I am still LOL’ing because Mr. Yates’ post about Market Monetarism and subsequent comment exchange reminds me of a time a colleague was trying to “help” me become more effective at work by telling me, “if you can’t dazzle ‘em with brilliance, baffle ‘em with bullshit.” The colleague, by the way is probably the best BS’er around, taking full advantage of the fact that quite a few people are too timid to say they don’t particularly understand what’s being said.

Really, Mr. Yates appears to be wedded to models and is probably just a little miffed that he spent his life mastering them, only to have them not tell him anything particularly useful about the Great Recession leading him to post a retort like this one:

You’re right that the models might not have anything to say about the real world; this is dealt with in my post. I mention them because i) if you want to mount a competing world view about how monetary policy works, you have to lay out a model and 2) MM protagonists frequently invoke the same models in support of their policies, but, in the instances I have seen, make incorrect statements about them.
‘I’ am actually not assuming anything; these models go back to Sidrauski, Brock and Clower in the late 1960s. ‘They’ make particular assumptions about how the non-pecuniary value of money fluctuates as the quantity of real balances change. Typical is the assumption that real balances are like other goods, that as their quantity increases, non-pecuniary returns [meant to capture the liquidity benefits of money], diminish. Just like the third Mars bar is less satisfying than the second. It is this property alone that generates the zero bound to nominal interest rates, and the inefficacy of monetary policy [here money injections] in the senses described in my post. So, initial assumption is about the functional form of the non-pecuniary returns to holding real balances with respect to their quantity. The conclusions drawn by ‘them’, are about the efficacy or otherwise of money injections. So I / ‘they’ are not assuming A to conclude A. There may be falsehoods, but no circularity. You and Dan are obviously both really interested in monetary economics. It baffles me why you haven’t taken the trouble – I reckon you would only have to give up a day’s work – to work through step by step just ONE of these models to see how they work, what you don’t like about them, what is meant by the statements in my post, and how you think the world would be better described. That way you could offer a serious challenge to mainstream monetary macro.

After posting a smear of Market Monetarism, he gets snarky saying that if you can’t speak his language, you are irrelevant. That is, after he makes a statement that is the epitome of the complaint lodged against him. Let’s take another look at it:

Typical is the assumption that real balances are like other goods, that as their quantity increases, non-pecuniary returns [meant to capture the liquidity benefits of money], diminish. Just like the third Mars bar is less satisfying than the second. It is this property alone that generates the zero bound to nominal interest rates, and the inefficacy of monetary policy [here money injections] in the senses described in my post.

Is this really true in all circumstances? No, I don’t think so. An example might be exploding demand for liquidity could be like a starving man in which the third Mars bar would still not be enough; he will go for a 4th and 5th until he is satisfied. Panics and market crashes cause “the starving” to demand liquidity. But I suppose that isn’t in his model. Wrapped up in all of the formality, he doesn’t seem to notice that the circumstances don’t match the model as he rattles off why the money injections aren’t effective (after saying they are – point 1# in the detailed rebuttal below). If the model cannot tell me anything about why the starving go looking for food and giving it to them satisfies, then I am not interested in his model.

His MM strawman doesn’t really matter, but this is it:

1.  Monetary policy is never ineffective at stabilising inflation or the real economy, even at the zero bound.

2.  Fiscal policy is ineffective [at, see above…] always.

3.  Fiscal policy is effective [at…], but not desirable.

4.  Those New Keynesian models omit to model money, and so don’t capture why monetary policy is effective.

5.  If you look at New Keynesian models carefully, they show that monetary policy is effective, even at the ZLB, which demonstrates why it, and not fiscal policy, should be used for stabilisation purposes always.

6.  Unlike in NK models, monetary policy isn’t just about OMOs, or even buying long dated government securities.  Expansions of the money supply can be used to buy all sorts of assets.

7.  Societies should adopt nominal GDP targeting.

8.  [And/or] It follows from some combination of 1-6 that societies should adopt some form of nominal GDP targeting.

9.  The crisis was caused by inflation targeting.  Following a MM perspective, including a nominal GDP target, would have averted it.

10.  Fiscal policy is ineffective away from the ZLB because it prompts an offsetting monetary policy response.

It’s what is left out that is the problem. Reading through this list it looks as though MM’ers just like to sit back and shoot arrows at New Keynesianism and fiscal policy (we just want to take away that fiscal punchbowl). I shoot arrows at people who adhere to models and the tenants of New Keynesianism even when they don’t match the circumstances, people who write out a thought (as quoted above) that is far out of touch with reality and the facts, and the light bulb doesn’t click on that something isn’t quite right. If one doesn’t understand that an idea doesn’t fit, there is scarce a chance to correct it. I shoot arrows at IT because it caused the financial instability that gets blamed on koo-koo bankers instead of policymakers admitting something is broken and correcting genuine error, and the daft notion that monetary policy is about interest rates. I shoot them because of the seriousness of the current issues involving real people who are suffering. While many of the things discussed among those engaged in theoretical macro are in the abstract, the effects are not in the abstract when put into practice. The public are not white lab rats and no amount of snark from ivory-tower economists will change that. People like Mr. Yates, who can’t tell when he is inconsistent, are in large part why I hang my hat in the MM camp. It’s full of people who are capable of updating priors when they seem to be wrong; and it has a rather nebulous periphery of those who at are at least willing to question what they believe.

