Larry Summers presents as an example of his contention that we know more than is in our models--that our models are more a filing system, and more a way of efficiently conveying part of what we know, than they are an idea-generating mechanism--Paul Krugman's Mundell-Fleming lecture, and its contention that floating exchange-rate countries that can borrow in their own currency should not fear capital flight in a liquidity trap. He points to Olivier Blanchard et al.'s empirical finding that capital outflows do indeed appear to be not expansionary but contractionary:
Macro Effects of Capital Inflows: Capital Type Matters: "Some scholars view capital inflows as contractionary...:
...but many policymakers view them as expansionary. Evidence supports the policymakers.... Bond inflows lead to currency appreciation and are contractionary, while non-bond inflows lead to an appreciation but also to a decrease in the cost of borrowing, and thus may be expansionary.... How can we reconcile the models and reality? ... Capital inflows may... reduce the cost of financial intermediation... be expansionary even for a given policy rate. In emerging markets, with a relatively underdeveloped financial system, the effect of a reduction in the cost of financial intermediation may dominate, leading to a credit boom and an output increase despite the appreciation....
The appropriate policies vis-à-vis capital inflows depend very much on the nature of the inflows. Sterilised foreign exchange (FX) market intervention... used in response to non-bond inflows... increases capital inflows, and thus increases the effects of inflows on credit and the financial system.... If the central bank is worried about both appreciation and unhealthy or excessive credit growth, FX intervention or capital controls are preferable to the use of the policy rate in response to an increase in bond inflows.... In response to non-bond inflows, our framework suggests that if the goal is to maintain exchange rate stability with minimum impact on the return to non-bonds, capital controls do the job best, followed by FX intervention, followed by a move in the policy rate...
If you asked me for a precis of what is going on in Blanchard et al., I would start drawing a graph in which we had:
- An IS ("Investment-Savings") curve, for which the level of production (a) increases when government purchases increase; (b) falls when the long-term risky real interest rate rises, discouraging investment and consumption directly and discouraging exports indirectly by raising the value of the currency; and (c) falls when desired capital inflows rise and thus raise the value of the currency.
- An LM ("Liquidity-Money) curve, relating demand for money and thus the short-term safe nominal interest rate as a function of the level of output given the money stock.
- A CC ("Credit-Channel) wedge between the two, consisting of (a) the term premium on interest rates, that is how much long-term rates exceed short-term rates; (b) the risk premium on investments; and (c) the expected inflation rate:
I would point out that, in Blanchard et al.'s setup, what they call "bond inflows" move the IS curve to the left by raising the value of the currency for any given short-term safe nominal policy interest rate. Thus they are contractionary--for a constant policy interest rate on the LM curve, and a constant double-arrow CC credit-channel wedge.
I would point out that what they call "nonbond inflows" both move the IS curve to the left and shrink the double-arrow CC credit-channel wedge. So--for a constant policy rate--are contractionary if the first and expansionary if the second effect dominates.
And I would point out that Krugman's Mundell-Fleming lecture deals with this case under the heading of "banking crisis":
Banking crisis: Several commentators – for example, Rogoff (2013) -- have suggested that a sudden stop of capital inflows provoked by concerns over sovereign debt would inevitably lead to a banking crisis, and that this crisis would dominate any positive effects from currency depreciation. If correct, this would certainly undermine the optimism I have expressed about how such a scenario would play out. The question we need to ask here is why, exactly, we should believe that a sudden stop leads to a banking crisis. The argument seems to be that banks would take large losses on their holdings of government bonds. But why, exactly? A country that borrows in its own currency can’t be forced into default, and we’ve just seen that it can’t even be forced to raise interest rates. So there is no reason the domestic-currency value of the country’s bonds should plunge. The foreign-currency value of those bonds may indeed fall sharply thanks to currency depreciation, but this is only a problem for the banks if they have large liabilities denominated in foreign currency...
Krugman is working in a framework in which the risk-premium part of the credit channel--the risk-premium part of the double-headed orange CC arrow--is small in normal times but can discontinuously shift to large in the event of a "banking crisis". For Blanchard et al. and Summers, the size of the risk-premium part of the CC arrow is not discontinuously 0-1, but rather moves gradually with "confidence": low when confidence is high and desired capital inflows are large, high when confidence is low and there is a sudden stop. Krugman says that since, in the absence of large foreign currency-denominated debt, there is no reason for there to be fears of a banking crisis in the event of a sudden stop.
Blanchard et al. appear to think that Krugman is largely right about developed economies. In them risk premiums are, they think, relatively small: financial markets work relatively well, and are not all that sensitive to amounts of new investment money flowing in. But, they say, they believe that in developing economies things are different. There, they believe, the risk premium component of the credit channel wedge is sensitive to international market conditions and thus the size the desired capital inflow.
In brief, Blanchard et al appear to think that only developing economies are "Minskyite" in the sense of a strong vulnerability of the CC risk premium to "confidence" even in the absence of fundamental shocks. Summers's judgment appears to be that all economies are "Minskyite". If I understand Paul's thinking, it is more that "confidence" is only likely to matter as an equilibrium-selection device in a model with multiple equilibria like that of Krugman's (1999b) third-generation financial-crisis model in "Balance Sheets, the Transfer Problem, and Financial Crises". No multiple equilibrium, no possibility of suddenly jumping to a bad equilibrium, and so little possibility of any sort of pure "confidence" shock causing large amounts of trouble.
Paul here is more on the "financial markets work like they ought to" side of the argument. Olivier and company are more on the side of "developing economies have financial-market vulnerabilities North Atlantic economies do not" side. And Summers is more on the side of Keynes here:
Enterprise only pretends to itself to be mainly actuated by the statements in its own prospectus.... Only a little more than an expedition to the South Pole, is it based on an exact calculation of benefits to come. Thus if the animal spirits are dimmed and the spontaneous optimism falters, leaving us to depend on nothing but a mathematical expectation, enterprise will fade and die.... This means, unfortunately, not only that slumps and depressions are exaggerated in degree, but that economic prosperity is excessively dependent on a political and social atmosphere which is congenial to the average business man. If the fear of a Labour Government or a New Deal depresses enterprise, this need not be the result either of a reasonable calculation or of a plot with political intent;--it is the mere consequence of upsetting the delicate balance of spontaneous optimism. In estimating the prospects of investment, we must have regard, therefore, to the nerves and hysteria and even the digestions and reactions to the weather of those upon whose spontaneous activity it largely depends...