Confidence as a political device

Some technical references but the key point does not need them

This is a contribution to the discussion about models started by Krugman, DeLong and Summers, and in particular to the use of confidence. (Martin Sandbu has an excellent summary, although as you will see I think he is missing something.) The idea that confidence can on occasion be important, and that it can be modelled, is not (in my view) in dispute. For example the very existence of banks depends on confidence (that depositors can withdraw their money when they wish), and when that confidence disappears you get a bank run.

But the leap from the statement that ‘in some circumstances confidence matters’ to ‘we should worry about bond market confidence in an economy with its own central bank in the middle of a depression’ is a huge one, and I think Tony Yates and others are in danger of making that leap without justification. Yes, there are circumstances when it may be optimal for a country with its own central bank to default, and Corsetti and Dedola (in a paper I discussed here) show how that can lead to multiple equilibria.

But just as Krugman wanted to emulate Woody Allen, I want to as well but this time pull Dani Rodrik from behind the sign. In his excellent new book (which I have almost finished reading) Rodrik talks about the fact that in economics there are usually many models, and the key question is their applicability. So you have to ask, for the US and UK in 2009, was there the slightest chance that either government wanted to default? The question is not would they be forced to default, because with their own central bank they would not be, but would they choose to default. And the answer has to be a categorical no. Why would they, with interest rates so low and debt easy to sell.

The argument goes that if the market suddenly gets spooked and stops buying debt, printing money will cause inflation, and in those circumstances the government might choose to default. But we were in the midst of the biggest recession since the 1930s. Any money creation would have had no immediate impact on inflation. Of course their central banks had just begun printing lots of money as part of Quantitative Easing, and even 5 years later where is the inflation! So once again there would be no chance that the government would choose to default: the Corsetti and Dedola paper is not applicable. (Robert makes a similar point about the Blanchard paper. I will not deal with the exchange rate collapse idea because Paul already has. A technical aside: Martin raises a point about UK banks overseas currency activity, which I will try to get back to in a later post.)

Ah, but what if the market remains spooked for so long that eventually inflation rises. The markets stop buying US or UK debt because they think that the government will choose to default, and even after 5 or 10 years and still no default the markets continue to think that, even though they are desperate for safe assets!? In Corsetti and Dedola agents are rational, so we have left that paper way behind. We have entered, I’m afraid, the land of pure make believe.

So there is no applicable model that could justify the confidence effects that might have made us cautious in 2009 about issuing more debt. There are models about an acute shortage of safe assets on the other hand, which seem to be ignored by those arguing against fiscal stimulus. Nor is there the slightest bit of evidence that the markets were ever even thinking about being spooked in this way.

Martin makes the point that just because something has not yet been formally modelled does not mean it does not happen. Of course, and indeed if he means by model a fully microfounded DSGE model I have made this point many times myself. But you can also use the term model in a much more general sense, as a set of mutually consistent arguments. It is in that sense that I mean no applicable model.

Now to the additional point I really wanted to make. When people invoke the idea of confidence, other people (particularly economists) should be automatically suspicious. The reason is that it frequently allows those who represent the group whose confidence is being invoked to further their own self interest. The financial markets are represented by City or Wall Street economists, and you invariably see market confidence being invoked to support a policy position they have some economic or political interest in. Bond market economists never saw a fiscal consolidation they did not like, so the saying goes, so of course market confidence is used to argue against fiscal expansion. Employers drum up the importance of maintaining their confidence whenever taxes on profits (or high incomes) are involved. As I argue in this paper, there is a generic reason why financial market economists play up the importance of market confidence, so they can act as high priests. (Did these same economists go on about the dangers of rising leverage when confidence really mattered, before the global financial crisis?)

The general lesson I would draw is this. If the economics point towards a conclusion, and people argue against it based on ‘confidence’, you should be very, very suspicious. You should ask where is the model (or at least a mutually consistent set of arguments), and where is the evidence that this model or set of arguments is applicable to this case? Policy makers who go with confidence based arguments that fail these tests because it accords with their instincts are, perhaps knowingly, following the political agenda of someone else.     

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. 

