This research looks at short-run inflation-output dynamics in developing countries, specifically whether demand shocks have nonlinear and asymmetric short-run effects on inflation. This question has implications for how monetary policy should be conducted and is particularly relevant as more discretionary monetary policy regimes take form in many developing countries. The research uses an innovative theoretical and empirical methodology and an extensive data set to provide new insights into this important, but largely unexplored policy-relevant question. In doing so it contributes substantially to knowledge in an important research area.
The research establishes a broad theoretical framework, which integrates specific developing-country features within a generic macroeconomic model, and which nests asymmetric terms within two alterative price adjustment specifications - a monetary model and an extended P* model. A number of empirical techniques are used to identify structural shocks and to test for asymmetries. These techniques include an identification of demand-shock proxies from reduced-form residuals, and from structural residuals using structural vector autoregression methodology. Panel data techniques are used in the estimation. This approach is innovative: the methodology has not been applied in the manner used here.
The innovative methodological approach is complemented by the use of a richer information set than has been used in other studies in this area. The research uses a large data set - a sample of over 80 developing countries, grouped at an aggregate level and divided into regional subgroups, covering a period of over 25 years. The large data set, and pooled estimation techniques, is necessary to uncover potential nonlinearities and asymmetries in the transmission of demand shocks to inflation; individual country data generally contain insufficient information to allow identification of more subtle processes.
The highly aggregated cross-country analysis, and the more disaggregated regional cross-country analysis, is supported by a country-specific case study, examining inflation dynamics in a small Pacific Island country, Fiji. This layered technique provides multidimensional insights into the policy questions addressed.
Overall, the results of the research provide a valuable contribution to knowledge, providing a number of new insights into inflation processes in developing countries, with important practical policy implications. The overall results, while subject to a number of caveats, provide substantial evidence of real and financial sector disequilibria in the transmission of demand shocks to inflation in developing countries and of significant nonlinearities and asymmetries in those processes. The asymmetries are large for most developing country groups: positive demand shocks have a much larger inflationary impact than negative shocks have a disinflationary impact. This pattern is consistent with structural and institutional rigidities and imperfections in developing countries, which see supply constraints emerge well before aggregate capacity constraints are breached. Other rigidities, such as downward nominal wage rigidities appear to impede downward flexibility of prices and wages following negative demand shocks, even where there is substantial excess capacity.
The asymmetry introduces a positive bias into inflationary processes in developing countries and potentially reduces long-term growth rates. If demand shocks are symmetrical, inflation will tend to ratchet up over time unless there is offsetting policy action. The costs of bringing inflation down will be above the short-term output gains that may accrue during periods of excess demand growth. Demand variability will have long-term non-neutral effects.
The evidence of convexity has clear implications for the conduct of monetary policy in those developing countries that are, or are contemplating implementing policy within discretionary policy regimes. For countries with relatively high inflation an asymmetric approach - in which the central bank would tighten aggressively when inflation or expected inflation is likely to be above desired levels and in which policymakers would actively set policy to lock in the gains from positive supply shocks and negative demand shocks when they occur - is likely to lead to better medium-term inflation outcomes.
For countries with lower inflation, the evidence of substantial convexity (considered against a backdrop of institutional and structural conditions in developing countries) suggests that monetary policy should be conducted more aggressively than would otherwise be the case, preemptively neutralising positive aggregate demand shocks and at least partly neutralising adverse aggregate supply shocks. Pre-emptive monetary policy to avoid a rise in inflation will be less costly than reducing inflation once it becomes entrenched. Proactive demand management policies will reduce the output costs associated with maintaining low inflation. The costs of delaying monetary policy adjustments are likely to be relatively high in developing countries.