Stylized Facts for the Argentine Economy

One of the most important objectives of the macroeconomic science is the comprehension of causes and consequences over the economic fluctuations. The statistical characterization of business cycles has been studied by many authors since the influential work by Hodrick and Prescott (1980). However, some problems can appear if we try to analyze the facts in this way. The first problem emerges as soon as we look at the time series. In the study of business cycles it is difficult to establish regularities. For instance, although Argentina was affected by crises more than several developed economies, the unemployment rate in Argentina has not shown large fluctuations. This is true, even though the series were incorrect. The second problem is to obtain reliable data. Maybe, unemployment in Argentina had been underestimated. Since the development of the Real Business Cycle theory, most of the modern macroeconomics has been doing large efforts to understand fluctuations of the macro variables. Criticizing Burns and Mitchell (1946) because of the absence of theoretical support, Lucas (1977) defined cycle movements as income deviations around its tendency and relative change in the macro variables behaviour respect to the income. Looking for behaviour patterns, Hodrick and Prescott (1980) found that some stylized facts announced by Burns and Mitchell (1946) were kept few years later. Actually, the results obtained by Burns and Mitchell (1946) and Hodrick and Prescott (1980) reinforced Lucas´ definition. However, other studies including more variables offered evidence about not keeping regularities. The latter depends on economic structural characteristics and the time period. Kydland and Prescott wrote another work in 1982. They pointed out the possibility that the same analytical tools used to study economic growth could be used to study business cycles too. In this way, the variables which explain business cycles are real factors, rather than nominal ones as in Lucas (1977). In 1990, Kydland and Prescott researched the stylized facts for the American economy. Then, there were many comparative and particular works. Backus and Kehoe (1992) have done a comparative study in which they compared a set of developed countries. Agénor, McDermott and Prasad (1999) have worked with medium-income countries and Loayza, Fajnzylber and Calderón (2004) have studied a set of Latin American countries. The present paper takes as particular references Mendoza (1991), Kydland and Zarazaga (1997), Sáez (2004) and Castillo Montoro and Tuesta (2006), since emergent economies have singular characteristics and all those papers deal with developing countries. Such characteristics could appear in those economies due to two reasons (Agénor et al., 1999). On one hand, the researcher must face the scarcity and low quality of available data. National account systems in developing countries likely have measurement errors, discontinuity on data and deficient estimations. On the other hand, these economies suffer sudden shocks because of their high volatility. Kydland and Zarazaga (1997) have wondered whether the business cycle in Argentina could be explained using the same arguments that explain the business cycle in developed economies. The authors have covered the 1970-1995 period, using two different estimations over product and its components. Even if their results may be considered ambiguous and crucial the period, Kydland and Zarazaga (1997) make up a reference point to this paper, and then our results will be compared to theirs. According to the literature, developing countries are characterized by the large volatility of macroeconomic variables and the major spreading degree of foreign shocks. Perhaps we can still find some regularity from the Argentine economy. In this work, we will try to introduce the stylized facts analysis on Argentina through the cross-correlation study in the 1990-2007 period. We used the Hodrick and Prescott filter, with and without applying the pre-whitening procedure. Intuitively, the results obtained in this paper seem to be correct although some of them do not have correspondence with business cycle theory and point out some differences with reference to Kydland and Zarazaga (1997) report on the Argentine economy. During consecutive crisis, Argentina has exhibited a specific behaviour which has been different from those patterns exhibited by developed countries. Nevertheless, the main variables have kept a similar behavior.