The entirely predictable recession

Sometimes when I write about the Eurozone, I get comments about how inappropriate it is to apply ‘an anglo-saxon model’ of how that economy works. I think the best translation of ‘anglo-saxon’ is Keynesian. In an important sense this is rubbish. All I am doing is using the framework that is used by most applied macromodellers everywhere. That framework says that if you have a large fiscal contraction like this

Underlying Government Primary Balances: source OECD Economic Outlook

without a large compensating relaxation in monetary policy, then you will get the stagnation in output that I showed at the top of this post, and a substantial increase in unemployment.

Can this scale of fiscal contraction in itself fully account for the second Eurozone recession? There are various ways of answering this question: see, for example, this work by Jordà and Taylor, or the analysis by Holland and Portes that uses a structural econometric model. A very recent paper (available here) by Ansgar Rannenberg, Christian Schoder and Jan Strasky does something different. It uses modified versions of three different DSGE models to analyse the impact of the Eurozone fiscal contraction from 2011 to 2013. One of these models is QUEST III at the European Commission, which Jan in‘t Veld used in analysis I described here. The other two are FiMod, developed by staff of the Deutsche Bundesbank and the Banco de Espana, and NAWM, the ECB’s New Area Wide Model. All very ‘anglo-saxon’!

The modifications the authors make to the models, and the details of their analysis, would probably only be of interest to inveterate macromodellers like me, so those who want to know more should read the paper. Here I will just quote the key conclusions:

“We find that fiscal consolidation caused a cumulative GDP loss of between 14% and 20% of annual baseline GDP over the 2011 to 2013 period in the Euro Area [EA], implying a cumulative multiplier between 1.5 and 2.2.” and “As a result, the simulated GDP effects of the EA’s fiscal consolidation are large, and would be more than sufficient to explain the recent recession in the EA.”

The idea that a large fiscal contraction shortly after a huge financial crisis would lead to a second recession is not the wild imagining of a group of ‘anglo-saxon’ economists, or a particular macroeconomic ‘school of thought’. It is just mainstream macroeconomics. And we must never forget that this is not the unfortunate cost of having to get debt down in a few periphery countries: as the chart above shows, this fiscal contraction occurred everywhere in the Eurozone. As the simulations described in this link (pdf) show, using the Belgian NIME model, these costs could have been largely avoided if the fiscal consolidation had been delayed until monetary policy was in a position to offset them. It is not just a predictable recession; it is a recession made by policymakers without good cause and therefore an entirely avoidable recession.

Why central banks use models to forecast

One of the things I really like about writing blogs is that it puts my views to the test. After I have written them of course, through comments and other bloggers. But also as I write them.

Take my earlier post on forecasting. When I began writing it I thought the conventional wisdom was that model based forecasts plus judgement did slightly better than intelligent guesswork. That view was based in part on a 1989 survey by Ken Wallis, which was about the time I stopped helping to produce forecasts. If that was true, then the justification for using model based forecasting in policy making institutions was simple: even quite small improvements in accuracy had benefits which easily exceeded the extra costs of using a model to forecast.

However, when ‘putting pen to paper’ I obviously needed to check that this was still the received wisdom. Reading a number of more recent papers suggested to me that it was not. I’m not quite sure if that is because the empirical evidence has changed, or just because studies have had a different focus, but it made me think about whether this was really the reason that policy makers tended to use model based forecasts anyway. And I decided it was probably not.

In a subsequent post I explained why policymakers will always tend to use macroeconomic models, because they need to do policy analysis, and models are much better at this than unconditional forecasting. Policy analysis is just one example of conditional forecasting: if X changes, how will Y change. To see why this helps to explain why they also tend to use these models to do unconditional forecasting (what will Y be), let’s imagine that they did not. Suppose instead they just used intelligent guesswork.

Take output for example. Output tends to go up each year, but this trend like behaviour is spasmodic: sometimes growth is above trend, sometimes below. However output tends to gradually revert to this trend growth line, which is why we get booms and recessions: if the level of output is above the trend line this year, it is more likely to be above than below next year. Using this information can give you a pretty good forecast for output. Suppose someone at the central bank shows that this forecast is as good as those produced by the bank’s model, and so the bank reassigns its forecasters and uses this intelligent guess instead.

This intelligent guesswork gives the bank a very limited story about why its forecast is what it is. Suppose now oil prices rise. Someone asks the central bank what impact will higher oil prices have on their forecast? The central bank says none. The questioner is puzzled. Surely, they respond, higher oil prices increase firms’ costs leading to lower output. Indeed, replies the central bank. In fact we have a model that tells us how big that effect might be. But we do not use that model to forecast, so our forecast has not changed. The questioner persists. So what oil price were you assuming when you made your forecast, they ask? We made no assumption about oil prices, comes the reply. We just looked at past output.

