The German minotaur

I was interested to read Simon Wren Lewis‘ post on the dysfunction in the Eurozone caused by German improved ‘competitiveness’.

Is this right?  Well, one thing I learned from those in my team at the Bank of England who were experts at international macro [main tutors were De Paoli and Lipinska, now of the Fed], is that with relatively little effort, you can get pretty much whatever result you like.

So the answer is:  who knows?  One cannot speak definitively about German dysfunction.

Nonetheless, here is another Wren-Lewis like story, told in the language of Balassa-Samuelson, plus downward nominal rigidities.

Germany gets better at making tradable goods.  This increase in German wealth drives up the price of German non-tradeables.  The relative price of tradeables does not change much, as, in this sector, the law of one price holds more nearly.  For a given inflation rate in the Eurozone as a whole, this Balassa-Samuelson runaway effect [runaway being the opposite of the ‘catch up’ we thought was happening before the crisis started] should mean above average inflation in Germany, and below average inflation in Greece.

If the runaway effect is large enough, or, put differently, if ECB monetary policy fails to generate enough inflation to accommodate the runaway, this may mean negative inflation in Greece.  If there are downward nominal rigidities, the force that would otherwise generate negative inflation [in particular negative nominal wage inflation] generates high unemployment.

In this story, the solution is to set the inflation target such that it is higher, the greater is the relative price change that technology runaways requires.  I doubt very much that, even if this were possible, it would have been allowed by the ECB.

But there was the additional problem that the collapse in demand caused by debt-deleveraging, coupled with the zero lower bound, meant that the ECB could not even meet the 2 per cent inflation target, let alone generate enough to accommodate the problems described here.

Mind changes: post script

Actually, I wonder about this mind changing business….

Two correspondents, Julian Echave, and another, privately, questioned the crudeness of this concept.

One way of responding to what they said is to note that, for example, from a Bayesian perspective, the binary shift in beliefs is a bit of a nonsense.

Mind changing perhaps could mean, from this perspective:  given a binary choice between two policy options [say, minimum wage or no], my posterior probability mass on the event that a minimum wage of a given level would have such and such an effect on employment outcomes shifted – in the light of evidence – such that my binary choice shifted from preferring no minimum wage, to supporting one.

 

Mind changes

It’s fashionable now to get out there what you’ve changed your mind on.  The calculation being, perhaps, that unless you do, you will get nowhere arguing hard for your other, unalterable priors.

In that spirit, here are a few thing I changed my mind on.

1.  The importance of financial ‘plumbing’.  I, like most macro people I talked to, used to think banking in macro was cluttering up models for the sake of it.
2.  The lack of any effect of minimum wages on employment.
3.  The minor effects on wages of immigration.
4.  How sticky price DSGE models are not, clearly, realistic enough to take seriously as devices on which to perform quantitatively relevant optimal control exercises.  For example, I used to admire what the Norges Bank and Riksbank did and thought we at the BoE were well behind the curve.
5. The usefulness of atheoretical empirical macro.  I used to think we were surely beyond just fact generation and sharpening.  But, the crisis having shaken my faith in the models, fact generation now seems attractively humble.
6. The efficacy of macro prudential, counter cyclical tools.  It’s early days.  But when I first read about these in 2009 or so, I thought [extrapolating from the ‘life is just a New Keynesian optimal control exercise] ‘here are some great taxes we can tweak about, problem solved’.  Now I am not so sure.
7. The effects of technology on the wage distribution.  The ‘hollowing out’ stuff is fascinating and I would not have guessed at that.  I came into those papers thinking that those at the bottom always lost.
8. How far authoritarian capitalism could get before imploding.  I would not have guessed China could have increased income per head this much without imploding first.  Maybe it will, yet, proving Acemoglu and co-authors right.
9.  The likelihood of a zero bound episode, and the optimal rate of inflation.  I was signed up for 2 per cent.  But look!
10.  The efficacy of quantitative easing.  There are lots of question marks still over what it did, and how durably.  But the initial impacts, and the unanimity, were a big surprise to me.

It’s probably not wise to get too excited about the moral grandeur of me declaring mind changes on 2 and 3, which revolve around the labour market, as I’m one of those macro people who have definitely not read enough labour economics.  A bit like me writing ‘I changed my mind about how quantum physics worked.’  [Incidentally, on that, I have, several times.]  Note that I ommitted a socialism-capitalism mind change, on the grounds that that occurred before the end of adolescence.

