We should have all copied the Swedish debt management office

I heard an interesting tale during my Riksbank visit, from Karolina Ekholm, Deputy Governor.  She explained to me that in the immediate aftermath of the crisis, the Swedish debt management office were OVER-issuing government debt [ie selling more than they needed to finance the deficit] and using the proceeds to finance the acquisition of riskier, private sector assets.  So they were doing negative QE.  If you accept the line of argument pushed by Vissing-Jorgensen and Krishnamurthy, and also Caballero and Farhi, that there is a special demand for safe assets with duration, then this is just what the authorities  should be doing at times of heightened financial stress and low risk tolerance.  Central banks commonly use the event study analyses showing that QE lowered yields (raised prices) on government bonds as evidence that what they were doing was a success.  But, on the contrary, this impact on yields could measure not the benefit, but the cost imposed by starving the economy of safe assets and of foregoing the benefits of the better, Swedish variety.’

Advertisements

Debt-ceiling fights destroy money, but strengthen monetary policy

In this post I explained why I thought that the fight over the debt ceiling acted like a monetary contraction (as well as, ironically for the Tea Party, acting like a wealth tax).  The argument, made by others too, is just that it reduces the usefulness of Treasuries as a means to park your wealth where it can be quickly turned into something else, as and when you want, and at a price you can predict.  It occurs to me that at the same time as this was going on, the fight also enhances the strength of unconventional monetary policy.  QE is argued by some to be a weak instrument because it involves swapping one zero interest, default risk free, money-like asset (reserves) for another (Treasuries).  So it leaves private sector balance sheets little different from before.  However, the debt-ceiling fight clearly made reserves and Treasuries less close substitutes. Treasuries became less default-risk free and less money-like.  At the same time as more monetary easing is needed, to make up for the money destruction, the liquidity-injecting properties of future monetary policy are enhanced.  Of course, the Tea Party wouldn’t like that either, because they are against the Fed interfering with monetary policy too.

In defence of panning Forward Guidance

Simon Wren Lewis writes ‘in defence of forward guidance‘.  He scolds Chris Giles, myself and others for ‘panning’ the Bank of England, for fear that it might retreat away from this bold experiment back into its natural, conservative, and insular mode of being.

Well, first, to recap on why I think they need panning.

1.  The MPC deserve a panning for not attempting Forward Guidance proper when it would have had most effect, shortly after Lehman’s.

2.  Mark Carney deserves a panning for appearing to take a position on the necessity for extra stimulus before he had assumed his role as chair of the MPC.

3.  He deserves another for refusing simply to back down, when he realised he could not carry others on the MPC with him.  The only way to explain the ‘no stimulus’ forward guidance policy is as a way for him to save face.  Central bankers with meteoric careers should be above needing to save face.

4.  The MPC as a whole deserve a panning for not realising that the garb of forward guidance should have been saved for when it was needed.  As I wrote in my mock Carney speech, they risk being misunderstood or not believed when they engage in Forward Guidance proper.

5.  Lastly, the Treasury deserve a panning for causing the mess in the first place.  The expectation is that central bank independence allows them to decide on operational matters, whether more stimulus is needed or not, and how, if it is, it should be injected.  The Treasury’s insistence on a review was designed to raise the costs of not undertaking Forward Guidance proper, and of not stimulating further.  If it was not designed with that intention, it was extremely ill-judged not to realise that it would have that effect.  Or, if that was realised, it was poor judgement to think that the other benefits of requesting a review outweighed the effect of raising the costs of not injecting more stimulus.  Although, come to think of it, perhaps it wasn’t poor judgement.  It was the short-termist political calculation that it was better to try to cave in central bank independence by kicking the MPC than to risk backtracking on monetary policy.  The former would only damage future monetary policy forever.  The latter would lose the Coalition this election for sure.

What is there to say in defence of the MPCs version of forward guidance?

