Limits to Expenditure Switching? Monetary Policy, Intervention, and Tradability
Submitted by Econbrowser
Currently, net exports are one of the few bright spots in the US economy. As Krugman points out, this is the one area where monetary policy is proving effective: by driving down the value of the dollar, expenditure switching is being induced.
Figure 1: Broad Trade Weighted Value of the Dollar, from January - March 2008. Source: Federal Reserve Board via FRED II, accessed March 17, 2008.So, dollar decline is a good thing, given that several of the other channels of monetary policy have been short-circuited. That being said, there are several constraints to this channel that should be mentioned.
- Exchange rates depend upon the policy rates of multiple central banks, not the least the ECB.
- Foreign exchange intervention is has only limited efficacy in influencing exchange rates over sustained periods, unless coordinated and signals changes in monetary policy.
- Not all US goods and services are highly tradable.
Multiple Central Banks. Consider first the dollar and the euro. In Figure 1, the log trade weighted values of the two currencies are plotted.
Figure 2: Log nominal trade weighted value of the USD against major currencies (blue), and nominal trade weighted value of the EUR against other industrial country currencies (red). Source: Federal Reserve Board via FRED II and IMF International Financial Statistics, accessed March 16.Returning to the first point, it bears repeating that the dollar’s value is a function of US variables relative to rest-of-world variables, whether it be monetary or Taylor rule fundamentals that are relevance.
Inspection of short term policy rates makes explicable — at least in part — recent dollar weakness.
Figure 3: Actual and forecast policy rates. Source: Deutsche Bank, Global Economic Perspectives, March 10.What the forecasts in Figure 2 also highlight is that there are bounds on how far the dollar can fall. As euro area policy rates rise, the dollar will tend to stabilize (although many other factors will continue to push the dollar downward, including possibly the departure of the GCC states from the dollar peg).
Forex Intervention. There has been some discussion of intervention [0]. Here I do not see this as a substantial constraint on dollar depreciation. That’s not because I believe foreign exchange intervention has no effect (see the discussion of the empirical literature on foreign exchange intervention here); rather effective intervention on freely convertible currencies requires the cooperation of major central banks. So we would need to know the motivations for central banks to intervene.
It’s not clear that the appreciation of the euro has quite the impact on Euro Area trade flows that comparable shifts in the dollar has on US trade flows; hence the impact from this channel of an appreciating currency on Euro Area GDP might be less than what one would glean from extrapolating from the US experience. Admittedly, the evidence on what exactly the price and income elasticities for Euro Area trade flows is not particularly developed, in part because of the relative youth of the euro area, and the likely structural changes attendant monetary union.
A quick eyeballing of the data suggests that there is a link between the euro (deflated by either CPI or by unit labor costs), lagged two years, and the Euro Area net exports to GDP ratio.
Figure 4: Euro Area Net Exports to GDP ratio (blue), and log real trade weighted euro exchange rate, CPI deflated (red), and unit labor cost deflated (green), lagged two years. Source: IMF, International Financial Statistics, accessed March 16, 2008. Note: the currency basket for the CPI deflated index is wider than that for the unit labor cost deflatedHowever, the extant econometric estimates suggest a more muted response of trade flows to exchange rates than in the US case [1]. Dieppe and Warmedinger (2007) provide some estimates of extra-Euro Area export and import elasticities. The estimation, couched in terms of error correction models, and incorporating homogeneity of exports to world demand for imports, and homogeneity of imports with respect to domestic (Euro Area) demand, suggests the sum of the absolute value of (relative) price elasticities is about 0.80. But we know that pass through of exchange rates into prices is typically substantially less than unity; hence, a back of the envelope calculation suggests a sensitivity of trade flows to exchange rates much less than that required for the Marshall-Lerner conditions to hold. Similarly, in a occasional paper from 2004, Anderton, di Mauro and Moneta cite relative price elasticities and pass through coefficients that imply (approximate) export elasticities with respect to exchange rates of about 0.25, and import elasticities of about 0.49. These results might explain the equanimity with which some European policymakers greeted the euro’s ascent. So, for now, no obvious floor to the dollar.
(Note, there are other channels whereby which changes in currency values could have an effect on output (see for a discussion of Europe and these linkages Lane and Milesi-Ferretti, 2007 [pdf]). The channel focused on above is on trade flows. Euro appreciation, with the Euro Area in a net creditor position, could a net capital gain or loss, depending upon what triggers the dollar decline. Hence, Euro Area policymakers might have other motives for halting euro appreciation.)
Tradability of US output. Finally, while we typically assume linearity in the relationships, when the changes in the exchange rates are sufficiently large, that might not be a tenable approach. In particular, the share of manufacturing, and the share of plausibly tradable goods, has been declining over time.
Figure 5: Manufacturing value added to GDP ratio (blue) and tradable sector value added to GDP ratio (red). Source: BEA industry statistics and author’s calculations; for details of calculations, see this post.Hence, so far there’s been a substantial response of trade flows to lagged exchange rate depreciation — although not in my particularly surprising given what we know from the econometrics. Whether the magnitude of the response will be smaller due to the shrinkage of the tradable sector over the years of the housing boom, along with adjustment costs associated with moving factors of production across sectors, or whether it will be larger, due to the increased tradability of services induced by the adoption of new information and communications technologies, remains to be seen. So it might be that subsequent dollar declines might not induce the same degree of trade adjustment.
And of course, the largest determinant of export demand remains rest-of-world GDP growth (remember, the contribution of exports to GDP growth in 2007Q4 was about equal to that coming from the decline in imports [2]). Continued export growth relies upon continued rest-of-world growth.
Visit 1800blogger to see all of our industry leading blogs.
Are you an investor? Have something you want to say about the economy? Register on Econoimist Blog now and get published within minutes. Before posting, it is recommended that you review our posting guidelines.