Why weak nominal wage growth is of concern to flexible inflation targeters

Today’s UK labour market data show nominal wage growth falling back – to about 2% annual growth, from a local peak last year of 3% – at the same time as the employment rate increases.  This is another important interregnum for those who thought the UK economy was following a trajectory that would justify a first hike in rates and the beginning of the process of normalising monetary policy.

Recent speeches by both Vlieghe and Carney on the MPC warned that nominal wage growth and other inflation indicators had to show clear signs of picking up before they were prepared to raise rates.

But students of modern monetary policy models will know that these models suggest monetary policy pay attention to nominal wages not just in so far as they are potentially indicators of future inflation, but for their own sakes too.  In these models, stabilising nominal wages around their long run growth rate is just as important as stabilising inflation around its target.

The mandate itself does not give the MPC a nominal wage growth target.  But its signal to support the government’s objectives of achieving ‘strong, sustainable and balanced growth‘, and the March 2013 review of the mandate by HMT which explicitly characterised what MPC should be doing as ‘flexible inflation targeting’ gives policy makers ample scope, in my view, to place independent emphasis on stabilising nominal wages.

You might wonder whether having this independent concern for nominal wages makes a difference, if they are anyway an important indicator of future inflation.

There are 3 things to say about this.

First, central banks have a long history of debating in fact whether short run correlations imply that nominal earnings follow inflation, rather than signal future inflation.  If you care independently about nominal wages, this debate is less relevant.

Second, a shift in labour’s bargaining power [or equivalent concepts] could well mean that nominal wages decelerate at the same time as firms’ prices accelerate.  Caring independently about nominal wages means that one would not respond to such a change by simply accommodating all of the nominal wage fall.  Policy would be faced with a trade-off.  [This is topical as on some measures core inflation has surprised on the upside].

Third, without a motive to care independently about nominal wages, these data become just one out of several pretty noisy indicators of future inflation that it may be easy to find reasons to downplay.   Not so if policy ought to care about nominal wage growth for its own sake.

Letter to Danny Blanchflower re the McDonnell BoE mandate review

This letter sent to Danny today, should contain no surprises.  But it gathers together the various views I’ve ranted about on the framework / mandate for monetary policy in the UK.

At some point I will update with a version that provides links to old posts that go into each topic in more detail.

Gerard Macdonnell on QE

A quick response to this guest post by Gerard Macdonell on Noah Smith’s blog.

There are two main thrusts to the post.

One is that the stated claims for the efficacy of QE are false.  I’m not a fan of QE myself, and concur on some points, but I think the event study analysis is more persuasive than he does.

Second, Gerard assumes that the Fed’s move reveals that it really believes QE to be ineffective.  I disagree on that entirely.

Actually, I’m pretty sure that the reason that the Fed has raised rates without QE sales first is exactly the reason they stated, and also strategised in private.

That is that they don’t want QE to be the marginal instrument of monetary policy.

The plan is, and was, to begin QE sales when it was felt that there was enough clearance of interest rates over the zero bound such that, if the economy began to worsen, this could be dealt with using interest rates only.  This plan would avoid having to restart QE purchases.   Conducting QE sales and then reversing course would complicate debt management policy, and, said the operational analysts, generate uncertainty in the securities market itself.

Think about it:  if the Fed really thought that QE asset stocks were not doing anything, they would surely be getting rid of them as fast as possible, as they have generated political hostility, and selling would be a costless way of eliminating the criticisms that come from the conservative permahawk lobby who perceive a tail risk in the form of a large surge in inflation on account of the extra money printed.

If you thought that only flows of QE mattered, but not stocks, then that would also be a motive for following the current Fed strategy, and so would rationalise Gerard’s interpretation of the Fed’s behaviour.  But in this case you have to believe in a watertight conspiracy to fail to explain the real reasons behind policy, participated in by all the voting and non-voting FOMC members, and avoiding leaks from disappointed staff.

Gerard thinks that if the Fed did still believe in the efficacy of QE, there would be far too much stimulus out there, bearing in mind that the Fed has told us rates won’t rise either fast or far.  However, this flies somewhat in the face of facts.  Although unemployment continues to fall in the US, there is credible evidence that there is ‘missing unemployment’ in the form of those outside the labour force who want to enter it and will, so Gerard’s claim that the US is close to ‘full employment’ is controversial.  And the Fed’s preferred inflation measure is weak, provoking the ire of Summers and co who think that the Fed’s rise in rates is premature anyway.  The data provide a pretty good case for there not being too much stimulus out there now.