1.  The eventual outcome reflects well on the MPC, and on the institutions of monetary policy.   Despite all the strain put on those institutions by Carney’s pre-Gubernatorial utterings, and the Treasury’s intrusion into operational matters, the MPC stood firm.  For a while most of the media seemed to think that the economy had been given a kick-start.  (I worry rather unscientifically that most of the country still do think this).  But, in fact, the MPC did not buckle.   Monetary policy is somewhat wounded, but will live on to fight another day.  A positive way to look at it is that the Treasury tried to bully the Bank of England into making up for its own fiscal policy incompetence by loosening monetary policy further.  But the Old Lady was not for turning.  If the MPC can stand up to this blatant trespassing on operational independence, it looks all the stronger for it.  Maybe people will conclude that the Treasury won’t try to play dirty again.

2.  The legacy of the whole affair will probably be that the MPC will find it harder to avoid communicating about future interest rates in the future.  This extra transparency will have many benefits, extolled elsewhere at length.  People will come to understand monetary policymaking and what the Bank does better.  The Bank will acquire extra legitimacy through exposing itself to more effective accountability, leaving the institutions governing not just monetary policy, but its other spheres, more robust.

3.  An even more profound legacy, harder to appreciate from outside, is that it will help reinforce to future incumbents of MPC office that you can’t do monetary policy without forming forward plans for your instrument.  This was not understood by many previous holders of that office.  Many staff members who tried to explain this were dismissed as hopelessly impractical nerds.  Fortunately, in the business of oil-tanker piloting, or rocket science, grasping that forward plans is seen as a necessity, not something that disqualifies you from the job.

4.  If Forward Guidance proper, of the lower for longer variety, is ever needed in the future, MPC will presumably find it easier to resort to it.  That’s a good thing.

What about Simon Wren-Lewis’ worry that the BoE needs gentle treatment?  Is there any truth in this?  Should we go easy on them, sparing them criticism that is due, in case it discourages further, more beneficial changes?

1.  If this were true, I’d suggest that it was grounds for serious outside action, not a case for scribblers like me to keep schtum.  It would surely be a sign of gross incompetence if an important organisation eschewed taking substantively justified steps to reform itself because it was worried about how outsiders would react.  As an aside, I don’t think the Bank is inherently conservative.  It has undergone radical convulsions in the last two decades.  It has suffered from the fact that people don’t like to admit they are wrong.  So once a policy decision is made, it’s hard to change it without a change of guard.  But the institution that governs term-lengths takes due care of that, in my view, balancing that problem with the need for each new leadership team to have enough time to decide to do something and persuade the staff to do it.

2.  I think there are good examples in the past to point to.   Chris Giles’ pointed remarks about the Bank’s fan charts are a good example.  These set analysts in the Bank scratching their heads and consulting statistics textbooks to come up with new fans, and prompted interesting new developments that might even constitute an improvement.

3.  The MPC did a good job of cheerfully ignoring everything I wrote when I was inside the building.  I doubt they will pay me any more heed now I am outside of it.

Either Forward Guidance IS a change in the reaction function, OR it’s misleading and empty

MPC have stressed that their version of Forward Guidance, unlike the original Woodfordian policy to which the term normally refers, is quite different.  Unlike the Woodfordian policy, the Bank of England are not communicating that rates would be lower for longer than one would predict based on a normal, historical reading of what the MPC cares about and how it sets about achieving it.   There’s no change to the MPC reaction function, we are meant to understand.  All that FG does is clarify what they were doing anyway.

Really?  How does that work?

Since when have MPC followed a scheme for responding to events as they unfold which amounts to not even considering responding to news, at all, unless unemployment moves through some threshold?  For example, since inflation targeting started in 1992, we know that you can fit a Taylor Rule (augmented with, say, a lag in the central bank rate) to the policy rate, and it will capture the main features of what the Bank of England tried to do.  That rule will have rates responding to real and nominal things.  You won’t get a better fit through those observed interest rates by trying to add indicator variables turning off the response to inflation unless an inflation forecast breaches 2.5 [a reference to the inflation ‘knockout’].  So unless the communication around forward guidance is completely empty, it DOES constitute a change in the reaction function.  [Thanks to Richard Barwell of RBS for articulating these points so clearly to me over coffee.]