None of the points are particularly wrong. And what I meant to say before I went off is that MM isn’t about bashing NK. But what is said in the response to his strawman that is more or less a smear.

1.  Monetary policy is never ineffective at stabilising inflation or the real economy, even at the zero bound.

Well, yes it is.  If the economy is expected to stay at the ZLB forever.  Or if a corresponding money injection is not expected to be permanent, and therefore is associated with interest rate not being expected to be any lower than normally, once the economy has escaped from the zero lower bound.

The economy will not “escape from the ZLB” if the injection is not expected to be permanent (or permanent to the point of outliving the ZLB). It will have zero effect. Here, even with tortured logic, the efficacy of monetary injections that are permanent is not really in question, thus not refuting the point. I’m not even really sure what he’s arguing because money would have to remain tight for the ZLB to bind forever, implying below trend NGDP growth forever. Temporary monetary injections would entirely defeat the purpose of doing it, like digging a hole and then filling it back up.

2.  Fiscal policy is ineffective [at, see above…] always.

This is false, both contrary to the theory and empirics, as stated above.  Versions of the theory which display Ricardian Equivalence  – the ineffectiveness of fiscal policy – are rejected by the data.  And common sense. [I can’t borrow against my future earnings in unlimited quantities, so I am not indifferent to the timing of taxes and spending].

Which theory, which data? If fiscal attempts to reflate the economy and the CB is targeting very low inflation or deflation, it is a cross-purpose, and monetary policy will control. The data from the last two and half decades in Japan tells us something about the ineffectiveness of fiscal policy when the CB is targeting very low inflation or deflation. Similarly, the US should have seen miniscule growth in 2013 with tax hikes and sequestration, but it was our best year out of the last six.

3.  Fiscal policy is effective [at…], but not desirable.

If this is an appeal to practical or institutional problems wielding fiscal instruments, then we are on to interesting territory.  But, I’d say that at the ZLB, they are dominated by the necessity of a stimulus, and the uncertainties surrounding the alternatives.

See my response to point #2 – fiscal policy has a multiplier as large as the CB allows it to have. It seems a bit strange to put together in the same idea being wedded to IT or PLT because of efficient allocation and use of fiscal policy as stimulus where politicians don’t have the same incentives. It would be more philosophically coherent for one of those to bend.

4.  Those New Keynesian models omit to model money, and so don’t capture why monetary policy is effective.

True.  But modifications of them to include roles for QE or credit easing which match the data don’t change the basic story – certainly that fiscal policy is always still effective.  Moreover, an expansion of money to purchase private assets is the sum of a conventional open market operation, which is ineffective in these models still, and a debt-financed purchase of private sector assets, which one might label fiscal policy anyway.

This is funny. It is “true” NK models don’t capture why monetary policy is effective, after he quibbled with the notion it is; and I don’t think the original point mentioned fiscal policy. Yates obviously misses the point that money demand matters (the starving guy with the Mars Bars); and for the point of efficacy, it doesn’t matter which assets are purchased. Credit allocation, from my point of view, is a problem that is beside the point.

5.  If you look at New Keynesian models carefully, they show that monetary policy is effective, even at the ZLB, which demonstrates why it, and not fiscal policy, should be used for stabilisation purposes always.

This seems to be what David Beckworth was saying by tweeting links to Woodford and Auerbach and Obstfeld to me in our exchange.   Well, no.  NK models show what I already explained.  Monetary policy can work if the economy is temporarily stuck at the ZLB, sure.  But so can fiscal policy.  And both are desirable.  And anyway, it’s a bit odd to throw 4 and 5 at us.  We thought you didn’t like the model?!

See my response to point #2 – and no, I don’t like the model – and “the data” is non-existent (who is “we”?).

6. Unlike in NK models, monetary policy isn’t just about OMOs, or even buying long dated government securities.  Expansions of the money supply can be used to buy all sorts of assets.

True.  But we dealt with this.  And it didn’t amount to concluding that fiscal policy wasn’t desirable.  And repeating my smug semantics, we saw that we could even call this fiscal policy if we were so minded.

See my response to point #4 – There’s the “we” again. Are “we” trying to slink off by calling OMOs of other-than-government securities fiscal policy? It must be an act of desperation to resort to splitting hairs.

