Monetary Policy in the Open Economy Revisited: Price Setting and Exchange-Rate Flexibility

Questions & answers

Q1. What are the two price-setting specifications the paper contrasts, and why does the choice between them matter?

Under producer-currency pricing (PCP), a firm sets a single price in its own currency, so that an exchange-rate change passes through one-for-one to the price importers pay; under local-currency pricing (LCP), a firm sets separate prices for domestic and foreign consumers, each fixed in the consumer’s own currency, so unexpected exchange-rate changes have no effect on consumer prices in either country (Sec. 1, pp. 3-4). PCP is “the traditional assumption in open economy macroeconomics,” central to Friedman’s and Feldstein’s classic arguments for exchange-rate flexibility, while the paper argues LCP “is more in accord” with recent empirical evidence on low short-run pass-through of exchange-rate changes to consumer prices (Introduction, pp. 1-2). The choice determines whether exchange-rate movements can perform their traditional expenditure-switching role at all.

Q2. What is the key methodological point the paper makes about exact versus linearized solutions?

Because the model is solved exactly (not log-linearized), monetary policy rules affect the expected LEVEL, not just the variance, of macroeconomic variables like consumption – a channel a simple Taylor-series approximation to utility would miss entirely, since such an approximation “would suggest that monetary rules should minimize consumption variance alone” (Introduction, pp. 2-3). It is specifically the incorporation of expected consumption into the objective that leads policymakers to care about exchange-rate variance and its covariance with productivity shocks, not consumption variance alone.

Q3. Under PCP, what is the optimal monetary rule in response to productivity shocks, and why does it require exchange-rate flexibility?

The optimal rule under PCP (eqs. 3.4-3.5, Sec. 3.a) has the home monetary authority expand money more than proportionately in response to a home productivity shock while the foreign authority contracts, which both raises world demand by the right amount AND depreciates the home nominal exchange rate – replicating the terms-of-trade adjustment that would occur under fully flexible prices (Sec. 3.a, pp. 15-16). “Exchange rate flexibility is a requirement for an optimal monetary policy rule, under PCP, in the presence of productivity shocks” – this is “the essence of Friedman’s argument for floating exchange rates.” The paper also shows that if only one government optimizes, acting alone, while the other follows a passive rule, the first-best cannot generally be achieved (Problem P2 versus P1, Sec. 3.a, pp. 16-17).

Q4. Why does the paper find that even a country with a perfectly stable money supply never wants a completely clean float under PCP?

The PCP monetary authority’s objective (eq. 3.3) is to minimize a linear combination of consumption variance, exchange-rate variance, and the covariance of consumption and the exchange rate with productivity shocks – exchange-rate variance enters as an INDEPENDENT cost, separate from consumption variance, because higher exchange-rate volatility raises expected marginal costs and hence expected prices even when the mean exchange rate is unchanged (Sec. 2, eq. 2.8, pp. 10-11; Sec. 3, eq. 3.7, pp. 17-18). As the home country’s own monetary variance shrinks toward zero, the optimal intervention coefficient gamma approaches a strictly negative limit rather than zero, so “a free float is not, in general, desirable” even then, because “the authorities are not trying to minimize consumption variance only, but are also concerned with exchange rate variance.”

Q5. How does the optimal response to productivity shocks differ under LCP, and what does this imply about Friedman’s argument?

Under LCP, the optimal rules (eqs. 3.10-3.11, Sec. 3.b) have both governments respond identically to a home or foreign productivity shock, and these rules “do not entail exchange rate adjustment” at all – “the optimal monetary rules are consistent with a fixed exchange rate,” because exchange-rate changes do not pass through to consumer prices under LCP and therefore cannot alter the relative demand for home versus foreign goods (Sec. 3.b, pp. 19-20). “The absence of exchange rate movement in the optimal rules in the LCP environment is at variance with the Friedman/Mundell argument for exchange rate flexibility” – even country-specific real shocks do not require exchange-rate adjustment as part of an optimal policy response when LCP holds. The paper notes this comes at a cost: LCP’s optimal rules match the flexible-price variance of consumption but not its mean level, so LCP imposes a first-order welfare cost relative to PCP for productivity shocks specifically.

Q6. What is the optimal degree of exchange-rate intervention in response to monetary shocks, under each pricing regime?

Under PCP, the optimal intervention rule (eq. 3.7) trades off the benefit of letting the exchange rate absorb foreign monetary shocks against the independent welfare cost of exchange-rate volatility itself; even when foreign monetary variance is very large relative to home variance, the rule stops short of a full float (Sec. 3.a, pp. 17-18). Under LCP, the optimal rule (Sec. 3.b, pp. 20-21) simply minimizes consumption variance, with no exchange-rate-volatility penalty, so “if there are no home monetary shocks at all, then the monetary authority should follow a clean float” – a full float becomes optimal when the home monetary sector is sufficiently stable, exactly the traditional presumption for a country with strong anti-inflation credentials, but reached here through the LCP channel rather than the PCP expenditure-switching channel.

Q7. Is PCP or LCP better for welfare overall?

The welfare comparison is ambiguous: with zero exchange-rate intervention and equal monetary-shock variances in both pricing regimes, expected utility is higher under LCP than PCP whenever phi-squared times rho is greater than 1, and higher under PCP otherwise – “either inequality may be satisfied,” so “the welfare comparison across the two pricing specifications is ambiguous: the adjustment benefits of the exchange rate in response to monetary shocks are tempered by the welfare costs of exchange rate volatility” (Sec. 3.b, pp. 21-22). When productivity shocks dominate, however, the paper shows welfare under LCP must be strictly lower than under PCP, since LCP cannot replicate the flexible-price mean level of consumption that PCP’s expenditure-switching channel achieves.

Q8. What does the paper conclude, and what does it flag as a limitation relative to contemporaneous sticky-price literature?

“The optimal response of monetary policy to exchange rates is an old theme… the novel aspect of this paper is that the analysis is undertaken within an expected-utility-maximizing framework in the presence of nominal rigidity… we find that the optimal monetary policy depends in a fundamental way on the type of price stickiness” (Sec. 4, Conclusions, p. 22). The paper explicitly contrasts itself with the contemporaneous Rotemberg-Woodford/McCallum-Nelson research program, in which prices adjust gradually over time rather than fully after one period and investment dynamics are modeled, noting “a challenge for the future is to merge the best elements” of both approaches, and flags endogenizing the currency of price-setting itself (rather than assuming it exogenously) as a further direction, citing the authors’ own follow-up work with Cedric Tille on pass-through and welfare.