You can see the problem. By using an intelligent guess to forecast, the bank appears to be ignoring information, and it seems to be telling inconsistent stories. Central banks that are accountable do not want to get put in this position. From their point of view, it would be much easier if they used their main policy analysis model, plus judgement, to also make unconditional forecasts. They can always let the intelligent guesswork inform their judgement. If these forecasts are not worse than intelligent guesswork, then the cost to them of using the model to produce forecasts - a few extra economists - are trivial.

On Macroeconomic Forecasting

Macroeconomic forecasts produced with macroeconomic models tend to be little better than intelligent guesswork. That is not an opinion – it is a fact. It is a fact because for decades many reputable and long standing model based forecasters have looked at their past errors, and that is what they find. It is also a fact because we can use models to generate standard errors for forecasts, as well as the most likely outcome that gets all the attention. Doing so indicates errors of a similar magnitude as those observed from past forecasts. In other words, model based forecasts are predictably bad.

The sad news is that this situation has not changed since I was involved in forecasting around 30 years ago. During the years before the Great Recession (the Great Moderation) forecasts might have appeared to get better, but that was because most economies became less volatile. As is well known, the Great Recession was completely missed. Forecasting has not improved, because our ability to explain variables like consumption or investment has not improved.

Does that mean that macroeconomics is not making any progress? I do not want to get sidetracked on this issue, but it could just be that as macroeconomists understand the economy as it was a little better, the nature of the economy also changes because of factors like financial innovation or technical progress. Does this mean macroeconomics is useless? No, in much the same way as medicine cannot predict year by year how your health changes but is quite good at responding to these changes.

What it does mean is that it is very difficult to use forecast performance as a means of judging between alternative models or organisations. Who is better in any one year is largely luck. You need to look at performance over at least a decade to be able to distinguish between luck and a better model or better judgement. Unfortunately models, and the people who use them, rarely remain unchanged for this length of time. The model and the modelling team that the Bank of England uses to forecast is different from its model and team ten years ago. (For much more on this see Ben Broadbent here.) 

I think it is safe to say that this inability to accurately forecast is unlikely to change anytime soon. Which raises an obvious question: why do people still use often elaborate models to forecast? Here it is useful to distinguish between policy making bodies like central banks and the rest.

It makes sense for both monetary and fiscal authorities to forecast. So why use the combination of a macroeconomic model and judgement to do so, rather than intelligent guesswork? (Intelligent guesswork here means some atheoretical time series forecasting technique.) The first point is that it is not obviously harmful to do so. (From my unsystematic reading the only consistent results from forecasting comparisons using alternative techniques is that all forecasts are pretty poor.) The second point is that forecasting using macroeconomic models allows forecasters to combine a large amount of information into a reasonably consistent story that links what has happened to what might happen, and policymakers find these stories helpful. (See, for example, this post by Corola Binder, or some of the work of Deirdre McCloskey.) There are interesting questions about what type of macromodel is best suited for forecasting, and whether the model used for forecasting should also be used for policy analysis, but I’ll leave those for another day.

Many other organisations, not directly involved in policy making, produce macro forecasts. Why do they bother? Why not just use the policy makers’ forecast? A large part of the answer must be that the media shows great interest in these forecasts. Why is this? I’m tempted to say it’s for the same reason as many people read daily horoscopes. However I think it’s worth adding that there is a small element of a conspiracy to deceive going on here too.

To set the scene, consider a similar little conspiracy. There is now a convention that when there are interesting aggregate moves on the stock or currency markets, the media will present some ‘expert’ - typically an economist working for some financial company - who will tell us why the market has so moved. The truth is that no one knows why the market goes up or down, because no one asks each trader why they are making a trade. So all the expert can give us is an unverifiable intelligent guess, but they never tell you that. This small deception suits the media and the experts.

In a similar way, the media likes to pretend that forecasts are much more accurate than they actually are, because that makes small changes newsworthy. In reality forecasts tend to follow each other and change slowly, for reasons Tim Harford notes, so presenting some new forecast every week makes little sense. (Occasionally this conservatism among forecasters can be usefully predicted, as I once noted in an anecdote.) But this conspiracy has the added bonus for the media that it can express horrified shock and surprise when forecasts go wrong.