Gerard also highlights what he thinks is an economic error in his thinking, what he deduces is a ‘misapplication of the quantity theory’.  He cites this passage from a 1999 paper.

“The general argument that the monetary authorities can increase aggregate demand and prices, even if the nominal interest rate is zero, is as follows: Money, unlike other forms of government debt, pays zero interest and has infinite maturity. The monetary authorities can issue as much money as they like. Hence, if the price level were truly independent of money issuance, then the monetary authorities could use the money they create to acquire indefinite quantities of goods and assets. This is manifestly impossible in equilibrium. Therefore, money issuance must ultimately raise the price level, even if nominal interest rates are bounded at zero. This is an elementary argument, but, as we will see, it is quite corrosive of claims of monetary impotence.”

The ‘misapplication’ Gerard homes in on is the statement that money ‘has infinite maturity’.  I read him as wanting to dispute that in the case that the economy turns out to need, subsequently, a reversal of the QE operation that created the increment to money.  But this simply misreads what Bernanke meant by ‘infinite maturity’.  What he meant, I have no doubt, was that there was nothing contractual in the ‘contract’ that the money embodies that triggers a future settlement of money in the form of something else at any particular date.

However, in pointing this out, we do get into territory where this passage raises some theoretical and policy issues of some controversy.  I don’t fully grasp what is bugging Gerard, but it might have something to do with the following, which recaps material briefly I have gone over before.

This gets to something that is actually ambiguous, though not necessarily incorrect, in Bernanke’s text.

In the standard model of money and finance, issuing more money will stimulate prices even at the zero bound if this issuance corresponds to a lowering in expected future rates.  If, on the other hand, expected future rates are already as low as possible, this won’t happen.  In the standard model, money is treated as a liability like bonds.   And so indefinite unlimited issuance is not possible or consistent with fiscal promises.  If attempted, agents believe that they will be reversed at some point, and there is no stimulus.

Policy makers have sometimes pointed to the fact that, in practice, money is not, or not considered a liability of anyone.  So its issuance [leaving aside the detail of interest on reserves] doesn’t create a liability.    Willem Buiter has a paper that formalises this logic.  In these circumstances, money issuance stimulates regardless of whether interest rates are expected to be at the zero bound forever or not.

For central bank historians interested in Bernanke rune-reading, the potential ambiguity in Bernanke’s text quoted above is this:  It’s not clear whether he means that money should not be included as a liability in infinite horizon consolidated central bank/government budget constraint, or whether he just means that money does not fall due at some particular date.  If he meant the latter,  then his statement is only true, theoretically, if future rates are not currently expected to be zero forever.  If he meant the former, then what Bernanke said is true regardless.

I say ‘regardless’:  this I’m just surveying two classes of non-micro-founded money models here.  Whether it’s true in micro-founded models, I don’t know.  And whether it’s true in the real world is anyone’s guess!

On models and forecasts, and how one implies the other.

The economic calendar is packed with forecast releases, and each time a new one comes out, social media bursts forth with comments denigrating forecasts.

Common themes are:  forecasting and economic modelling are different activities;  forecasting is trash-talk, modelling or analysing the present is the serious business;  forecasting requires one set of skills, modelling another (viz discussion of whether you need ‘foxes’ or ‘hedgehogs’ for them).

My response to this is to point out that there is not a good separation between modelling and forecasting.  A model implies a forecast.

An example of what prompted me to write this is this piece by John Kay, which, while not sinning, is open to a mis-reading by sinners.

Tom Sargent was fond of repeating every lecture that ‘a model is a probability distribution over a sequence’.  What he meant was that if you have written down a model, it will contain in it a statement about how likely certain things [defined in your ‘sequence’] like output or inflation are to take on different values at different times.  Or, if you project into the past, it will tell you how likely it was that output and inflation in the past ended up being what it was.

This statement was always couched in terms of macroeconomic time series.  But it is more general than that.  The ‘sequence’ could be the set of individuals in a workforce, whose behaviour you haven’t measured yet, but you are trying to predict from what you have measured.

Take the Bank of England’s model as an example.  This model involves a decomposition of all past output and inflation into a set of ‘shocks’.   The model will tell you the contribution productivity shocks were making at any point in time in the past.  But also, and since the model will tell us that shocks take time to have their full effect, we can work out without any extra maths, assumptions, or clairvoyance, what the model tells us about the chance of output and inflation being within certain ranges out into the indefinite future.