7.  Societies should adopt nominal GDP targeting;  8.  [And/or] this follows from some of 1-6.

Well, in some model set ups, nominal GDP targeting is the right thing to do, but in many, in fact one might say usually, it is not.  Even in the general case where it is not, Woodford has advocated it as a means to managing expectations of short rates, so that people get the idea that the inflation undershoot has to be followed by an inflation overshoot, for which a reasonable approximation is that people get the idea that a nominal GDP growth undershoot is followed by a nominal GDP growth overshoot [as is the case with a nominal GDP levels target].  So there is something to NGDP targeting.  But it is really only a special case that emerges occasionally.  That’s not to say that what central banks actually do matches what is more generally supported in theory either.  Who knows what they do precisely, for that matter.   However, there is no result screaming out there to justify a major change in frameworks.  [The Bank of Canada used this argument in favour of rejecting Price Level Targeting].  And the most mysterious thing about the interest in nominal GDP is the strange, magical, mythological jump from money to nominal GDP.  Formally, the interest in nominal GDP targeting is, so far, a non-sequitur as regards the MMs worries about NK treatments of money.  The fact that views 7 is grouped with the others undermines them as a ‘movement’.  Where, in the literature, we do find support for nominal GDP targeting, it’s not because of any of 1-6.

I get the sense that he just does not understand NGDPLT. It doesn’t have anything to do with under or overshooting inflation. A NGDPLT regime doesn’t care about inflation for the specific purpose of not having the complexity of worrying about productivity and supply shocks that cause a lot of confusing noise in inflation measures. A CB doesn’t have to switch to NGDPLT to communicate that undershooting inflation would be followed by overshooting. David Beckowrth has pointed to the “result that screams for a framework change;” and if correct, perhaps the case that there isn’t a compelling enough reason should be for public debate and the political system to work out. If IT did indeed cause the financial crisis, then the entire point is particularly glib.

9.  The crisis was caused by inflation targeting.  Following a MM perspective, including a nominal GDP target, would have averted it.

There are BIS-like claims that inflation targeting caused the crisis, through its alleged neglect of financial stability concerns, and asset prices.  These are debatable.  [Answer:  it would have been too costly to avert the crisis with tight interest rates].  But the MM claim I am aware of relates to a different point, to do with the fact that the Fed was insufficiently stimulative, and would have been more so had it appreciated the efficacy of monetary policy even at the ZLB even more than it did, and sought more energetically to generate a boom to make up for the slump.  I think this argument has more to it than the others.  But mainstream New Keynesian macro would say it differently.  Policy might have been better had we already had in place a prescription for a future, post recession inflation overshoot, which would have managed expectations in such a way as to make the initial undershoot less severe.  And, anyway, although some would not agree with me on this, the fact that inflation stayed pretty close to target indicates to me that demand-side policy was doing not far short of what it should do.

The last sentence is the illustration exactly of the shortcomings of IT. Inflation stayed close to target and the world went to the dogs anyway. The reason it went to the dogs is because the demand for liquidity outpaced accommodation, which is a passive tightening of policy when the short nominal rates remain unchanged. Even at the ZLB monetary policy can be too tight if the natural rate is sufficiently negative; and zero is just a number – a cognitive illusion. Money would be too tight if the short nominal rate were 5% and the natural short rate were 1%. And I am not sure, but it seems Yates is arguing with the idea that nominal shocks have real effects. If they do, it’s hard to believe that it wouldn’t impact the demand side. And this is another example of writing out the idea and having nothing click.

10.  Fiscal policy is ineffective away from the ZLB because it prompts an offsetting monetary policy response.

It’s true that away from the ZLB a fiscal expansion would prompt a partially-offsetting monetary contraction, but, this wouldn’t make it ineffective [referring to NK theory and VAR studies of historical policy] or undesirable [referring to theory here].  You can find fiscal instruments [eg the sales tax] that under some settings can be wielded in such a way as to be identical to monetary policy.  But in general, it isn’t, and from this flows the statement that begins this paragraph.

As a retort to each and every one of these points, MMs, or anyone else for that matter, could say ‘I don’t like your models, I’m talking about the real world’.  [This has been the flavour of some of the MM critiques].  Well, I’m not claiming that these models are right.  But they are relevant.

I said I don’t like the models. The reason I don’t like the models is because they don’t illustrate that statements like, “And, anyway, although some would not agree with me on this, the fact that inflation stayed pretty close to target indicates to me that demand-side policy was doing not far short of what it should do” indicate there is a problem with the model. Models may be useful and relevant for something, but applying them to the circumstances and being lead to absurdity means that they are not relevant to the circumstances. Market Monetarism offers an alternate explanation of events that explains most of the facts of the circumstances. The models don’t appear to do that.

Overall, it appears to me, from the drumming on about fiscal policy, that the real complaint about MM is that it doesn’t advocate fiscal policy to remedy economic ills. It’s a veiled ideological barrage that consists of splitting hairs because any logical conclusion that government should insist that monetary policy function effectively rather than spend public resources unnecessarily is heretical blasphemy; hence no reference to data, a blind eye to the mess that is Japan and no explanation as to why the US had great GDP for 2013. Certainly Yates is entitled to his opinion, but he isn’t entitled to his own facts.

[Update] I corrected point references in my rebuttal to the detailed points, and added a reference in the parenthetical for inconsistency on whether monetary injections are effective for clarity.

Advertisements