Mundell Fleming Muddle ?

Robert Waldmann

Wow an argument in favor of protection from Paul Krugman. I never expected to read that. Now he’s against protection all the same, but he does admit that there is a good argument that right now a bit of protection would be good for the world.

The argument, basically, is that governments aren’t coordinating macro policy and that, if stimuli spill out, then they won’t stimulate enough so everyone will be better off if every country keeps its stimulus to itself. He concludes however, that protection would be a bad policy, because once countries start they don’t stop so a little bit of protection right now is not an option. Anyway it’s brief so read it.

Nick Rowe has an interesting counter argument. He argues that if all countries are in a liquidity trap, then fiscal stimuli don’t spill over.

According to Krugman “His argument is based on the proposition that since interest rates are fixed under a liquidity trap, capital flows are fixed, and the exchange rate will adjust to offset any change in the trade balance.” I will comment on this alleged argument. Rowe’s post refers to a paper which I haven’t read. Actually the post seems less sophisticated than Krugman’s presentation, as Rowe seems to argue that flexible exchange rates always imply balanced trade. He can’t be arguing that, and I haven’t read the paper.

The argument that this is true specifically if we are in a liquidity trap seems to be specific to Rowe as presented by Krugman.

Rowe’s argument as presented by Krugman seems backward to me (from now on “he” means Rowe as presented by Krugman). Basically he seems to be arguing that demand for say 3 month t-bills is very elastic and so the price of 3 month t-bills won’t change and so the amount of 3 month t-bills in investors’ portfolios won’t change — that is he is going from the assumption people are willing to shift huge amounts of wealth into 3 month t-bills to the conclusion that people won’t shift any wealth into t-bills.

I repeat this argument in various ways after the jump.

The evidence that we are in a liquidity trap is simply that short safe rates are approximately zero. However, the application in macro models is the assumption that, therefore, safe short term interest rates won’t increase if more short term public bonds are issued. That seems to me to be the same as the assumption that people and firms are willing to buy much more of them.

For some reason, Rowe assumes that this applies only to domestic people and firms and that whatever makes investors so eager to buy short term public debt is confined nation by nation. Not to put to fine a point on it, this strikes me as absolutely absurd given, for example the fact that UBS is endangered by the US real estate bubble and that AIG was brought down by its London office.

The argument is interest rates are fixed because a miniscule change in interest rates would cause a huge flow therefore changes in interest rates will be miniscule therefore flows will be miniscule.

I’d say this is a case of the dear old Mundell Fleming model overcoming common sense as if it were an RBC model.

The same goes for the anchoring model.

s = sbar + (i*-i)/k

s is the exchange rate, sbar a steady state exchange rate determined long after the recession and stimulus are over, i domestic interest, i* foreign interest rate and k a constant.

Krugman assumes that k is constant which, for one thing, implies that capital flows are linear in interest rates. Does that assumption seem reasonable right now ? seems to me that an interest rate fluctuating between 0 and 10 basis points is a strong time that k is a bit low right now.

However I guess k doesn’t matter to Krugman. I guess that Krugman’s argument amounts to saying that with constant interest rates, bonds become perfect substitutes so the amounts of different bonds that people will hold (tried to avoid writing “stock of bonds” there) can change a lot and balance large trade deficits at a fixed relative price. That is Krugman is ahead of me as usual always so far.