The rather boring truth is that it is entirely predictable that forecasters will miss major recessions, just as it is equally predictable that each time this happens we get hundreds of articles written asking what has gone wrong with macro forecasting. The answer is always the same - nothing. Macroeconomic model based forecasts are always bad, but probably no worse than intelligent guesses.

Misunderstanding macroeconomic models

The first half of this post is meant for non-economists, but it ends with a couple of points on OLG modelling

I recently wrote a post on the Eggertsson and Mehrotra paper on secular stagnation, because I thought the paper was interesting. A much more critical post from Unlearning Economics (UE) has just appeared in Pieria. UE says it “helps to illustrate the troubles faced by contemporary macroeconomics”. One of UE’s complaints seems to reflect a misunderstanding, often shared by non-economists, about what much academic macromodelling is designed to do.

UE objects to the fact that the model assumes that the amount the young can borrow (the degree of leverage) is exogenous, which means that there is no attempt to explain where this constraint on the borrowing of the young comes from. UE also complains that the model contains no banks, and no investment in physical capital. In other words, the model is much too simple. It is a natural enough idea: to explain what might be currently going on, you need a more complex model that includes everything that could be important.

There is certainly a place for this kind of more elaborate model. Christiano, Eichenbaum and Trabandt in this paper want to argue that a model based on New Keynesian theory can track what has happened over the last ten years. Their model has 40 equations. If I was trying to do a similar exercise, I would want to augment the standard New Keynesian framework with at least the following: nominal wage stickiness as well as price stickiness, a financial sector that endogenised both the cost and rationing of credit, a model of consumption which allowed for credit constraints and precautionary saving, a housing market, a model of the labour market that combined matching with rationing (as here), and something that allowed recessions to have long lasting (hysteretic) impacts on labour supply and technical progress. However large models like this will involve many macroeconomic ‘mechanisms’, and it will generally be unclear which mechanisms are important at driving particular results or explaining particular facts. We do not want to treat the elaborate model as a black box, but instead we want to understand its properties.

To understand complex models, we need much simpler models. (I once - in this paper - called the process of relating complex models to simpler models ‘theoretical deconstruction’.) In fact it is often sensible to start with the simpler model. For example, a particular issue with secular stagnation is to show how the natural real interest rate can be negative for decades rather than years (i.e. beyond the Keynesian short term)? What mechanism can do this? As I explained in my post, neither a standard representative agent model nor a standard two period overlapping generations model (OLG model, where the two generations are those earning and those retired) will give you that result. What Eggertsson and Mehrotra show is that a very simple three period OLG model (which adds a young generation that borrows) where borrowing by the young is constrained (they would like to borrow more but cannot) can provide just that mechanism.

That is a key point of the paper. The paper is not designed to explain where borrowing constraints come from: there is now a big literature on that. Thankfully the authors do not feel compelled to microfound these constraints. Instead the paper simply offers and explores a mechanism whereby an increase in these borrowing constraints could move the natural interest rate into negative territory, and for it to stay there. Having established that result, it is for subsequent work (which the authors intend to do) to see if that mechanism survives complicating the model, by for example adding investment.

Suppose the endeavour is successful, and a more complex but realistic model is able to provide an account of secular stagnation that includes other important mechanisms and which is based on a realistic set of parameter values. That would be a success, but those not familiar with all the work would ask: why does this model allow real interest rates to be negative when the standard models we know do not. The reply would be that the three period OLG structure was critical, and to see why have a look at the original, simple model.

Now you might say the authors should wait until they have built the more realistic model before creating what could turn out to be a research path that might fail to achieve its goal. That would be quite wrong, because the more debate there is within the academic community when ideas are at their early stages the better. I want to give an example of this, but here I will go into territory that will probably only interest macroeconomists.

It might be the case, for example, that the authors intuition that their results will survive introducing other assets like physical capital can be shown to be wrong very quickly. Indeed, Nick Rowe has already made such a claim, arguing that the presence of land as an asset ensures a positive real interest rate. If Nick was right this could be enough to kill the research programme, without any more time being wasted. Whether he is right is another matter: this paper by Rhee may be relevant in that respect.

