The Simple Post-Keynesian Monetary Policy Model: An Open Economy Approach

Abstract

Monetary policy with an inflation targeting rule is analyzed through a simple small-scale Post-Keynesian model that incorporates open economy issues. In contrast with previous Post-Keynesian attempts, the model embodies policy authorities that are committed not only to hitting inflation and/or output targets, but also to the achievement of the external balance. To take account of the external balance objective, we model the real exchange rate as an endogenous and moving target, with the nominal exchange rate being the instrument of that target. The model shows that in response to an adverse external shock the central bank has to consider first the required real exchange rate adjustment that will preserve the external balance, and secondly the level at which the interest rate must be set in order to maintain inflation stabilization. Keeping inflation to target requires higher interest rates and strong reliance on the unemployment channel which, under certain circumstances, also has adverse side effects on income distribution. We show that to deal with an exogenous external shock a policy mix of real exchange rate targeting and income distribution targeting outperforms inflation targeting.

Acknowledgments

I am grateful to the Editor and Roberto Frenkel for suggestions that led to improvements to this paper. I also wish to thank an anonymous referee for valuable comments on earlier drafts.

Notes

¹According to Hammond (2011), at the start of 2011, some 27 central banks could be considered fully-fledged inflation targeters, and many others were in the process of establishing a full inflation-targeting framework. Member countries of the European Monetary Union, Hammond notes, do not belong to the universe of inflation targeters. The set of countries using inflation targeting as their main monetary policy frameworks accounts for about 25 per cent of world GDP.

²Two rare exceptions are Cordero (2008) and Porcile, Gomes Da Silva & Viana (2011) who develop formal monetary policy models in which the nominal interest rate is chosen to hit either an inflation target or a real exchange rate target. However, both attempts concentrate on the comparison of two alternative monetary regimes: inflation targeting versus real exchange rate targeting. Their approach therefore avoids interactions among macroeconomic policy instruments and the possibility of conflicting targets, a pattern that differs from the one we follow here.

³Edwards (2006, p. 2), discussing whether the exchange rate should affect the monetary policy rule in inflation targeting countries, observes that ‘almost every central bank takes exchange rate behavior into account when undertaking monetary policy.’ Aizenman et al. (2011, p. 713) similarly remark that ‘Real exchange rates are likely to play an important role in the formulation of optimal monetary policy in emerging markets’.

⁴From a macroeconomic point view addressing the external balance is equivalent to addressing the competitiveness issue.

⁵This is done through the inclusion of a two-equation dynamical system that describes the wage bargain and price setting process of organized workers and firms in terms of conflicting claims over the distribution of income. The system allows the endogenous and simultaneous determination of the equilibrium rate of inflation and the equilibrium wage share.

⁶Palley (2011) points out that there is a long history of empirical support for the proposition that the coefficient of inflation expectations in Equation 1 is less than unity.

⁷Hence, to preserve the tractability of the model we incur one cost: the omission of the distributive channel. The presence of the terms γ₁ in Equation 3 is meant to capture all factors that, abstracting from the interest rate and exchange rate channels, can influence aggregated demand and employment; but it is clear that it cannot capture the endogeneity of the rate of unemployment with respect to changes in the wage share. One of the consequences this strategy, and in clear contrast with Neo-Kaleckian models, is that it is not possible to designate an open economy as having wage-led or profit-led demand regimes as do Neo-Kaleckian models.

⁸Frenkel (2004) summarizes a number of recent empirical studies of Latin American countries in which real exchange rate depreciations and the rate of employment are positively related. Zeng et al. (2011) find similar results for China.

⁹Beyond the uncontroversial treatment of the foreign interest rate as exogenous, we also assume that capital flows, whatever form they may take, are explained by conditions outside the domestic economy. Indeed, co-movements of capital flows across countries, or ‘contagion’ effects, can hardly be explained by return differentials or expected rates of return. An early assessment of the role of external factor accounting for the observed capital flows in Latin American countries can be found in Calvo et al. (1993). Taylor & Sarno (1997) investigate the determinants of large portfolio flows from the US to Latin American and Asian countries during 1988–1992 and find that ‘global factors are much more important than domestic factors in explaining the dynamics of bond flows’ (Taylor & Sarno, 1997, p. 451).

¹⁰There are good reasons to presume that this is how most floating systems work in practice today. In a fully flexible system, as the experience in mature and developing countries has shown, speculation may be destabilizing in the sense it may cause the exchange rate to diverge from the rate that would assure balance of payments equilibrium. It is this risk of destabilizing speculation that explains why most countries that declared themselves to be independent floaters in the IMF statistics actually intervened in the foreign exchange market, often on a large scale and on a regular basis (Bofinger & Wollmershäuser, 2003).

¹¹Although the NAIRU has receded as a concept for guiding policy and has been increasingly replaced by the notion of inflation targeting, it still underlies the inflation-generating process of mainstream macroeconomic models.

¹²Setterfield (2007) uses a more complex specification in which workers’ bargaining power is also affected by institutional changes in the labor market that create income insecurity or employment insecurity among the working class.

¹³This result validates the findings made by Hein & Stockhammer (2010) in their closed economy model in which inflation-targeting monetary policies—the main stabilization tool proposed by the New Consensus Model—are in the short-run adequate only for certain values of the model parameters.

¹⁴Cultivating more wage or profit share flexibility does not require the destruction of labor unions, or the use of force against the business enterprise in a way that undermines profitability and the investment climate. It does, however, require that the flexibility be based on a credible and stable compromise between labor and capital.

¹⁵There are three ways in which the government might contribute to the management of conflict between conflicting interest groups: by helping to define the bargaining frontier, by providing public information relevant to determining the distribution of the gains from agreement, and by contributing to its enforcement.