Economic Management and Social Policy 183 it,I\OAL,6 Ai84. May 2001 Findings reports on ongoing operational, economic, and sector work carried out by the World Bank and its member govemments in the Africa Region. It is published periodically by the Knowledge and Leaming Center on behalf oF the Region. The views expressed in Findings are WOR L D BA N K those of the author/s and should not be attributed to the World Bank Group. Revisiting Growth 22681 8oN AN9D and Convergence: Is Africa Catching Up? he neoclassical Solow frame- ogenous technical change. Endog- T work has been the workhorse enous growth models have provided for empirical analysis of mechanisms through which eco- growth in industrial and develop- nomic and social policies can af- ing countries. In this framework, fect long-run growth through their steady state economic growth (de- effects on human and physical fined as the state in which output capital accumulation. Recent grows at a constant rate) depends cross-country empirical work on on exogenous technological growth has been inspired by these progress and population growth. In extensions of the neoclassical particular, without technological model extended that include gov- progress, output per capita does emient policies, human capital. not grow. and technology diffusion. An important feature of the neo- classical model that has been the Objectives central focus of empirical work is 0i _ the convergence property: output This study examines rates of con- levels of countries with similar vergence for African and OECD technologies converge to a given countries and investigates the de- level in the steady state. In the end, terminants of per capita growth V ceteris paribus, the lagging poor rates drawing on the neoclassical countries will tend to catch up with and endogenous growth theories, the rich. Using cross- sectional and focusing on issues of economic analysis the majority of the litera- estimation. Specifically, the study ture seems to have reached a con- addresses the following: sensus on the issue of convergence: the poor do catch up with the rich, * Are countries converging towards at a rate of 2-3 percent per year. their steady states, and, if so. The obvious shortcoming of the how fast are they getting there? U neoclassical model is that long-run * Are differences between coun- - * per capita growth is determined by tries' per capita growth rates ex- the exogenous rate of technologi- plained by how far away the cal progress. Work on endogenous countries are from their steady A ] growth theory has introduced al- states or by the determinants of ternative models that explain long- their steady states? w - run growth, and provide a theory * Is the neoclassical growth frame- of technological progress: growth is work adequate to describe cross-- generated by factors other than ex- country differences? * Human capital accumulation en- divisions influence economic form, is not consistent with the hances economic growth because growth, the rationale being that empirical evidence and therefore it is a direct input to research or polarized societies have more cannot account for the important because of positive externalities. difficulties agreeing on the pro- features of cross-country income In light of this, if a measurement vision of public goods such as differences. of human capital accumulation is infrastructure, education, and . Various economic factors such added to the analysis, is the "aug- growth-enhancing policies. as initial conditions, investment, mented" neoclassical framework population growth, human capital adequate to describe cross-coun- Conclusions development, government con- try differences? sumption, openness, financial de- * Can economic policies affect per The paper notes that cross-coun- velopment, and the political envi- capita income growth or do poli- try empirical work that fails to ac- ronment, are found to contribute cies have only transitional ef- count for country-specific effects to economic growth. Focusing on fects? In particular, how do the and endoge neity of explanatory the Africa sample alone, there is following affect long-term eco- variables yields inconsistent esti- evidence to support the hypothesis nomic growth: mates of rates of convergence. The that African economies with higher Financial development. Appro- ericon methose incthis savings rates, lower population priate monetary policy pro- paper corrects for those miconsis- growth rates, more outward-ori- mnotes a stable financial envi- tencies. In particular, taking into ented policies, a faster pace of fi- ronment necessary for eco- account country specific effects, as nancial development, and a more nomic growth by maintaining a well as potential endogeneity pro- democratic environment, have low inflation rate. High and van- duces strikingly higher rates of con- tended to grow faster; there is only able rates of inflation are ex- vergence than the ones reported in tentative evidence that government pected to lower the monetary the literature. consumption impinges negatively authorities' credibility and re- The specific results of this study on growth. duce the returns on private sav- are as follows: * Differences in the rates of ings and investment. physical and human capital devel- Government consumption. * Rates of convergence are esti- opment as well as population Higher budget deficits crowd mated at above 10 percent, which growth rates between the OECD out private investment as result implies that countries are very close and Africa samples highlighted the of higher real interest rates. to their steady states. The theoreti- greater importance placed upon Also, higher government spend- cal implication of finding rates of higher human and physical capi- ing creates expectations of fu- convergence in excess of ten per- tal development as well as low ture tax liabilities and hence, cent for both African and OECD population growth rates for the Af- distorts incentives and lowers samples is that countries are very rican countries. growth. close to their steady states. This Trade openness. The proposi- result (i) suggests that observed tion that more outward-ori- differences across per-capita in- ented economies tend to grow comes between countries can be faster has been tested exten- explained primarily by differences sively in the literature, and the in their steady state values, not the This article is based on the publi- majority of the evidence tends distance from their steady states, cation "Revisiting Growth and Con- to support this proposition. and (ii) calls for more policy activ- vergence: Is Africa Catching Up?" by ism: policies can not only acceler- Charalambos G. Tsangarides, Afiica Plieethcat plibercal andeconomi atbhepceo-outieeahn Region Working Paper Series, No. liee tat oliica an ecnomc ae te pce f cuntiesreahin I0, December 2000. For more infor- freedoms are mutually reinforc- their long run levels of incomes, but mi,pease O ing; in this view, icesn moth lmongruantincom lheves ca fetCtsangarides@worldbank.org and/ political rights promotes eco- h 1 ' 1 1 or visit the website, http.// nomic rights and therefore . The neoclassical framework, wwo sworldbank orge afrt ps// stimulatesgrowth.Also,ethcboth in its textbook and augmented p stimulates growth. Also, ethnic bohi t etokadagetd index.htm