رابطه غیر خطی بین تورم و رشد اقتصادی

Using a large panel data set from both developed and developing economies and employing the PSTR and dynamic GMM techniques, this study highlights two aspects of the inflation–growth relationship. First, it analyzes the nonlinearity of the relationship and identifies several thresholds for the global sample and for various income-specific sub-samples. Second, it identifies some country-based macroeconomic features that influence this nonlinearity. Our empirical results substantiate both views and validate the fact that inflation–growth nonlinearity is sensitive to a country's level of financial development, capital accumulation, trade openness and government expenditures. Moreover, these country-specific characteristics result in some marked differences in this nonlinear relationship.

The relationship between inflation and economic growth has long been a debatable element of the discussion between policy makers and researchers. The idea of positive long-run money growth due to nominal rigidities has been so widely accepted that the idea of a negative inflation rate—proposed by Friedman, 1969 and his followers—has never held a place on the policy agenda of any central bank around the world.2 Nevertheless, accelerating inflation has also been unanimously discouraged by all schools of economic thought for its undesirable re-distributional and welfare effects. A few questions remain unanswered. First, are the detrimental effects of inflation on economic growth immune to the level of development of an economy? Second, at what level does inflation start inhibiting long-term growth? And lastly, what country-specific characteristics alter the direction and intensity of inflation's effects on economic activity? Empirical studies conducted in the last two decades have unanimously confirmed the negative and nonlinear impact of inflation on the economic growth beyond some threshold levels, although various inflexion points have been reported in the literature.3 One possible explanation for this lack of consensus is the fact that the inflation–growth relationship depends upon country-specific characteristics. The nature of the relationship and its degree of sensitivity are therefore influenced by differences in the degree of economic development of different countries. This all implies country-specific and time-specific structural breaks in the inflation–growth relationship (Khan and Senhadji, 2001). These authors also advance the view that because inflation could be considered a characteristic of an underdeveloped economy, this structural break is higher for developing economies than for their more advanced counterparts. However, inflation's effects on growth are subject to certain macroeconomic developments that can vary substantially from one country to another. One such feature is the level of financial deepening, which directly affects growth through capital formation. In more financially developed economies, inflation exacerbates the price variability in goods and money markets, further increasing the cost of hedging financial assets among potential trading partners.4 Thus, the adverse effects of inflation on growth are more severe for economies with higher levels of financial development. The same is true of other macroeconomic developments that condition this nonlinearity, such as trade openness, public expenditures and capital accumulation. A higher degree of trade openness also exacerbates the cost of inflation through exchange rate volatility and export competitiveness. Similarly, a higher level of public expenditure results in inescapable inflation through seigniorage tax and cost overruns on public projects, reinforcing the adverse effects of inflation on growth. Inflation also reduces capital accumulation by lowering the real interest rate and saving rates. All of these country-specific characteristics condition inflation–growth nonlinearity. Nevertheless, these conditional variables have been overlooked in the literature. Another reason for the imprecise threshold determination is that, in the majority of the previous studies, the threshold has first been assumed exogenously and then tested for empirical significance. This drawback of the external threshold determination has been resolved by Omay and Kan (2010) and López-Villavicencio and Mignon (2011), using a panel smooth transition regression (PSTR) model for 6 and 44 countries, respectively. Nevertheless, their studies focused only on limited sets of countries and ignored the country-specific characteristics discussed above. Our paper fills these gaps by enlarging the dataset and analyzing how the degrees of trade openness, financial development, capital accumulation and government expenditures can influence the nonlinearity of the inflation–growth relationship. Most of the previous studies have also been based on the assumption that the inflation–growth relationship can only be affected by cross-country variations in the level of inflation.5 Consequently, these studies neglected the changes that occur in inflation and economic environments over time. To overcome these deficiencies, we use a PSTR model that authorizes a smooth transition for a weak number of thresholds over a continuum of regimes. This approach has two main advantages. First, PSTR model enables us to transcend variation among countries and over time. This provides a simple way to appraise the heterogeneity of the inflation–growth relationship by country and over time. Second, this approach allows a smooth change in the country-specific correlation, depending on the threshold variables. Our final contribution is to analyze the role of income level in determining the nonlinearity of the inflation–growth relationship by splitting the data into sub-samples of countries based on their per capita GDP. Our main findings support the findings of previous studies concerning inflation–growth nonlinearity and propose different thresholds for rich and emerging countries. We also tested the hypothesized role of macroeconomic developments for our sub-samples. The rest of the paper is organized as follows. Section 2 summarizes some important previous research and provides a brief discussion of the role of country-specific characteristics on the inflation–growth relationship. Section 3 presents our PSTR and GMM model settings and the specifications tested. 4 and 5 present the data and the empirical findings, respectively. Lastly, Section 6 offers conclusions.

This study provides new evidence of the nonlinearity of the relationship between inflation and growth by applying the PSTR and dynamic GMM models to a broad data set for 102 developed and developing countries. The recent literature has confirmed that the relationship between inflation and growth is nonlinear and that there therefore exists a certain threshold above which inflation is harmful and below which it enhances growth. However, a precise estimation of the threshold level and the macroeconomic environment that influences this nonlinearity has not been attempted in previous research. We have mainly addressed two aspects of this relationship: the threshold estimates for the whole sample, as well as for different sub-samples, and some country-specific characteristics that can possibly affect the degree of sensitivity between inflation and growth. Our first-stage findings confirmed those in the literature that hold that the inflation–growth relationship is nonlinear, and our threshold estimates decrease with the level of income. This systematic inverse relationship illustrates the fact that inflation indexation and inflation tolerance are higher in developing economies than in developed economies. Our threshold estimates are consistent with the recent work of López-Villavicencio and Mignon (2011) for a relatively small data set of 44 countries. With respect to the other country-specific macroeconomic variables that affect the degree of sensitivity in the inflation–growth relationship, our results validate the usefulness of these channels in explaining the differences in the intensity of the relationship between inflation and growth. Indeed, the degrees of financial development, trade openness, capital accumulation and government size were found to be the main factors responsible for altering the nonlinearity of the inflation–growth relationship over time and across countries. Our evaluation of these factors also highlights the fact that the issues of the welfare cost of inflation and the optimal inflation rate cannot be settled in a vacuum. Effectively, the specific features of the macroeconomic environment related to a certain country determine both its optimal level of inflation and the welfare cost of inflation, once it crosses this threshold. For example, the optimal inflation in an open economy can be very different from that in a closed economy, even if both countries have the same level of GDP. Lastly, consistently with López-Villavicencio and Mignon (2011), our direct analysis of the inflation–growth relationship indicates a very high optimal inflation level for low-income countries. However, contemplating other macroeconomic conditions makes it clear that adverse inflation effects can appear through other channels well below this level. Thus, our results suggest that the issue of inflation optimality requires a broader and deeper analysis.