The Stabilizing Properties of a Nominal GNP Rule

Abstract

In a recent paper Frankel and Chinn (1995) (hereafter FC) examine stabilizing properties of a number of different policy rules. Within context of their model one of their major conclusions is in error. Specifically, conclusion that in an open economy the nominal GNP unambiguously dominates money rule (Frankel and Chinn 1995, p. 326) is incorrect. The reason for mistaken conclusion of FC has to do with an error in their expression for expected loss under a Nominal GNP Rule. It will be shown below that terms involving variance of velocity shocks are inappropriately missing from their expression for expected loss. 1 It should, however, be emphasized that although this result changes specific algebraic expressions for all comparisons involving Nominal GNP Rule versus other rules for open economy, bulk of FC's substantive conclusions remain unchanged.2 In next section FC model is presented and correct expression for expected loss under a Nominal GNP Rule derived. In section 2 expected losses under Money and Nominal GNP Rules and under Nominal GNP and Exchange Rate Rules are compared. Using parameter estimates presented by FC in their model, it is shown that resolution of question as to whether a Money or Nominal GNP Rule is superior is an empirical issue within corrected FC model.3 Also, since a comparison of expected loss under an Exchange Rate Rule versus ex-