Truth and Robustness in Cross-country Growth Regressions

Abstract

We thank Oscar Jorda, Judith Giles, Wayne Joerding, and participants in seminars at the University of California, Irvine, and the University of Victoria for helpful comments on an earlier draft. We also thank Orley Ashenfelter for his help in getting this project off the ground. AbstractThe work of Levine and Renelt (1992) and Sala-i-Martin (1997a, b) which attempted to test the robustness of various determinants of growth rates of per capita GDP among countries using two variants of Edward Leamer's extreme-bounds analysis is reexamined. In a realistic Monte Carlo experiment in which the universe of potential determinants is drawn from those in Levine and Renelt's study, both versions of the extreme-bounds analysis are evaluated for their ability to recover the true specification. Levine and Renelt's method is shown to have low size and extremely low power: nothing is robust; while Sala-i-Martin's method is shown to have high size and high power: it is undiscriminating. Both methods are compared to a cross-sectional version of the generalto-specific search methodology associated with the LSE approach to econometrics. It is shown to have size near nominal size and high power. Sala-i-Martin's method and the general-to-specific method are then applied to the actual data from the original two studies. The results are consistent with the Monte Carlo results and are suggestive that the factors that most affect differences of growth rates are ones that are beyond the control of policymakers. JEL Classification: C4, C8, O4 Truth and Robustness in Cross-country Growth RegressionsNo. Country

Searching for the Causal Structure of a Vector Autoregression*

“…What is more, rates of omission rise somewhat as nominal text size falls, confirming the usual tradeoff between size and power. These results are consistent with findings of well‐behaved size in non‐causal search algorithms (see Hendry and Krolzig, 1999; Hoover and Perez, 1999, 2002; Krolzig and Hendry, 2001). For a general defense of well‐regulated search as a respectable econometric practice, see Hoover (1995) and Hoover and Perez (2000).…”

How Does Foreign Direct Investment Really Affect Developing Countries' Growth?

“…To determine which of these variables are important for explaining the cross‐country variations in the effects of FDI on economic growth, this paper uses the general‐to‐specific model‐selection approach suggested by Hoover and Perez (2004). The general‐to‐specific approach is adopted here because a comprehensive theory to explain the cross‐country variations in the growth effects of FDI does not exist.…”

“…Following the Hoover and Perez (2004) approach, this paper starts by estimating a general specification, in which all variables are included, and subject the estimated model to a series of specification tests. The test battery includes a Jarque–Bera test ( JB ) for normality of the residuals, a Ramsey RESET test for general nonlinearity and functional form misspecification ( RESET ), a Breusch–Pagan–Godfrey test for heteroscedasticity ( HET ), and a sub‐sample stability test ( STABILITY ) using an F ‐test for the equality of the variances of the residuals of the first half versus the rest of the sample.…”

The Principal Components of Growth in the Less Developed Countries

“…Some of these interrelationships could also be indicative of complex interaction effects, as the results below demonstrate below. And finally, this EFA‐based approach is a sensible alternative to the Leamer‐type (1985) tests used by Levine and Renelt (1992) and Sala‐i‐Martin (Sala‐i‐Martin, 1997, Sala‐i‐Martin et al, 2004) and general‐to‐specific approaches of Hoover and Perez (2004) and Hendry and Krolzig (2004), both of which have their problems 14…”