New Insights with Opposing Results
The new open economy macroeconomics framework led to several new insights on international welfare spillovers of monetary and fiscal policy. Yet, with its micro-foundation structure there are many more modelling choices to make which affect the results.
In particular two issues are controversially debated: the source of the nominal rigidity and the pricing behaviour of exporters. Nominal rigidities might not necessarily stem from pre-set prices. They might also stem from sticky wages which are negotiated in advance (Obstfeld and Rogoff 2000; Corsetti and Pesenti 2001). Unfortunately, empirically there is no unambiguous answer where the rigidities stem from. Both distortions in wage setting and in price setting are possible. The same applies to the second issue: price setting by exporters is neither only conducted with respect to the home market (fixed mark-up in home currency) nor only with respect to the foreign market (fixed mark-up in foreign currency). Thus, both assumptions can be justified. Unfortunately, the choice of the assumption has non-trivial consequences for the resulting policy recommendation. This is particularly nasty, since the new open economy macroeconomics aimed at providing a tractable tool for policy analysis.If prices are pre-set and firms price in local currency as in Devereux and Engel (2001), nominal exchange rate changes have small or no short-run effects on international trade. This contrasts sharply with the traditional Keynesian view, with much of the policy advices, but also with the result from Obstfeld and Rogoff (1995) presented above. With local currency pricing the exchange rate pass-through, that is, the responsiveness of export prices to exchange rate changes, is zero instead of one as in the models assuming producer country’s currency pricing. The consequence is that there is no expenditure switching effect in local currency pricing models.
Without expenditure switching, the exchange rate moves stronger in the short run than in models relying on producer’s currency pricing. Over-shooting is therefore possible in local currency pricing models.An unanticipated monetary expansion in the home country improves home country’s terms of trade in the short run, since home firm’s export prices in home’s currency increase with the nominal depreciation while the import prices at home are unaffected owing to local currency pricing. This is the opposite result to that in Obstfeld and Rogoff (1995). The consequence is that “beggar-my-neighbour” policy results from these models; not because of an expenditure switching effect but because the home country’s monetary expansion reduces the terms-of-trade of the foreign country (Betts and Devereux 2000). A monetary expansion in such a setting induces a positive real wealth effect for individuals in the home country, which results from the improved terms of trade and increases output and consumption in the short run.
An interesting consequence of fixed prices in local currency is the deviation of the interest rate in both countries after an unanticipated monetary expansion in the home country. This deviation occurs despite a perfectly integrated asset market. Since the interest rate is directly linked to consumption growth, the interest rate at home falls relative to the initial rate and relative to the foreign interest rate. The reason is the diverging consumption pattern. While consumption grows proportionally to money in the initial period in the home country, foreign consumption is unaffected. Since future consumption at home is unchanged with pre-set prices for only one period, the interest rate at home must fall in the shock period because with rising prices in the future, shock period’s savings are worth less in terms of consumption units. Foreign consumption is unchanged in all periods.
With no consumption smoothing in either country, there are no current account dynamics.
The home country produces and consumes more in the initial period but does not spread the higher income to the following periods. The consumption path in the foreign country is completely unchanged. The monetary expansion does not create a trade imbalance with local currency pricing. The full effect of the expansionary policy occurs in the initial period. After this, money is neutral which stands in sharp contrast to the results of Obstfeld and Rogoff (1995).It is the reaction of the current account following a major change in economic policy which leads Obstfeld and Rogoff to herald producer country’s currency pricing (Obstfeld and Rogoff 2000). Although they admit that the pass-though is not one or not even close to one, they argue that improvements in the current account usually follow depreciations. Depreciations are empirically related to deteriorations and not to improvements of the term of trade (as the local currency pricing would predict). Yet, producers’ currency pricing implies that the law of one price holds which is also not supported by the data (Froot et al. 1995). Both modelling strategies can however be tested in a nested empirical model which might be the appropriate jury in this debate.