by Alan Manning. This empirical evidence was sufficient to encourage the development of alternative theoretical models: principally but not only monopsony.

So here we see theory and evidence interacting in a Popperian type way, hopefully leading to better theory. [1] Yet with economics there will always be ideological resistance, so there will always be those who want to stick to the basic model and who select those empirical studies that support it. For the discipline to survive, those ideologues have to be a minority. But even if this condition is met, a healthy discipline has to recognise the influence of that minority, rather than try and pretend it does not exist or does not matter.

There is a slight twist for macroeconomics. As governments are the monopoly providers of cash, and provide a backstop to the financial system, they are involved in the ‘market’ whether they like it or not. Complete non-intervention is not an option: instead the next best thing (from a laissez-faire point of view) is some kind of ‘neutral’ default policy rule, like keeping the stock of money constant.

The Great Depression was the empirical wake-up call (the equivalent of the Card and Krueger study) for macroeconomics. So profound was the impact of this empirical event that it led to a whole new way of doing the subject. Keynesian economics was methodologically different from much of microeconomics: it put much more weight on aggregate evidence (through time series econometrics), and much less on microeconomic theory. One way of putting this is that in the 1960s, general equilibrium theory of the Arrow-Debreu-McKenzie type seemed a complete contrast to what macroeconomists were doing. That an event as powerful as the Great Depression should have had such a profound methodological impact is not really surprising.

The Great Depression also meant that those advocating non-intervention had to make an exception of macroeconomics. It was for the generation after the Great Depression abundantly clear that here was a colossal market failure. This is one sense in which the term neo-classical synthesis can be used: to allow the state to combat the market failure represented by Keynesian unemployment (albeit, in the case of Friedman, in as rule like way as possible), but to maintain advocacy of non-intervention elsewhere. Note however that this is a synthesis servicing a particular ideological point of view, rather than being anything inherent within economics as a discipline.

Was this ‘ideological synthesis’ tenable among those supporting the ideology? There were two natural tensions. First, the position that macro intervention should be rule based and minimal was contestable. Second and more importantly, as the memory of the Great Depression faded (and neoliberalism spread), the temptation grew to ask ‘do we really have to accept the need for state intervention at the macro level’. However I’m not sure the latter would have become critical had it not been for another tension within macroeconomics itself. 

What was not tenable from a methodological point of view was the distance between the very empirical orientation of macroeconomics, and the more axiomatic foundation of much of microeconomics. What was required here was a different kind of synthesis, one which allowed for a healthy dialogue between theory and evidence. My impression is that in many areas of microeconomics this happened: that is partly why I gave the minimum wage example, but it is also worth noting that general equilibrium theory lost the primacy that it might once had among microeconomists. But these are impressions, and I’ll happily be corrected.

I think the same thing could have happened in macroeconomics. Heterodox economists (and Robert Waldmann) would almost certainly disagree, but I think macroeconomics has gained a great deal from the project to add microfoundations. Where I hope heterodox economists would agree is that a dialogue where theorists engaged with macroeconomics and tried to persuade macroeconomists of the importance of following particular theories would have been healthy. But that was not the way it turned out. What could have been a dialogue of the Popperian kind became instead a theoretical and methodological counter revolution. Instead of asking ‘what can we do to get better microfoundations for sticky prices’, the assertion became ‘without good microfoundations we should ignore sticky prices’.

Why was there a counter revolution in macro rather than a Popperian dialogue? I think it is here that the second tension in the ‘ideological synthesis’ I identified above is important. Those who wanted to dispute the need for macro intervention realised that the microfoundations for macro market failures that existed at the time were poor (adaptive expectations in a traditional Phillips curve), and so any macroeconomics based on ‘rigorous’ (textbook, imperfection free) microfoundations would not be Keynesian. They also realised that they could produce models which generated real business cycles which were entirely efficient. These models assumed all unemployment was voluntary, which in any normal science would lead to their rejection, but in an axiomatic based approach where some evidence can be ignored it was acceptable.

New Classical economics did not want to improve Keynesian economics, but to overthrow it. It is very difficult to believe this motivation was not ideological. Does the fact that this counter revolution was largely successful among academic macroeconomists imply that the majority of macroeconomists shared this ideological outlook? I suspect not. What New Classical economists succeeded in doing was framing the issue as one where a choice had to be made, between an eclectic empirically orientated approach where theory was weak and empirical methods shaky, and an alternative whose methodological foundations were solidly based within the discipline of economics. So we moved from a position where macroeconomics and Arrow-Debreu-McKenzie seemed worlds apart, to one where at least some see the former arising naturally from the latter. Ironically this happened at the same time as many microeconomists saw Arrow-Debreu-McKenzie as less relevant to what they did.

Of course we have moved on from the 1980s. Yet in some respects we have not moved very far. With the counter revolution we swung from one methodological extreme to the other, and we have not moved much since. The admissibility of models still depends on their theoretical consistency rather than consistency with evidence. It is still seen as more important when building models of the business cycle to allow for the endogeneity of labour supply than to allow for involuntary unemployment. What this means is that many macroeconomists who think they are just ‘taking theory seriously’ are in fact applying a particular theoretical view which happens to suit the ideology of the counter revolutionaries. The key to changing that is to first accept it.

[1] By Popperian type, I just mean that a theory proves inconsistent with data and so a better theory is developed. The Popperian ideal where one piece of evidence (one black swan) is enough on its own to disprove a theory is never going to apply in economics (if it applies anywhere), because evidence is probabilistic and fragile. There are no black swans in economics.