Here I just want to add a final thought. Within an OLG framework, it may not be necessary to establish the existence of a steady state with negative real interest rates. The typical period in an OLG model lasts two or more decades. So if the dynamics of such a model involved some overshooting, it might be possible to generate prolonged periods (in years) of negative interest rates even if the steady state real interest rate was positive. To be honest I’m not sure what might give rise to overshooting of this kind, but that may just reflect my inadequate imagination.  

The Fed’s macroeconomic model

There has been some comment on the decision of the US central bank (the Fed) to publish its main econometric model in full. In terms of openness I agree with Tony Yates that this is a great move, and that the Bank of England should follow. The Bank publishes some details of its model (somewhat belatedly, as I noted here), but as Tony argues this falls some way short of what is now provided by the Fed.

However I think Noah Smith makes the most interesting point: unlike the Bank's model, the model published by the Fed is not a DSGE model. Instead, it is what is often called a Structural Econometric Model (SEM): a pretty ad hoc mixture of theory and econometric estimation that would not please either a macro theorist or a time series econometrician. As Noah notes, they use this model for forecasting and policy analysis. Noah speculates that the Fed’s move to publish a model of this kind indicates that they are perhaps less embarrassed about using a SEM than they once were. I’ve no idea if this is true, but for most academic macroeconomists it raises a puzzling question - why are they still using this type of model? If the Bank of England can use a DSGE model as their core model, why doesn’t the Fed?

I have discussed the question of what type of model a central bank should use before. In addition, I have written many posts (most recently here) advocating the advantages of augmenting DSGE models and VARs with this kind of middle way approach. For various reasons, this middle way approach will be particularly attractive to a policy making organisation like a central bank, but I also think that a SEM can play a role in academic analysis. For the moment, though, let me just focus on policy analysis by policy makers.

Consider a particular question: what is the impact of a temporary cut in income taxes? What kind of methods should an economist employ to answer this question? We could estimate reduced forms/VARs relating variables of interest (output, inflation etc) to changes in income taxes in the past. However there are serious problems with this approach. The most obvious is that the impact of past changes in taxes will depend on the reaction of monetary policy at the time, and whether monetary policy will act in a similar way today. Results will also depend on how permanent past changes in taxes were expected to be. I would not want to suggest that these issues make reduced form estimation a waste of time, but they do indicate how difficult it will be to get a good answer using this approach. Similar problems arise if we relate growth to debt, money to prices (a personal reflection here) and so on. Macro reduced form analysis relating policy variables to outcomes is very fragile.

An alternative would be for the economist to build a DSGE model, and simulate that. This has a number of advantages over the reduced form estimation approach. The nature of the experiment can be precisely controlled: the fact that the tax cut is temporary, how it is financed, what monetary policy is doing etc. But any answer is only going to be as good as the model used to obtain it. A prerequisite for a DSGE model is that all relationships have to be microfounded in an internally consistent way, and there should be nothing ad hoc in the model. In practice that can preclude including things that we suspect are important, but that we do not know exactly how to model in a microfounded manner. We model what we can microfound, not what we can see.

A specific example that is likely to be critical to the impact of a temporary income tax cut is how the consumption function treats income discounting. If future income is discounted at the rate of interest, we get Ricardian Equivalence. Yet this same theory tells us that the marginal propensity to consume (mpc) out of windfall gains in income is very small, and yet there is a great deal of evidence to suggest the mpc lies somewhere around a third or more. (Here is a post discussing one study from today’s Mark Thoma links.) DSGE models can try and capture this by assuming a proportion of ‘income constrained’ consumers, but is that all that is going on? Another explanation is that unconstrained consumers discount future labour income at a much greater rate than the rate of interest. This could be because of income uncertainty and precautionary savings, but these are difficult to microfound, so DSGE models typically ignore this.

The Fed model does not. To quote: “future labor and transfer income is discounted at a rate substantially higher than the discount rate on future income from non-human wealth, reflecting uninsurable individual income risk.” My own SEM that I built 20+ years ago, Compact, did something similar. My colleague, John Muellbauer, has persistently pursued estimating consumption functions that use an eclectic mix of data and theory, and as a result has been incorporating the impact of financial frictions in his work long before it became fashionable.

