Exchange rates and the balance of payments: Reconciling an inconsistency in Post Keynesian theory

Abstract

There already exist Post Keynesian alternatives to neoclassical trade and exchange rate theories. That focusing on the former explains the direction of trade as a function of absolute advantage, which is, in turn, driven by cost and technological differences. There is no automatic force causing these differences to diminish over time meaning that—unlike in Orthodoxy—it is possible for trade imbalances to be large and long-lasting. Meanwhile, Post Keynesian exchange rate theory argues that currency prices are set almost entirely by autonomous financial capital flows. If international investors’ forecast of profit from dollar-denominated assets improves, they buy dollars and the dollar appreciates. On the surface of it, these appear to be satisfactory real world-based approaches that offer much more explanatory power than comparative advantage, purchasing power parity, et cetera. When viewed together, however, an inconsistency emerges—for, a theory that is predicting the trade balance is simultaneously predicting the capital account balance (and the exchange rate at which these transactions are occurring). If Post Keynesian scholarship suggests that a particular nation would have the absolute advantage and should be experiencing a trade surplus, then it must also argue that autonomous capital flows will cooperate and create a corresponding deficit. But Post Keynesian exchange rate theorists see absolutely no reason to expect the latter except by coincidence. Although a tentative solution to this problem has been forwarded, it conflicts with other well-established Post Keynesian tenets. The goal of this article is to resolve the matter once and for all and by a means that leaves all the essential conclusions unchanged. The key lies in explicitly modeling the manner in which trade flows are financed. Once it is acknowledged that the necessary liquidity is endogenously created and that they do not have any direct impact on currency prices, then it is possible to show that trade flows are a function of absolute advantages while exchange rates are driven by international finance.

Keywords:

• Exchange rate
• trade
• Post Keynesian
• endogenous money

JEL CLASSIFICATIONS:

• E12
• F10
• F31
• F32

Notes

1 Or they would purchase spot and forward contracts or currency swaps with mismatched maturity dates, which incorporates interest differentials.

2 Empirical studies often use LIBOR or other interbank rates.

3 One of the most important is that events today must have a bigger quantitative impact on forecasts of the future than they do on today’s spot prices.

4 While considerable work has been done on how currency market participates form their expectations, that will not be detailed here (see Harvey 2006, 2009a, 2009b, and 2012).

5 Harvey 2004 offers empirical evidence for the fact that currency market participants are less willing to commit funds under conditions likely to reduce their levels of confidence.

6 Prasch 1996 reviewed an entire series of problems with the theory of comparative advantage.

7 Though the referenced paper is focusing on exchange rates, their ultimate goal is to explain why these do not automatically adjust to generate balanced trade and hence the latter plays a key role. By regulating producer, they mean the best-practice producer. In other words, the relevant cost for market competition is that generated by the most efficient firm.

8 To be fair, it is clear from the tone of the occasional references to this method of solving the problem, Post Keynesian trade theorists are not nearly as convinced as the Marxists that this represents a defensible explanation.

9 Note that although the two capital markets are distinct, they can still affect one another. For example, trade finance could be securitized and thus affect the speculative capital market.

10 Note that the trade-financing capital flows are drawn parallel to the trade flows on the assumption that changes in the price of foreign currency would have no effect (though it clearly would on the trade flows themselves).

11 The latter actually results specifically because ${\mathrm{D}}_{\$}\left({\mathrm{X}}_{\$}+{\hspace{0.17em}\text{tK}}_{\$}^{\mathrm{i}}+{\hspace{0.17em}\text{sK}}_{\$}^{\mathrm{i}}\right)$ and ${\mathrm{D}}_{\text{fx}}\left({\mathrm{M}}_{\$}+{\hspace{0.17em}\text{tK}}_{\$}^{\mathrm{o}}+{\hspace{0.17em}\text{sK}}_{\$}^{\mathrm{o}}\right)$ shifted the same distance, but for the identical reason: M_$ and ${\text{tK}}_{\$}^{\mathrm{i}}$ shifted by the same amount.

12 Had some portion of the imports not been financed then either D_fx(M_$) would not have shifted as far to the right or those U.S. importers would have, indeed, had to compete for foreign currency, thereby causing the actual exchange rate to appreciate.

13 See Leigh et al. (2015) for an comprehensive and particularly well-designed review of the impact of exchange rates on trade flows.