And are we meant also to consider that the MPC foregoing the normal convention of reconsidering their policy stance afresh each period amounts to not changing anything relative to before?  If so, what is the point of ditching the ‘face each ball as it comes’ language?  Unless that commitment is empty, this also constitutes a change in the MPC’s reaction function.  Either they have laid out a data-contingent plan and intend to stick to it, as they say, in which case this is different from what went before, when there were no such pre-agreed plans, or they haven’t, in which case the words are hollow.

So, in short, either forward guidane does constitute a change in the reaction function, contrary to some of what has been claimed by MPC themselves.  Or it does not, in which case the commitment made is empty.  Which is it?  Can someone explain?

One way out for the MPC is by reference to the ‘unwarranted’ rise in interest rates at that the MPC described as priced in to the yield curve at the August MPC meeting.  Perhaps all the terminological tomfoolery is about trying to get the yield curve to behave as it always used to, for the two decades prior to August?  Perhaps that is why the MPC needed to borrow a policy label used for something else, conjor up ‘wayposts’ and ‘knockouts’, to tether the yield curve so that it sways in the economic wind just as it always used to.  No, that doesn’t seem to ring true either, does it?  Are we to believe that the yield curve always did pretty much what MPC wanted it to, until now, when a new policy framework was needed?  This is not very believable.  The summer giration was hardly that unusual.  What happened is that a new Governor arrived, one prepared to talk about future interest rates and compare them to the yield curve when previous Governors were not.  Unfortunately, that same new Governor had also boxed himself into needing a new policy framework, or begin his Governership with a defeat.

The speech Mark Carney should have given but didn’t

Folks.  I found myself in a tricky position when I arrived.  The bosses in Whitehall were really taken with this thing about keeping interest rates lower for longer, and, frankly, since there was no way they could loosen fiscal policy, they were in the mood to try anything.  They had backed themselves into a corner by saying that there was going to be no more fiscal loosening.  And though the economics of this sucked, the politics was clear.  And maybe, at a stretch, you could understand it, since it was a legacy of the pretty scary times they started out in back in 2010.  So HMT were really hoping to get more stimulus through Forward Guidance.  They knew I was into it and we hit it off.

