Wednesday, October 17, 2012

Are recoveries always slow after financial crises and why

Carmen Reinhart and Ken Rogoff have an interesting new Bloomberg column, "Sorry, U.S. recoveries really aren't different." They point to the great Barry Eichengreen and Kevin O'Rourke "Tale of two depressions: what do the new data tell us" columns. (Hat tip, commenter Tim to "slow recoveries after financial crises" who asked what I think. Here's the answer)

Reinhart and Rogoff go after the sequence of studies who have questioned their assertion that recessions after financial crisis are deeper and recoveries slower.
As you can imagine, the argument quickly runs into measurement and sample debates. Reinhart and Rogoff say "Part of the confusion may be attributed to a failure to distinguish systemic financial crises from more minor ones" and "The distinction between a systemic and a borderline event is well established."

Me, I'm still waiting for an economically meaningful definition of "systemic" better than "we'll know it when we see it."  So, the premise here is that there is a clear, visible separation between business cycles associated with (let's all be careful not to jump to "caused by") "systemic" financial crises, "non-systemic" financial crises, and other recessions. I've been doing macro for 30 years, and this clear understanding of the shocks behind any recessions has eluded all the rest of us.

Then, we fight about "how a recovery is measured, and how success is defined." You can feel your eyes getting heavy.

But all this is really beside the point. This latest update does not address or change my basic problem with the whole exercise so far, as it has evolved in the policy debate. (Collecting data is unobjectionable!)

First of all, let us distinguish between recessions following financial crises are "on average" worse and longer or "always" worse and longer, as RR are often misquoted to say -- but they don't do much to clear up the misquote. Recessions following even their definitions of "systemic crisis" vary tremendously in length and depth.

Second, it therefore does not follow that recessions even if  "on average" are worse and longer are "inevitably" worse and longer. Hey, it's not our fault, it's just a law of nature. No. Some recessions are bad and long. Others less so.

Third, then, and most of all, I don't know how we can have this conversation at all without even whispering what the mechanism  might be.

Here's my tentative view: Sure, recessions are worse and longer after financial crises...because governments go completely haywire and screw things up after financial crises. They bail out banks. They hike taxes on "the rich." They transfer wealth. They bail out borrowers. They stomp all over property rights (GM.) Thus, they kill capital markets for a generation. They clamp down on the financial system in horse-left-the-barn efforts to regulate "safety." (We are in this paradox of the 3% mortgage that nobody can qualify for.) They try big "stimulus" plans. They often end up with unsustainable government debts leading to sovereign default or inflation. I'm not making this up. Most of this is in Reinhart and Rogoff's book!  So, perhaps if recessions are longer and deeper after financial crises, not as a matter of economics, but as a matter of particularly bad policy. This is the opposite of inevitability!

You don't have to agree with me, but agree it's logically possible. If so, then the refrain of "recessions are always longer and deeper after financial crises" starts to ring pretty hollow, doesn't it? There is an unwitting implication that the historical average measures some law of economics, that has nothing to do with economic policies. That seems like a pretty big assumption!

Do RR disagree? If so, let's hear loud and clear what they think the economic mechanism is, let's see evidence, and let's see and why policy after such events is only a combination of more or less benevolent responses. Not a word.

The great depression is a wonderful example here. It followed a "systemic" financial crisis by any definition, even mine. (I know it when I see it too!) It was longer or deeper than anything we've seen since as the lovely graph from Eichengreen and O'Rourke on the left here reminds us.

 Now, every single writer  on the great depression thinks it was long and deep primarily because.... wrongheaded government policy made it long and deep. Monetarists point to the Fed, not expanding the money supply enough, reserve requirements, the gold standard, etc. Keynesians think Roosevelt didn't stimulate enough, and only world war II saved us. The wave of new "neoclassical" scholarship points to the disasters of the NRA, cartelization of industries, "war on capital," 70% marginal tax rates,  Smoot-Hawley, financial regulation, unions, etc. Nobody, but nobody thinks the great depression was deep and long because, oh well, that's always the way things are after financial crises.

So why is the 2008 - now recession exempt from the same critique? Why are we following some law of economic nature?  If we don't even talk about mechanism, I don't see how all the averages in the world mean anything.

So what is the message? The closest RR get is to tippy toe around it at the end
This doesn’t mean that policy is irrelevant, of course. On the contrary, at the depth of the recent financial crisis, there was almost certainly a risk of a second Great Depression. However, although it is clear that the challenges in recovering from financial crises are daunting, an early recognition of the likely depth and duration of the problem would certainly have been helpful, particularly in assessing various responses and their attendant risks. 
This is very coy. So, what more than $1.5 trillion deficits (in Keyneisan economics, the whole deficit counts, not just the part labeled "stimulus") and $1.5 trillion monetary expansion do they think the Obama administration and Ben Bernanke should or would they have done, if only they had panicked a little bit more? And are they not acknowledging here that the length and depth of recessions has everything to do with policy, and is not measuring some law of economic nature?

Eichengreen and O'Rourke echo a more explicit sentiment
..policymakers should note that the level of industrial production is still 6% below its previous peak (figure 1). (At the trough it was 13% below its previous peak.) It follows that considerable excess capacity remains in a number of important economies. Exiting now from policies of stimulus in those countries would therefore be premature.
Well, the historical record as they have presented it says absolutely nothing about the efficacy of fiscal stimulus. Neither set of authors claims that, holding the severity of "systemic" financial crises fixed, they have evidence that countries with larger stimulus programs exited more quickly. In fact, RR's compilation of sovereign defaults and inflations after financial crises might suggest exactly the opposite, as a certain regret over stimulus might be settling in in Europe right now.

Without mention of cause, the historical record is just as consistent with my view: a financial crisis can lead to a deep and prolonged recession... so we can't indulge in the usual stimulus/bailout quack medicine that governments follow, which worsens recessions after financial crises.

But Reinhart and Rogoff, with the great data at their fingertips, could be writing about which policies are associated with quicker and slower recoveries. OK, I said "associated,"  and they do say
It is not our intention to closely analyze policy responses that may take years of study to sort out,
If the correlations  come back in ways that don't make sense, will will have a good argument whether  governments that got lucky on recoveries were able to indulge in silly policies.  But since they're basically opining that the Obama administration should have tried stimulus north of 10% of GDP, that cat's out of the bag, and we might as well be debating these facts, not just the averages.

Bottom line, without thinking about mechanisms I don't think we learn anything from these averages. And both sides of the debate are making some big, and often contradictory assumptions. If you conclude "recessions are always long and deep after financial crises" then you're saying policy doesn't really you shouldn't be advocating different policies! If policies matter a lot to the length and severity of recessions, then "recessions are always deep and long after financial crisis" is a meaningless statistic, and a poor fig leaf of an excuse.