Priors, Plans, and Other Small Things

Brad DeLong has a long and thoughtful discussion on the confusing process of updating his priors in the past decade. The macronarrative is the tension between his prior and posterior of his confidence. At a 2006 lecture on NAFTA, DeLong starts by noting “Usually whenever I standup and talk in lecture, I am talking about something I’m sure I know what I think, and I’m also arrogant enough that when I’m sure I know what I think, I’m also sure I know I’m right”. This post does not exude that same confidence. But the micronarrative is the tension between his prior and posterior on the state of neoliberalism: have we solved the business cycle, are we at the end of economic history?

It’s a long post, and I won’t comment on it all. It’s definitely worth reading and, if you keep up with his blog, not all the sentiments or slides are brand new. I will respond to Paul Krugman’s pithier post disagreeing (for once) with some of DeLong’s comments.

Brad DeLong has a long meditation on policy that, surprisingly, includes some things I strongly disagree with. I guess I should start by saying that when he describes his fairly complacent view of macroeconomics on the eve of the crisis, he’s describing a view that many economists shared — but I wasn’t one of them. Japan’s troubles and the Asian financial crisis destroyed my faith that we had the business cycle under control; I wrote a whole book about the return of depression economics back in 1999.

Paul Krugman first detailed this argument, a year before his book, in 1998 at BPEA, for which Kenneth Rogoff’s opening response was “this is a truly inspired paper on Japan’s ongoing ‘Great Recession,’ although I have to keep pinching myself to ask if the main thesis can really be true”. I hope Rogoff is no longer pinching himself. (Note, one year later, in March 1999 the Bank of Japan became the first central bank in history to engage in quantitative easing policies).

However, Paul Krugman’s general prior with regard to fiscal dominance at the turn of the 21st century – which is the primary object of DeLong’s introspection – could not have been far from DeLong’s, and other “complacent” neoliberals. Indeed I wonder who, in 2003, wrote:

[L]ast week I switched to a fixed-rate mortgage. It means higher monthly payments, but I’m terrified about what will happen to interest rates once financial markets wake up to the implications of skyrocketing budget deficits…my prediction is that politicians will eventually be tempted to resolve the crisis the way irresponsible governments usually do: by printing money, both to pay current bills and to inflate away debt.And as that temptation becomes obvious, interest rates will soar. It won’t happen right away… But unless we slide into Japanese-style deflation, there are much higher interest rates in our future.

Like any intellectually honest man, he recognizes his flaw and notes:

My thinking has evolved. If you haven’t updated your views in the face of new experiences, you’re not doing your job.

At least insofar as the blight of fiscal dominance, which was the import of the discussion, both Paul Krugman and Brad DeLong have evolved substantially. Krugman indeed was definitely more vocal about the return of depression economics as a phenomenon. However, if I sat with Krugman and DeLong, each, a decade ago and asked “what are the odds America will fall into a liquidity trap”, I expect both would find the chance negligible.

That Krugman believed disinflationary expectations challenged the end of business cycles does not mean he thought America would fall victim. There is an intellectual difference, granted, in accepting it could fall victim and rate the probability as negligible and rating the existence of such a probability itself as negligible. However, I have greater concern with Krugman’s take on confidence “imps”:

First, our expectations argument is a hope; theirs is a plan

[…]

Which brings us to the second point: those of us hoping to summon the expectations imp want to do so with policies that are at worst harmless, such as expanding the monetary base under conditions where this has no direct inflationary impact. The austerians, on the other hand, have pushed directly destructive policies — fiscal contraction in depressed economies — in order to achieve their hoped-for shift in expectations.

So this is the difference between “Let’s try this possibly ineffective remedy, it might work and in any case won’t do any harm” and “Let’s do the opposite of what standard analysis says we should be doing, just trust me”.

On the one hand, I’m bound to agree because I think confidence vigilantes are hardly a problem with the massive demand for safe assets and the American dollar. On the other hand, it misses the ideological Turing Test in its sweeping indictment. We must always consider the priors of those who advance a certain theory. This is like Krugman saying “bond vigilantes are wrong because, well they’re wrong”. DeLong is not unlikely to agree with Krugman on this point, but notes the structural and mechanistic similarity between the confidence fairy and expectations imp (god damned “animal” spirits!). One advancing the confidence fairy argument:

  • May take a Treasury view prior, or believe in monetary offset and hence disregard austerity as bad for growth.
  • May believe inflation is a serious risk.

Neither of these views lend themselves to Krugman or DeLong, but do not divorce the mechanistic similarities between the two spirits. So it’s really not “the difference between “Let’s try this possibly ineffective remedy, it might work and in any case won’t do any harm” and “Let’s do the opposite of what standard analysis says we should be doing, just trust me”.” Because that would be like (logically) begging the question.

The distinction between “hope” and “plan” is even more confusing. And it also really depends if you talk to New or Old Keynesian Krugman (or Krugman the academic vs. Krugman the blogger). He has as much a plan – via a standard IS-LM – as do the austerian types. It’s just a better plan, but that makes it a plan nonetheless. But his word choice is even more confusing. Krugman says that “we have a hope”, “they have a plan”, and yet uses this language to describe their “plan”:

I want the Fed, the Bank of Japan, etc. to target higher inflation, in the hope that it might help, but it’s a hope, and meanwhile we need to fight demands for fiscal austerity and even push for stimulus. The expansionary austerity types, on the other hand, are (or were) actually counting on the supposed rise in confidence to avoid what would otherwise be nasty recessions, which have in fact materialized.

Replace “counting on” with “hope” and tell me if the semantics change? Am I equivocating something? Have I misunderstood? Perhaps. But also, Krugman’s hope is as much predicated on a plan for something else. Furthermore, insofar as we’re only talking about one country, it’s entirely possible to kickstart inflation. As he himself brilliantly put it, the Bank of Japan must “credibly commit to be irresponsible”. That sounds like a plan. A hopeful one, but a plane nonetheless.

To me the similarity between confidence fairies and expectations imps ends at the mechanistic level. But I didn’t even think about it until Brad DeLong pointed it out, so I feel the smarter for it. I wrote this post because after reading Krugman’s post I was quite confident I didn’t feel stupid for feeling smart earlier. So I wrote this post; and maybe I haven’t updated my confidence levels as much as DeLong has, but that’s a sign of ignorance more than anything!

By the way, I think Paul Krugman has been the most astute academic through this crisis. Economic history textbooks will note his prescience and forceful blogging. That doesn’t mean he’s always right, and I think this post is a good example.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s