They made a good show of being cautious about appearing to change the monetary framework, commissioning a review.  But they knew what I thought already – I’d gone on record about it -and we both knew what the answer was going to be.  Except, when I got here I found that there was a majority against any more monetary stimulus.  I was hoping to persuade them but I couldn’t.  It’s a little tough to accept, since I started out thinking that this was what was going to ‘secure the recovery’, but the reality is that the other MPC members think the recovery is in safe hands.  So our review is an in principle review.  I think that many of my colleagues accept that if we needed more monetary stimulus, it would be worth having this tool in your kit.  However, most think that we don’t need to try to engage in ‘lower for longer’ right now.  But we are not going to leave things there.  I have managed to persuade my colleagues that we should be more open about future interest rates in general, and about what we are trying to do with our policy instruments.  The former Governor had strong views about this, but we think it’s time for a change.  The Reserve Bank of New Zealand have been publishing interest rate forecasts for two decades without any difficulty.  The Scandinavian central banks are facing challenging times, as market forecasts of what they will do have differed from official ones for some time.  But on balance we feel that this has proved educative.  Why shouldn’t those forecasts differ?  And the debate there has focused on the economics of how the conjucture will play out, not the competence or credibility of the institution.  So, this is what we will do from now on.  Moreover, we are going to explain more about how we arrive at our policy plans.  For instance, although everyone has got the idea that the central bank should stabilise the real economy, as well as inflation, we have been totally silent, and, frankly, evasive, up until now about just how much weight we should put on each concern.  There’s no great clarity in the literature about this, but previous regimes have tended to hide behind this observation to avoid taking a punt.  But without taking a punt, frankly, it’s impossible to know what should be happening to your policy instruments.  [How were all those votes cast before, you might very well ask yourselves, and how did our Treasury bosses tolerate handing over all that power without even specifying what we should do with it, or asking us to be clear about that?].  So, taking that punt, we are going to go for two to one weights in favour of inflation.  The consensus in the modern macro literature is for weights much higher than that.  They have to do with the models’ assumed very high costs of invalidating the forecasts of firms whose prices can’t adjust period by period.  We think those very high costs probably over-egg it.  After all, no-one seriously believes firms will not change prices no matter what.  Those comments need a proper dive into the technicalities to explain them properly, and we’ll do that, but for now take it on faith.  Two to one might change, as we learn more about what that means for our policy plans.  Having set out (actually for the first time!) what our objectives are, we next need to present you with some tools we use to decide on our plans, and to help you reflect on how we do it too.  To begin with, we present a range of simple policy rules, the coefficients of which have been honed to achieve the best on the two to one weights objective, given how we see the economy.   And, just to probe things further, we’ve repeated the exercise for slight changes of the two to one weights assumption around our base case, and interestingly, also for a two to one assumption replacing price inflation with nominal wage inflation.  (One of the predictions from these models is that central banks should care as much about nominal wage inflation as price inflation).  Obviously, these rules can’t capture all the complexity of what we think we face, particularly now.  And you can see our central forecast for interest rates differs from what all of the rules suggest.  But we go into reasons for that in our Inflation Report.  To reinforce these openness steps, we are providing all the code and judgements applied to the Bank’s suite of forecasting models that are used to produce each forecast.  That way you’ll be able to check that we are being honest, and see how we use and abuse these models.  Amongst other things, you’ll be able to see important things like what the assumed effect of the Funding for Lending Scheme is; what the assumed effect of changes in capital requirements made by FPC is;  the assumed effect and transmission of quantitative easing.  So you can get an idea how all the different instruments interrelate.

So, the upshot is, we are not engaging in Forward Guidance as other central banks have chosen to use the term, ie we are not trying to keep rates lower for longer than you would guess we would based on normal modes of setting interest rates.  That’s because there is a majority against injecting more monetary stimulus.  We have all of us agreed that there may come a time in the future when this kind of policy would be effective.  Probably, the best time to do this would have been when it became clear that interes rates were going to fall a lot after the Lehman’s Crisis.  That was when the yield curve was still projecting steep rises from their trough, fairly soon after the crisis hit.  So there was a lot to be gained by flattening that curve.  Some of my colleagues pointed out that embarking on this policy now, so long after interest rates have already been pinned at the zero bound, would be to do so at exactly the time when this policy would have least effect.  So, we are not doing Forward Guidance.  But we are entering a new era of transparency about what we do.  This will make monetary policy ‘more effective’, in the following specific sense.  It will make it easier for you to judge whether we are keeping our promises and acquitting the important duties assigned to us.  It will bring the technology of holding central banks accountable up to the standards now expected of modern democracies.  It will hopefully guard against people worrying that we have gone soft on inflation, because you’ll have a better idea what we are doing, and why allowing inflation to be consistently higher than target for so long is still the right thing to do, despite our remit.    It’s extremely important for us that our new policy is not confused with the Woodford lower-for-longer policy, and that for that reason we will completely eschew the ‘Forward Guidance’ language.  This is semantics.  But important semantics.  Because we don’t want to mistakenly inject monetary stimulus we collectively judge not to be needed.  And, more importantly than that, if we are to keep lower-for-longer Forward Guidance as an effective tool for the future, we don’t want to blow it by somehow riding on the coattails of other central banks and pretending what we are doing is like what they are doing.   If we did that, and we were found out by close questioning that there were no more monetary stimulus, we might get a reputation for trying to spin our policies for things that they were not, which would be an extremely hazardous thing to carry forward into the near future.  Although my colleagues are against more monetary stimulus now, they recognise that we might have to rethink in the future.  For example, there are great risks of things working out in the Eurozone badly over the next couple of years.  If that happens, we will need every monetary stimulus tool at our disposal, and our reputation for plain speaking intact.  That would be the time to present a tool that we label ‘Forward Guidance’.  And when we do, we want people to take us at our word that more stimulus was intended.  If we presented a ‘no-stimulus’ Forward Guidance policy now,  there would have been a risk that people thought this was Mr Carney rising to rescue the recovery and then finding out that I wasn’t doing anything at all. Down the road, when the recovery really needed rescuing, and we launched the ‘kick start Forward Guidance’, they might then think we were trying to pull a fast one again, and wrongly conclude that we were in fact sitting on our hands again.  This would be a hoax on the scale of trying to pretend that buying gilts with electronic money was some great new stimulus policy, when all that was going on were some fancy open market operations!