So I suspect the Fed uses a SEM rather than a DSGE model not because they are old fashioned and out of date, but because they find it more useful. (Actually this is a little more than a suspicion.) Now that does not mean that academics should be using models of this type, but it should at least give pause to those academics who continue to suggest that SEMs are a thing of the past.

ZLB Models?

There was a little interchange between Noah Smith and Paul Krugman a couple of weeks ago on what kind of models could explain Japan’s stagnation, and perhaps by implication the Great Recession. (Original Noah post here, Paul’s response here, and second round here and here.) I thought it was interesting, but it has taken me a bit of time to put my finger on why I thought it was interesting.

Noah began by saying there were two dominant macro models: RBC and New Keynesian (NK). The problem with applying NK models to Japan is that in NK models recessions last for as long as it takes for prices to fully adjust.  So how can NK models explain a lost decade or more? (You see this now in economists asking ‘how can the US, UK or Eurozone still be in a demand induced recession, from a shock that occurred 5 years ago’? Often the implication is that this is implausible, so the explanation must be supply side.) The answer, as Paul pointed out, is the Zero Lower Bound (ZLB). Noah replied that “They [ZLB models] are not yet well-developed or well-explored”.

Now I think Noah makes a lot of valid points, but I was unhappy about how his discussion was framed. I should also say that this framing is common to a lot of macroeconomists, so if I think it is unhelpful it is important to understand why.

It is often said that NK models just add price stickiness to RBC models, and if prices are sticky in the short run, aggregate demand matters in the short run. [1] I like to express it differently. What is the mechanism by which we can or cannot ignore aggregate demand? That mechanism is monetary policy, and how that is influenced by price adjustment. The way NK models can work is that price adjustment induces a monetary policy response, and it is the monetary policy response that ensures demand shortfalls are not persistent. Break the monetary policy response, because you hit the ZLB, and you break the correction mechanism, particularly if the monetary policy regime also involves inflation targets.

The ZLB therefore allows NK models to generate much more persistent recessions, if the recessionary shock is itself large and persistent. But the implications of the ZLB for RBC models are just as profound.  Implicit in their construction is that demand shocks ‘do not matter’, because the correction mechanism to get demand back to supply works sufficiently quickly that we can just focus on supply decisions. If the correction mechanism is broken because of the ZLB, then the foundation on which the model is built becomes problematic. It is no good saying ‘we assume price flexibility’, when even if prices adjust rapidly monetary policy cannot get demand back up. Or to put it another way, you cannot assume that the real interest rate will always be at the natural level if there is no way that real interest rate can be achieved.

That is one of the benefits (there are also costs) of the NK model encompassing the RBC framework. We can see the conditions under which the ‘special case’ of RBC works. And at the ZLB with inflation targets, it does not.

Of course you can ignore this point, and try to use RBC models to explain the current recession or Japan’s lost decade. But there are two huge problems with this. First, it ignores a big piece of evidence - these economies are at the ZLB! Well, that could just be a coincidence, or an inconsequential by-product. But second, the ZLB under inflation targets undercuts a key principle on which RBC models are built. In that sense, the model is not microfounded. [2] Thinking about mechanisms rather than models helps you see that second point. [3]

We can use NK models to analyse the implications of the ZLB, by hitting them with a large and persistent negative demand shock of some sort and adding the ZLB constraint. But what is clearly missing here is any understanding of the large and persistent negative shock. There is much current work looking at ‘financial frictions’, and the balance sheet implications that these may have. This may help explain the persistence of ZLB recessions. But they may also explain much more, and help improve the ability of NK models to track trends before the Great Recession. So to describe this endeavour as ZLB modelling seems inappropriate (or at least premature).

This approach to modelling ZLB recessions still has a unique steady state, and sees prolonged recessions as involving a natural real interest rate below its steady state value. An interesting possibility is that the ZLB constraint can create an alternative steady state, where a positive real interest rate is associated with deflation (see this paper by Mertens and Ravn (pdf), for example). The central bank (unlike Milton Friedman) is not happy with this steady state, because inflation is below target, but cannot shift to its preferred steady state by lowering interest rates. Whether you would call this alternative steady state a recession, and whether it could be applied to Japan, are interesting questions.

I do not think it is very informative to describe both this approach, and the more standard persistent demand shock approach, as ‘ZLB models’? The mechanism behind a persistent recession in either case is very different. But more fundamentally, they both use similar NK models, but just take the ZLB constraint seriously in that model. So it seems very odd to talk about NK models on the one hand, and ZLB models on the other, when the ZLB is an undeniable fact.

