Comments and Questions

Congratulations to Presbetero for his hard work. Some suggestions of the referee to improve exposition are also useful. In addition I think it is better to avoid too many abbreviations because it is not difficult to expand (with macros) expressions like HIPC for highly indebted poor countries etc. />Although Presbetero follows a standard approache in growth econometrics—i.e., collect data on a number of countries, use one or two refined econometric methods and estimate some cross-country regressions by regressing the growth rate on a few selected variables like the debt ratios etc., there are some serious methodological weaknesses in this and similar approaches. These studies ignore that an economy may take a long time to reach its steady state. Straightforward simulations with the closed form solutions show that this transition period may be 50 or more periods. Therefore, unless growth rates over longer periods of 20 or 30 years are used, ignoring the growth effects of factors of production is a serious misspecification of the basic growth equation. Bosworth and Collins (2003) is an excellent example for cross-country work. They include capital per worker as an explanatory variable although they have used average growth rates over 20 years. The aforesaid specification weakness is serious because growth accounting exercises show that as much as 70% or more growth is due to factor accumulation in the developing countries. What can actually be reasonably estimated with annual or 3 to 5 year averages is only the production function—not the steady state growth equation. Estimates of the parameters of the (modified) production functions can be used to derive the steady state growth equations and test if variables like the debt ratio or aid ratio etc., have any significant growth effects. For policies I am skeptical about the findings based on ad hoc growth equations used in many cross-country studies. I agree with the methodology of Greiner, Semler and Gong (2004) and favour country specific time series models. Some of my methodological views are explained in Rao (2006, 2007). References Bosworth, B. and Collins, S. M. (2003) The empirics of growth: An update, Brookings Institution, September, 2003. Greiner, A., Semler, W. and Gong, G., (2004) The Forces of Economic Growth: A Time Series Perspective, (Princeton, NJ: Princeton University Press). Rao, B. B. (2006 ) Time Series Econometrics of Growth Models: A Guide for Applied Economists, available from http://logec.repec.org/scripts/paperstat.pl?h=repec:pra:mprapa:1547 Rao, B. B. (2007) Estimates of the Steady State Growth Rates for Selected Asian Countries with an Endogenous Growth Framework, available from http://econpapers.repec.org/paper/pramprapa/2389.htm.