A recap on the Forward (or Fudged) Guidance communications fiasco

On Friday, Bank of England Chief Economist Spencer Dale is set to respond to twitter questions posted to #AskBoE.  This seems like a useful time to recap on previous posts on the fiasco, as I see it, surrounding the launch of forward guidance (FG).  All this has been said before, by me, and others.   But there is no harm in repeating questions that are begged, especially when they don’t get answered the first time.

1.  Forward guidance presumably must have been a ‘inject no more’ stimulus policy, otherwise how would the hawks agree to it?

2.  If it was a policy to inject no more stimulus, why bother with all the smoke and mirrors?  Was it worth risking faith in the monetary framework to inject no more stimulus?  Or risking that this was seen as a sop to a Governor-elect that had already declared himself in favour of a stimulatory forward guidance before taking up post, and would be embarrased to find he could not have it?

3.  If it was a policy to inject no more stimulus, it was mighty unfortunate, as it looked a lot like other forward guiding central banks’ policies, which were designed to inject more stimulus.  The similarity was hazardous and confusing.

4.  If it was a policy to inject no more stimulus, what does it mean that FG was meant to make policy ‘more effective’?

5.  If ‘more effective’, by contrast, does mean, ‘more stimulative’, and yet FG is meant to inject no more stimulus overall, where is the compensating tightening to leave the net stimulus the same?

6.  If ‘more effective’ does mean more stimulating, where are the quantifications of the impact of reducing uncertainty on the macroeconomy?  Most people would expect such effects to be very small. What prevents them from being small in this case?  Have the MPC reflected on the reaction to FG and deduced that all the benefits from reduced uncertainty about policy have accrued as planned?  If so, what evidence do they have to substantiate that.  If not, what do they intend to do about it?

7.  How can Governor Carney’s initial press conference which was full of almost hyperbolic language about securing the recovery be squared with an ‘inject no more stimulus’ version of FG?

8.  If the intention was to inject more stimulus, how can Carney’s answer to Faisal Islam’s question in that same conference be squared with that intention?.  Recall that he asked Carney whether FG should be considered a monetary loosening or not.  Carney pointedly avoided saying it was, and instead described it is an attempt to make policy ‘more effective’.

9.  If ‘more effective’ does not mean ‘more stimulative’, what does it mean?  Why is it so important that it is ‘more effective’ if it is not stimulatory?  Why does adopting a framework that is allegedly ‘more effective’ warrant the cost of all the confusion and uncertainty about the MPC’s long run intentions regarding the inflation target?

10.  If FG is not meant to be more stimulatory, how does that square its launch and evaluation relative to the initial request by HMT, the intention of which can clearly be read as a request to evaluate an instrument for stimulating the economy that the BoE had not yet deployed, in contrast to the Fed and the Bank of Canada.  (They did not say:  FG seems to have been very effective in stimulating the economies of the US and Canada.  Can you evaluate the equivalent policy tailored so that it does not stimulate at all, and get back to us in the August Inflation Report?)