Now at this point you may be thinking that I am just being a bit pedantic about labels. I am not sure I should apologise if I am, but I do have another motivation. Talk of different models that can be applied to the same problem harks back to ‘schools of thought’ days in macro. I think macro should be better than that now. For better or worse, the microfoundations project and the new neoclassical synthesis gave us a common language, where we could talk about different mechanisms within a shared approach. That should make the process of matching evidence to theory more straightforward.

[1] Of course NK models often ignore the capital accumulation process, which is much more central to RBC analysis. But the key point is that we can always add sticky prices to any RBC model.

[2] There could be some other mechanism which justifies ignoring aggregate demand, but the whole point of microfoundations is that this mechanism needs to be spelt out. In its absence, all that is left is to just assume that large negative demand shocks never happen. Which is a bit like assuming nominal interest rates can be negative. 

[3] Chris Dillow’s comment that I link to here was really helpful in allowing me to appreciate why seeing macro in terms of competing models can be so confusing. In a way I just had to remember what it felt like learning macro for the first time, but that is easy to forget when you spend the rest of your life building and analysing these things.

Data, Theory and Central Bank Models

As the Bank of England finally publishes [1] its new core model COMPASS, I have been thinking more about what the central bank’s core model should look like. (This post from Noah Smith I think reacts indirectly to the same event.) Here I will not talk in terms of the specification of individual equations, but the methodology on which the model is based.

In 2003, Adrian Pagan produced a report on modelling and forecasting at the Bank of England. It included the following diagram.

One interpretation is that the Bank has a fixed amount of resources available, and so this curve is a production possibility frontier. Although Pagan did not do so, we could also think of policymakers as having conventional preferences over these two goods: some balance between on the one hand knowing a forecast or policy is based on historical evidence and on the other that it makes sense in terms of how we think people behave.

I think there are two groups of macroeconomists who will feel that this diagram is nonsense. The first I might call idealists. They will argue that in any normal science data and theory go together – there is no trade-off. The second group, which is I will call purists, will recognise two points on this curve (DSGE and VARs), but deny that there is anything in between. I suspect most macroeconomists under 40 will fall into this group.

The purists cannot deny that it is possible to construct hybrid models that are an eclectic mix of some more informal theory and rather more estimation that DSGE models involve, but they will deny that they make any sense as models. They will argue that a model is either theoretically coherent or it is not – we cannot have degrees of theoretical coherence. In terms of theory, there are either DSGE models, or (almost certainly) incorrect models.

At the time Pagan wrote his report, the Bank had a hybrid model of sorts, but it was in the process of constructing BEQM, which was a combination of a DSGE core and a much more data based periphery. (I had a small role in the construction of both BEQM and its predecessor: I describe the structure of BEQM here.) It has now moved to COMPASS, which is a much more straightforward DSGE construct. However judgements can be imposed on COMPASS, reflecting a vast range of other information in the Bank’s suite of models, as well as inputs from more informal reasoning.

The existence of a suite of models that can help fashion judgements imposed on COMPASS may guard against large errors, but the type of model used as the core means of producing forecasts and policy advice remains significant. Unlike the idealists I recognise that there is a choice between data and theory coherence in social science, and unlike the purists I believe hybrid models are a valid alternative to DSGE models and VARs. So I think there is an optimum point on this frontier, and my guess is that DSGE models are not it. The basic reason I believe this reflects the need policymakers have to adjust reasonably quickly to new data and ideas, and I have argued this case in a previous post. 

Yet I suspect it will take a long time before central banks recognise this, because most macroeconomists are taught that such hybrid models are simply wrong. If you are one of those economists, probably the best way I can persuade you that this position is misguided is to read this paper from Chris Carroll. [2] It discusses Friedman’s account of the permanent income hypothesis (PIH). For many years graduate students have been taught that while PIH was a precursor to the intertemporal model that forms the basis of modern macro, Freidman’s suggestion that the marginal propensity to consume out of transitory income might be around a third, and that permanent income was more dependent on near future expectations than simple discounting would suggest, were unfortunate reflections of the fact that he didn’t do the optimisation problem formally.

