ASYMMETRIC RISK

All the buzz these days about the dangers of deflation and the strong associated down side risks has set me thinking that this is all really only a modern rehash of what has long been known as Pascal's wager. Named after the 17th century French mathematician, philosopher and co-founder (with Fermat) of probability theory. The wager consists of an argument according to which belief in God is rational whether or not God exists, since falsely believing that God exists leads to no harm whereas falsely believing that God does not exist may lead to eternal damnation. In its modern version this argument runs: it is preferable to risk a few years of mild inflation rather than expose ourselves to the dangers of eternal 'Japanese style' deflation.

Of course as my readers will know, I take the dangers of deflation very seriously indeed, even if I can't help feeling that while the mechanics of how you fall into a liquidity trap type hole are quite well understood, the reasons why the honour fell to Japan, and not someone else are not. And if we don't really understand why they're in the hole, how do we know who may be the next: the US or Germany for example, or no-one.

Meantime this week produced a little quibble between Brad de Long and Paul Krugman over the 'output gap': no-one would want to be seen putting around 'overproduction theories' now would they?

The fuss started with a piece from Paul Krugman - under a title more commonly associated with a pre-recorded voice at a famous London tube station - which contained a certain amount of loose wording:

Mind the Gap: The U.S. economy's "potential output" -- what it could produce at full employment -- has lately been growing at about 3.5 percent per year, thanks to the productivity surge that began in the mid-1990's. But according to the revised figures released a couple of weeks ago, actual growth has fallen short of potential for seven of the last eight quarters. The conventional view is that we had a brief, shallow recession last year, and that recovery has begun. But the output gap tells a different story: Two years ago we went into an economic funk, and it's not over.

In a way the whole double-dip controversy is a red herring; the real question is when G.D.P. will start growing fast enough to narrow the output gap. And so far there's no sign of that happening. There's no mystery about the causes of our funk: the bubble years left us with too much capacity, too much debt and a backlog of business scandal. We shouldn't have expected a quick and easy recovery, and we're not getting one...

Of course one can understand the point and it is pretty sound, but just in case there could be room for doubt, Brad clears it up:

I do, however, have one criticism to make of Paul Krugman's argument. He talks about "excess capacity." There can be--and is--excess capacity in individual industries, like telecom (although this will be quickly taken care of as telecom firms go bankrupt, their assets are taken over by other companies with rational capital structures, and telecom service prices fall through the floor). There's no such thing as "excess capacity" for the U.S. economy as a whole. Further declines in the dollar would spur demand for U.S. exports and for U.S. producers of import-competing goods. Further declines in interest rates would (with proper management of the state of our financial institutions and operating companies) boost demand for investment goods.

The mantra: there is no such thing as (aggregate) excess capacity, there is only insufficient demand.
LINK

Now in case this last mantra seem a little confusing, it might be worth pointing out that the concept of 'output gap' is widely used in the Taylor rule literature over inflation targeting. Logically in any well functioning economy there is going to be a given level of unused capacity, otherwise the thing, to use the nechanics metaphor, would overheat. So in theory there is a certain level of output more or less commensurate with a certain level of inflation, somewhat like that other idealized portmanteau concept of economics the NAIRU, and if output is below that level then the tendency is deflationary. In principle there is nothing wrong with deflation, sometimes it is desirable. The preoccupation comes from the existence of the zero bound, and the danger of overshoot. In any event Krugman takes the trouble to try and sort this out:

A followup on Friday's NYT column: first, Brad DeLong is quite right that there's no such thing as excess capacity for the economy as a whole. I've been vociferous about that myself in the past. What I meant to say was over-investment in short-lived business capital, mainly tech, relative to other resources - which is the sense in which an over-investment model of the business cycle can be justified - the same sense in which business as a whole can find itself with excess inventories.

The experience of disinflation during the 1980s suggested that the economy's "sacrifice ratio" - the number of point-years of output gap needed to bring inflation down by one point - was about 4. That number may well have fallen since, as the economy has become more flexible. Now suppose that the average output gap over the next year is at least as high as it is now, surely a realistic possibility. Then by a year from now we could be looking at zero inflation as measured by the GDP deflator - we could indeed be looking at mild deflation. In short, concerns about future deflation in the U.S. are not wild and crazy: you can quite easily construct a deflation scenario using completely conventional models and not-too-pessimistic numbers.

Is this a bad thing? Zero inflation is not a cliff-edge, with an abyss on the other side. But the lower the inflation rate embedded in peoples' expectations, the higher the real interest rate when the Fed funds rate is zero - that is, a slide from mild inflation into mild deflation makes a liquidity trap that much more likely. Every time I read a news report describing low inflation numbers as good news, I shake my head in disbelief.
LINK

All this is fine, and indeed could be explained by the point that an aircraft entering turbulent weather is going to experience a drag on its velocity. And since we've just had the 'mildest' of recessions, then ironically this may mean that the excesses may take longer to work of. But there is still something which bugs me. Why has deflation really become a danger? In the late seventies if anyone had had a year of negative growth and mildly falling prices we wouldn't have gone overboard on the thing. But today we do, we're apprehensive and we're not sure why? I can come up with two possible lines of inquiry. Firstly a problem from Newtonian mechanics. If an extremely large body enters the vicinity of other smaller bodies, even if moving relatively slowly, it can wreak havoc on the anticipated trajectories of the lesser bodies. For large body read China (and then think India coming right in on its tail). Secondly think population ageing and the rising dependency ratios of most of the economically developed countries. These two factors combined could begin to give us a clue.