11.  [Aside:  I wonder what HMT make of FG, if in fact it is not supposed to be  more stimulatory?!  Presumably it’s something of a disappointment.  All the charade of a formal review of a policy option that amounts to clarification?]

12.  The MPC are right to defend themselves against the accusation of a flop when the yield curve disagrees with them.  But they have yet to explain what they had forecast the reaction of the yield curve to be, and how they intend to respond to it.  [Watch for an answer to Chris Giles’ well-aimed #AskBoE question on this].

13.  How have the MPC coped with the extreme unreliability of their model (and all like it) when simulated under constant interest rates?  Where are we to find the extra uncertainty, for example, in the Fan Chart?

14.  If FG was meant to be more stimulatory, how do the Hawks justify it?  What changed their mind?  And just when the economic data was hotting up so clearly?  That data was deflty, if (for me) unconvincingly, described as a hotting up of demand and supply that were perfectly in harmony (and thus inflation-neutral).  But really, what else happened?

15.  If what changed their mind was the ‘unwarranted’ tightening in the yield curve, why have such factors not been mentioned before, nor any description of their effect on the policy decision been presented?  As Chris Giles asks in his #AskBoE tweet, are subsequent movements in the yield curve ‘warranted’ or not?  How much difference do the MPC estimate the unwarranted tightening made to anthing, eg, inflation?

16.  If FG was a ‘no more stimulus’ policy, and instead just about making policy more effective, (which we will assume for the moment just means, ‘better presented’, or ‘easier to understand’, but not implying any first-order difference in terms of stimulus), but yet the yield curve tightening was unwarranted, what was supposed to be the response to it?

17.    If FG makes monetary policy more effective now, why was it not introduced before?  Why did it take the request of the Treasury and the appointment of Carney to look into it?

18.  One presumes that before Carney arrived, there was a majority against using FG.  What persuaded those who were formerly against FG to support it subsequently?

19.  The MPC must have realised that there would be a risk that if nothing further were said, it would look as those that were formerly against FG had not in fact changed their mind, but were simply falling in behind the new Governor, matching popular press accounts of how the MPC works, contrary to the doctrine of individual responsibility that has reigned thus far.  What can the FG flip-floppers say to outsiders to explain their change of heart?

20.  In the absence of any explanation for why those formerly against FG subsequently supported it, there is a risk that observers would conclude that perhaps instead FG was simply never discussed.

19….. in which case, supposing it wasn’t, such an outside observer might wonder why it wasn’t discussed?  Were those supportive of FG before simply not given an opportunity to discuss it?  If not, why not?  Or were they too timid to raise it?  Were they putting aside their preferences in favour of the former Governor?

20.  In communicating FG, the Bank describes it as enabling the MPC to ‘explore the scope for economic expansion without putting price or financial stability at risk’.  What does it mean to ‘explore the scope for economic expansion’?  Did the MPC not guess that such a term has no recognised meaning amongst professional economists, let alone amongst those ‘down the Dog and Duck’?  In what sense was previous policy not able to undertake such exploration without its attendant risks?

21.  If FG was simply about clarifying what the MPC were already doing, (the corollary of FG not being equivalent of injecting more stimulus, and therefore not being about ensuring interest rates were ‘lower for longer’ a la Woodford), why all the fuss about staging posts?  If FG was simply about such clarification, why not (instead) make reference to the kind of rules that outsiders might use to try to forecast what MPC are doing, and compare what they actually do to those?  Why not explain how the MPC weigh its competing short run goals of inflation and real economic stability?  For example, why not tell us how much more worrisome is a deviation of inflation by one percentage point relative to  deviation of unemployment from its natural rate by the same amount?  If MPC have an answer to this question, why don’t they tell us what it is?  What argument could justify concealing their calculations about this from us?  (For that matter, what do the MPC’s custodian, the Treasury, think MPC should be doing? Why are they happy for this to be kept from us?).    If they don’t, how on earth are they setting monetary policy?  How can they be confident that the course they have set is going to achieve the best combination for their goal variables?