The Monetary Policy Committee don’t just do this.  They introduce off model-judgements to bend the forecast to what they think is most likely.  But this doesn’t mean that forecasting is something else apart from modelling.  It means that they are averaging across models;  or modifying the model.

So, when John Kay writes:

“A bane of this economist’s life is the belief that economics is clairvoyance. I should, according to this view, be offering prognostications of what gross domestic product growth will be this year and when the central bank will raise interest rates.”

I’d say:  Fair enough.  But, without any clairvoyance, your understanding of the macroeconomy [your model] means that you are implying something – even if you haven’t stated it explicitly – about what would, other things equal, given what you know now, happen to growth and interest rates in the future, and when.

Or when he writes:

“It is usually easy to move the subject on to something more interesting than macroeconomic forecasting.”

I’d say:  maybe so.  But your conversation partners are failing to parse the logic of economics fully if they think that you really have changed the subject away from forecasting.

John is right to say that economics is not clairvoyance.  But economics – explanations of the economy’s present workings – contains within it statements about the future.

There are of course lots of differences between models [or, equivalently, between forecasts].  Models that are chosen to best fit in sample.  And those chosen to best fit out of sample.  Models with or without explicit Bayesian priors.  These are models where the probability distribution over the sequence – to indulge in the Sargent language – is bent to achieve different criteria.

Following this line of thought, it doesn’t make sense to talk of modelling being interesting but forecasting boring.  Or models requiring one set of skills and forecasts another.  Or forecasts, but not models, being trash.

Although I admit that it would be a bit of a turn off to try to pitch a column about probability distributions over sequences bent to achieve one criterion needing foxes, while those bent to meet another needed hedgehogs.

The Krugtron, confidence and models

Paul Krugman, Brad DeLong, and Larry Summers have been having a debate about just why they think the Fed might have been premature in raising interest rates.  The conversation surfaced a disagreement between Krugman and Summers over the legitimacy of worries about ‘spooking the markets’ with overly large fiscal deficits at the outset of the crisis.

The conversation echoes comments that Krugman, Simon Wren-Lewis, Jonathan Portes and perhaps others made about the narrative account given by the Coalition government – in, for example, a published piece by outgoing Treasury permanent Secretary Nick Macpherson – of the move to reduce the size of the deficit inherited from plans made by Gordon Brown’s Labour administration.

PK insists that the worry about confidence is not wise thinking that is informing models, but unwise thinking undisciplined by models.

I am going to reiterate my position in the Summers-Macpherson camp.

PK’s train of thought is that markets will never fear default by a sovereign that can print its own currency.  The bonds on which it contemplates default are merely claims to that currency.   Printing presses can be run at virtually zero cost.  So what would ever stop them from being run to cover whatever financing gap emerged?

The case in principle that markets might be spooked seems straightforward enough to me.

That case notes firstly that there have been multiple occasions in the past where countries issuing an independent currency have defaulted.   It also notes that inflation is costly, and very high inflation prohibitively so.  There therefore comes a point where monetary finance – particularly expected monetary finance, which is a very inefficient way to pay for government – imposes greater social costs than a partial default.

PK explains that the expected inflation coming from the ‘spooking’ would be stimulative.  He refers to the effect that this has in lowering the real interest rate – the nominal rate being pinned at the zero bound for now.

However, the default risk and the amplified inflation uncertainty that spooking would bring with it [in all but an uninteresting perfect foresight version of this story] would raise the real cost of finance, eating into the stimulus.

These effects are missing from some models – like the simplest, linearised New Keynesian models of monetary policy.  But they are not un-modellable.

How much it should have borne down on fiscal policy then, and should now, is a quantitative question only, and tricky to answer.

This paper, by Corsetti and coauthors, illustrates the trade-off I sketch above and how it produces the possibility of spooking [multiplie rational expectations equilibria, here].  As a point of detail, [I think] it assumes fixed money velocity in order to simplify algebra somewhat, and therefore, in my estimation, would overstate the benefits of monetary finance.  [Real money demand would more normally be thought to shrink as inflation and nominal interest rates eventually rise].

But the idea is clear enough.  And it lives inside a model.

If it were possible to index the unit of account, relatively costlessly – Shiller described how, for example, Chile attempted to do this with the unidad de fomento – and to correspondingly eliminated inflation’s other costs, the Krugman argument would go through.

But in today’s economy, I maintain it doesn’t.  Markets rightly worry about independent currency issuers’ default.

