Bearing in mind what has happened to me almost every single time since 1997 when I have concluded that Paul Krugman is wrong…
The key, I think, is something hidden in Paul’s column. It is the fact that the effect of pretty much any shock depends on what the private financial market and the public monetary and fiscal policy response to it is:
Still Confused About Brexit Macroeconomics: “OK, I am still finding it hard to understand the near-consensus among my colleagues…:
…about the short- and medium-term effects of Brexit…. [While] Brexit will make Britain somewhat poorer in the long run, it’s not completely obvious why this should lead to a recession in the short run…. So let me give an example of the kind of analysis that I think should raise eyebrows: BlackRock…. “‘Our base case is we will have a recession’, Richard Turnill, chief investment strategist at the world’s largest asset manager, told reporters…. ‘There’s likely to be a significant reduction of investment in the UK,’ he said, adding that Brexit will ensure political and economic uncertainty remains high…
When we say ‘uncertainty’, what do we mean? The best answer I’ve gotten is that for a while, until things have shaken out, firms won’t be sure where the good investment opportunities in Britain are, so there will be an option value to waiting… Brexit might have seriously adverse effects on service exports from the City of London. This would mean that investment in, say, London office buildings would become a bad idea. On the other hand, it would also mean a weaker pound, making investment in industrial properties in the north of England more attractive. But you don’t know how big either effect might be. So both kinds of investment are put on hold, pending clarification.
OK, that’s a coherent story, and it could lead to a recession next year. At some point, however, this situation clarifies. Either we see financial business exiting London, and it becomes clear that a weak pound is here to stay, or the charms of Paris and Frankfurt turn out to be overstated, and London goes back to what it was. Either way, the pent-up investment spending that was put on hold should come back. This doesn’t just mean that the hit to growth is temporary: there should also be a bounce-back…. But that’s not what BlackRock, or almost anyone else, seems to be saying; they’re projecting lower growth as far as the eye can see. They could be right. But I still don’t see the logic. It seems to me that ‘uncertainty’ is being used as a catchall for ‘bad stuff’.
When asset managers–indeed, when anyone anywhere in the world who is not a trained economist–uses the phrase “more uncertainty”, they do not mean what me trained (or mistrained) economists mean: they do not mean that the future distribution of the random variable has a larger variance but the same mean. What they mean, instead, is that the distribution has a larger and longer lower tail. The variance is up and the mean is lower. The principal thing they see as pushing down investment in the near future is the fear of this lower tail–not capitalizing on the option value of waiting until more knowledge comes in.
Back in 1992 Britain exited the ERM. ERMexit had two effects: (1) a small reduction in the desirability of locating in Britain to serve the continental European market because one now faced exchange rate risk, and (2) a large easing of conventional monetary policy and thus lower interest rates and a lower value of the pound because the Bank of England no longer had to maintain the pound at an overvalued parity. The result: boom.
Today Brexit looks to have two effects: (1) a large reduction in the desirability of locating in Britain to serve the continental European market, and (2) ???? (we are not going to get a large easing of conventional monetary policy):
In a proper neoclassical flex-price zero-debt world that was, somehow, at the zero lower bound on nominal interest rates, the response to Brexit would be to bounce the real value of the pound down and to bounce the internal price level down and follow that bounce with higher inflation. The much more strongly negative real interest rate produced by the price level bounce-down-followed-by-inflation would cushion the decline in investment. And the boost to exports from the bounced-down real value of the pound would soak up workers exiting investment-goods industries and maintain full employment.
Of course, the proper neoclassical flex-price zero-debt world is one in which the full operation of Say’s Law is a metaphysical necessity, and so full employment is always attained. We, however, do not live in a proper neoclassical flex-price zero-debt world. It is the job of fiscal and monetary authorities to follow policies that push real prices–real exchange rates, real interest rates, real wage levels–to the values that would obtain in such a world, and so preserve full employment. We can imagine:
*1. Expansion of government purchases: preserve full employment by replacing I with G. Not going to happen in any Britain ruled by anything like this generation of Tories.
2. A helicopter drop: the Bank of England buys bonds for cash, cancels the bonds, and the government cuts taxes by the amount of cancelled bonds. Might happen even with this generation of Tories if they were less thick. But they seem to be very thick indeed.
3. Continued whimpers from Mark Carney that he would not chase away the Inflation-Expectations Imp were she to somehow appear. Not likely to be effective.
4. Everybody becomes so terrified about the safety of their assets in Britain that the real value of the pound bounces low enough that expanding exports soak up all of labor exiting from investment-goods industries.
Paul seem to be betting that (4) is a real live high-probability possibility: the short-term safe real interest rate is pinned at -2%/year for the foreseeable future, but the pound will bounce low enough for expanded exports to preserve full employment. It could happen–the world is a surprising place. But that possibility seems to me to be a tail possibility, not something that should be at the core of one’s forecast.