22.  FG is about expectations management.  This brings to the fore the question of what the MPC believe about how those expectations (of what it is going to do, of inflation, etc) are formed?  So what do they believe?  Academic work on forward guidance is set in the context of rational expectations (an assumption that means that agents in the model know all there is to know about the model).   MPC have distanced themselves from this work (in explaining that FG is not about lower for longer) and would anyway presumably distance themselves from the extreme assumption of rational expectations.  But they have replaced it with what?  And could the MPC tell us how they think expectations are amenable to MPC’s comunication and policy decisions?  Have these views been formalised and ecoded in the simulations underpinning the MPCs estimates of the likely impact of forward guidance, thus replacing the existing assumption in COMPASS, the Bank’s suite of forecasting models?  If not, how have they arrived at their view about the length of time it will take to reach the ‘staging post’ of 7 per cent unemployment?

One could go on.  It need hardly be said that this episode is laced with an unfortunate irony.  Forward guidance is, as the name suggests, about trying to explain what you are going to do in the future, ie, it is about using communication to enable others to understand better than they would otherwise be able what you are going to do.  Unfortunately, as I see it, FG has achieved exactly the opposite.  It looks, mechanically, to be a ‘no more stimulus policy’ from the outside.  Yet there was somehow an agreement struck that it could be presented as the opposite.  And that there would be no intervention to stem the tide of media coverage interpreting FG as the thing that would rescue the economy from the doldrums.  FG was no more stimulus for the hawks, but appears to have been spun as lots more stimulus for the media, or at least that is how it looks.  And in risking that this was how it looks the MPC put in jeapordy its reputation for plain speaking.  (This is why I called it ‘Fudged Guidance’.)  With inflation measurably, and, on balance, justifiably higher than target for some years now, and expected to be so for at least another two, now was not the time to tamper with its reputation for plain speaking.

There’s no war now. Pragmatic microfoundationists and an empirical macro that uses minimal theory has taken over.

Simon Wren Lewis’ latest post on Mainlymacro rails against what he sees as an overemphasis on microfoundations in macro models.  Paul Krugman picked up this post, and it was reblogged and retweeted by Brad DeLong and probably others, referring to the notion of ‘macro wars’, between those who microfound, and those who don’t.  Simon points in particular to the failure to incorporate downward nominal rigidity in wages, and other features that spring from behavioural economics.

To recap for the uninitiated, a model that is ‘microfounded’ is one in which you write down explicitly what it is the firms and consumers in the model are trying to do, and what constraints are placed on them when they try to do it.  Microfoundations came to be seen as important for a very good reason.  Namely Robert E Lucas Jnr’s work showing that models that didn’t articulate exactly what agents were doing, but instead just relied on correlations between macro time series, could give very unreliable policy advice.  The example topical at the time was the apparent tendency for unemployment to fall as inflation rose (the long run Phillips Curve).  A tendency that disappears (at least in post Lucas Critique models) if policymakers try to exploit it to lower unemployment permanently.  The risk with non-microfounded models is that all you have is a set of difference equations that don’t really have any economic meaning at all.

First point.  Some notable microfounders have looked at downward nominal rigidity in otherwise standard and microfounded real business cycle models.  For example, this paper by Schmitt-Grohe and Uribe, models the periphery of Europe as suffering from downward nominal wage rigidity;  a credit boom pumps up nominal wages, and then when the crash comes, unemployment soars because wages don’t fall.  Another example:  this paper by Junil Kim, looking a the optimal inflation rate.  Or take this paper by the two Maliars, prominent in the literature on numerical methods for real business cycle models, putting downward nominal wage rigidity into a heterogeneous agent model.   In finance, the idea is alive and well too.  For example this paper by Cohen and coauthors, which goes back to an old 1975 paper by Modigliani and Cohn, arguing that stock prices behave as though investors have nominal illusion.  (Probably one of the necessary ingredients for downward nominal wage rigidity).  Or this, on a similar theme, by Monica Piazzesi, a prominent finance academic who will have started out life learning Lucas’ microfounded model of asset pricing.