Instead, Carroll suggests that the PIH may be reasonable approximation to how an optimising consumer might behave as they anticipate the inevitable credit constraints that come with old age. There is also a lot of empirical evidence that consumers do indeed quickly consume something like a third of unexpected temporary income shocks. In other words Friedman’s mix of theoretical ideas and empirical evidence would have done rather better at forecasting consumption than anything microfounded that has supplanted it. If that can be true for consumption, it could be true for every other macroeconomic relationship, and therefore a complete macromodel.

[1] A short rant about the attitude of the Bank of England to the outside world. COMPASS has been used for almost two years, yet it has only just been published. I have complained about this before, so let me just say this. To some senior officials at the Bank, this kind of lag makes perfect sense: let’s make sure the model is fit for purpose before exposing it to outside scrutiny. Although this may be optimal in terms of avoiding Bank embarrassment and hassle, it is obviously not optimal in terms of social welfare. The more people who look at the model, the sooner any problems may be uncovered.  I am now rather in favour of delegation in macroeconomics, but delegation must be accompanied by the maximum possible openness and accountability.

[2] This recently published interview covers some related themes, and is also well worth reading (HT Tim Taylor).

Some notes on macro modelling

             This post is prompted by this post by Robert Waldmann, and this by Noah Smith, commenting on an earlier post by Wieland and Wolters. 

1) Forecasting and policy analysis

Noah repeats what is a standard line, which is that microfounded models are for policy analysis and not forecasting, and for forecasting “we don't need the structural [microfounded] models, and might as well toss them out”. The reason he gives is policy invariance: microfounded models address the Lucas critique.

While the Lucas critique is important, it is not in my view the reason we have microfounded models. The need for internal consistency drives the microfoundations project. Often internal consistency and addressing the Lucas critique go together, but not always. The clearest example is Woodford’s derivation of a quadratic social welfare function from agents’ utility. This is not needed to address the Lucas critique, but it is required for an internally consistent analysis of what a benevolent policy maker should do.

Why is internal consistency important? Because we think that agents in the real world are internally consistent, so models that are not can make mistakes. They can make mistakes in forecasting as well as policy analysis.

However, in an effort to achieve internal consistency, we may well ignore important features of the real world. ‘Ad hoc’ models that capture these features may be better models, and give better policy advice, even though they are potentially internally inconsistent.

So microfounded models could be better at forecasting, and ‘ad hoc’ models could give better policy advice. In that sense I think Noah is repeating a common misperception.

2) On a pedantic point, there is a long tradition of comparing different macromodels, both for forecasting and policy analysis, so Wieland and Wolters is hardly a first step. In the UK for 16 years we had an excellent research centre that did just that, run by Ken Wallis. There is a wealth of expertise there, which anyone doing this kind of comparative analysis needs to tap.

3)  Just in case anyone reading Robert’s post gets the wrong impression, the idea of the core/periphery structure for the Bank of England’s model came from economists at the Bank (strongly influenced by the antecedents from other central banks that I mentioned in my post), and not me. My role was mainly to give advice on theoretical aspects of the model to a very competent team who needed little of it. However Robert and Noah are wrong to suggest that because the Bank uses the core/periphery structure for forecasting, there is no point in having the microfounded core. For example, you can do policy analysis with both the complete model and just the core.

4) This final comment is just for those who read Robert’s post, and is very pedantic. Robert starts off by saying “As far as I can tell, Simon Wren-Lewis has been convinced by Paul Krugman”. The first point is that all my posts on this issue have come from a consistent view. I think microfoundations modelling is an important thing to do, but I do not think it is the only valid way of modelling the economy and doing policy analysis. I think Paul Krugman and I are on absolutely the same page here, and always have been. Robert is however right that my aim has been to convince those doing microfounded modelling of this point.

I’ve disagreed with Paul Krugman (and Robert) on the empirical success of the microfoundations approach, and I still disagree. But given that we agree that analysing microfounded models is useful, I don’t think this is terribly important. I picked up on the ‘mistaking beauty for truth’ phrase, because – taken literally – I don’t think that this is a problematic force behind the way the microfoundations project progresses. All scientists like simplicity, and they also get complicated when they need to, and DSGE models do the same. What I think is problematic is the weak role played by external consistency that I illustrated here, and the role of ideology. On the latter I think I’m once again on the same page as Paul Krugman.