Kocherlakota and the credibility calculus of raising the inflation target

Krugman, Blanchard and others have suggested raising the inflation target as a way to prevent – or at least reduce the severity of – future episodes where the monetary policy interest rate is trapped at zero.

This is a position I have supported as a measure not to help alleviate current difficulties, but as something to consider once the current target is demonstrably met.

On Twitter last night, commenting on Stanley Fischer’s contribution to a panel at the American Economic Association meetings in San Francisco, outgoing FOMC member Kocherlakota expressed his scepticism about the wisdom of raising the inflation target.

Quite rightly, he pointed out that doing this might raise the probability that observers would put on a further inflation target rise sometime in the future.  This would be a bad thing because of the resources consumed as people try to insure against the implied inflation uncertainty, and because it raises the real costs of delivering inflation on target.

However, credibility worriers also need to remember [and here I don’t finger Prof Kocherlakota for failing to] that in some respects raising the inflation target may improve the credibility of monetary policy and reduce inflation uncertainty.

By persisting with the current 2 per cent target, the Fed and other central banks risk further long episodes at the zero bound, and further protracted periods in which inflation is substantially below target [in the UK headline inflation has been about 0 for a year now], and corresponding uncertainty about whether the central bank can ever regain control over inflation.  If setting a higher target means reduced time at the zero bound, then it most surely means better inflation control, and enhanced ‘credibility’, in the sense of the reputation for competence and inflation forecasts that would follow from inflation turning out to be closer to the new, higher objective.

Prof Kocherlakota commented that the Fed’s performance in returning inflation to target during this zero bound episode will inform the expectations that observers form when the next episode ensues.

True enough.  If inflation is returned to target, finally, this ought at least to demonstrate that what was possible this time may be possible in the future, and that reaching the zero bound does not mean automatically becoming like Japan and getting oneself trapped there for 2 decades.

However, this episode will nonetheless underscore the chances of another one of several years, during which it may be forecast that there is a lot of time for further crises to entirely derail monetary policy.

Moreover, expectations will also be informed by the knowledge that the central bank won’t have the advantage of a 5 percentage point cut introductory stimulus [the resting point for rates in the future will be more like 2-3%].  And that there will be less fiscal space, and probably also less political appetite for helping hand from fiscal policy.   And also, should the zero bound be revisited in the next several years, that central bank balance sheets will start out swollen, and so may be expected not to afford the possibility of equally large unconventional monetary policy measures.

 

Carney’s conduct at the BoE

So during the first of the two short slots I did last night on Bloomberg with Joe Weisenthal, it was put to me, via an audio of Paul Donovan at UBS, that Mark Carney was adopting too dictatorial a style.

The way Paul put it was that Carney was treating the BoE like it was the Bank of Canada, when that wasn’t how things were done at the BoE.

I haven’t dealt personally with Mark, but I have heard a lot about him from others, and I did not form the same takeaways as Paul.

My impression from those conversations is that he’s energetic and very forceful.  But how many leaders of large organisations, that have important policy levers, where making a mistake is costly, do not have these characteristics?

I also don’t see any evidence that he has overstepped the mark regarding his roles on the policy or oversight committees of the Bank, where his powers are clearly proscribed (eg by voting).

It’s also worth pointing out that in his role as manager, he’s entitled to manage as he sees fit.  Clearly, he has to give due regard to maximising the chance of taking others with him in changing up the Bank.  But that would not necessarily mean letting those underneath him decide.  And one would anticipate conflict, with those who have thrived under and are more comfortable with the old way of doing things.

That is not to say I am a particular fan of what he has done at the Bank.

I doubt that the new matrix management model, which tilts towards all directorates serving all functions of the Bank, would have made any difference in the last crisis, or have any impact on the next one.  And the confusion and blurring of accountability that such systems can cause may generate as many costs as benefits.  The best that can be said about these changes are that they are changes, without which institutions like the Bank can take on lives and objectives of their own.

And I think forward guidance was a mess.  Confused over whether it was intended to impart stimulus or not.  Appropriating the language of a tool that should have been saved for when it was really needed [ie when they really needed to stimulate].

As I’ve blogged ad nauseam, I also think Carney has made a habit of speaking off topic, which is not advisable for the head of a public body which already has a broad remit.

However, I don’t see any justification for the that Carney is a dictator, clumsily infringing on BoE cultural norms.  I don’t recall his predecessors adhering to this stereotype of softly, softly Britishness.