There are many other examples too where authors drop the assumption that agents are maximisers or know all about the workings of the model.  Tom Sargent, George Evans and Seppo Honkapohja made their living studying what happens when agents have to learn about the model in the same way that econometricians do.  It is common for modern DSGE models to have rule of thumb consumers that make no complicated intertemporal decision, but simply eat what they earn.

The practice of building in a friction into macro models without explaining where it comes from is extremely widespread.  The ‘New Monetarists’ that revolve around Randall Wright and Steve Williamson would say that all modern New Keynesian models fall foul of this.  The apparently microfounded practice of including (symmetrically) sticky prices and wages is a case in point.  Most users of these models acknowledge that this aspect is not seriously microfounded, and is in fact simply a way of getting real business cycle models to fit the data.  (Even Nobuhiro Kiyotaki, who co-invented it with Olivier Blanchard, called the sticky price part of the model ‘a fairy story’).

In the heterogeneous agents literature, which began as a direct outgrowth of the old Lucas/Kydland/Prescott business cycle models, a key aspect of the model is how agents can store their wealth, if at all.  What kinds of assets can they save into?  Many assume, realistically, that there are very few assets that the typical consumer has access to.  But without explaining why in the model (without microfounding in other words).  In their elegant summary of the state of the art in this literature, Heathcote et al call this a division between ‘model what you see’ and ‘model what you can microfound’.  The same issue colours open economy macro, where results differ markedly between environments where markets are complete and where they are assumed, without microfounding, that they are not.  The literature on credit frictions which the financial crisis has breathed life into also contains many examples of ad-hocery;  the assumption that firms issue standard non-contingent debt, and usually only one period debt.  Borrowing constraints that relate what someone can borrow to the value of their collateral.  All common sense, but not microfounded in the models themselves.

Simon worries that the influence of these microfounded models is too great in our policy institutions.  Well, I’d like to reassure him on that point.  In the Bank of England, generous use is made of other models, and judgement overlaid on the main DSGE model.  Second, many of those who produce the forecast and many of the policymakers have had no direct exposure to these models in their careers, and consequently don’t really understand them.  Many think exactly in terms of older, non microfounded models that Simon and PK would prefer.  There may be much less to worry about than Simon thinks! (At least in the UK).

Perhaps a third of macroeconomists either specialise or have a toe-hold in ’empirical macroeconomics’.  Here the bread and butter model is the vector autoregression. (A system where you regress everything on lags of itself and lags of everything else).  This tool was explained to macroeconomists by Chris Sims, Nobel laureate.  The whole point of it is to avoid making ‘incredible’ assumptions that emerge from particular models.  And instead to interpret them using assumptions that would be true of many models, perhaps all models that the researcher would accept as having something sensible to say.  Debates rage in this literature about just how safe these assumptions are, but this is the proper place for atheoretical modelling, Sims would say.  Not tacking together equations that the researcher thinks roughly describe how firms or consumers go about life.

Simon’s tag-wrestlers in this complaint about microfoundations talk of ‘macro wars’.  I think by this they conceive of macro as a battle between those who are content to use IS/LM or similar models, and those who want to use real business cycle models and their descendants.    But I don’t see it like this at all.  Although the blogging community might manifest this war, and although the IS/LM model is alive and well in introductory macro courses and textbooks, in the community of those publishing in peer-reviewed journals there is no longer any war.  The pragmatic microfounders and empirical macro people have won out entirely.  (At least if you ignore a few policy institutions!).