Aggregate production functions and growth economics

Temple, Jonathan R. W. (2006). Aggregate production functions and growth economics. International Review of Applied Economics, 20(3), July, 301-317.

Rigorous approaches to aggregation indicate that aggregate production functions do not exist except in unlikely special cases. This paper considers the awkward implications for growth economics. It provides a conventional defence of growth theory in terms of “parables” and then considers how empirical growth research might avoid the need for aggregate production functions.

[See also the paper immediately below]

Aggregate production functions, growth economics, and the part-time tyranny of the identity: a reply to Felipe and McCombie

Temple, Jonathan (2010). Aggregate production functions, growth economics, and the part-time tyranny of the identity: a reply to Felipe and McCombie.
International Review of Applied Economics, 24(6), November, 685-692.

A recent paper by Jesus Felipe and John McCombie argues that I have understated the importance of the value added identity for conventional estimates of production relationships. This reply seeks to clarify the areas of agreement and disagreement. It suggests that the problems raised by the accounting identity are genuine and deserve to be more widely known, but Felipe and McCombie have sometimes exaggerated the scope of the argument.

A positive effect of human capital on growth

Temple, Jonathan (1999).  A positive effect of human capital on growth. Economics Letters, October, 65(1), 131-134.

Recent studies have found that economic growth appears to be unrelated to increases in educational attainment. In this note I show that there is a correlation in one data set, but it is typically hidden by unrepresentative observations.

You can download the main data set by clicking on benspieg.xls. This should allow all the results in the paper to be replicated. The data set is that created by Benhabib and Spiegel for their 1994 JME article; see their article for details of the sources and data construction.

Some additional data benspieg2.xls may also be useful for those examining the results of Benhabib and Spiegel. In particular this data set also includes population figures.

At the threshold: the increasing relevance of the middle-income trap

with Patrick Imam

Imam, Patrick and Temple, Jonathan R. W. (2024). At the threshold: the increasing relevance of the middle-income trap. CEPR discussion paper no. 19346.

We investigate the existence of a middle-income trap using finite state Markov chains, constant growth thresholds, and mean first passage times. As well as studying output per head, we examine the dynamics of its proximate determinants: TFP, the capital-output ratio, and human capital. We find upwards mobility for the relative capital-output ratio and human capital, but not for relative TFP. The lack of upwards mobility in relative TFP, at least from an intermediate level, suggests that escaping the middle-income category can take many years. A middle-income trap may become increasingly apparent in the years to come.

The autocratic gamble: evidence from robust variance tests

with Fabio Monteforte

Monteforte, Fabio and Temple, Jonathan R. W. (2020). The autocratic gamble: evidence from robust variance tests. Economics of Governance, 21(4), December 2020, 363-384.

This paper examines whether autocracy is a gamble. Using robust variance tests and other analyses, we find that long-term growth varies more across autocracies than across democracies. This remains true even when richer countries are excluded from the sample. We also investigate channels, to see if the higher variance of growth outcomes across autocracies can be traced to a higher variance in investment ratios or productivity growth. Overall, the paper’s findings suggest that growth prospects are more uncertain for autocracies than democracies, and especially for closed autocracies. From the viewpoint of a domestic population, autocracy remains a gamble.

Convergence behaviour in exogenous growth models

with Jochonia Mathunjwa

This paper analyzes several aspects of convergence behaviour in the Solow growth model. In empirical work, a popular approach is to log-linearize around the steady-state. We investigate the conditions under which this approximation performs well, and discuss convergence behaviour when an economy is some distance from the steady-state. A formal analysis shows that convergence speeds will be heterogeneous across countries and over time. In particular, the Solow model implies that convergence to a growth path from above is slower than convergence from below. We find some support for this prediction in the data.

Download the latest version (April 2007) jsm_jt10april2007.pdf. This is a revised version of Mathunjwa, Jochonia S. and Temple, Jonathan R. W. (2006). Convergence behaviour in exogenous growth models. University of Bristol discussion paper 06/590.

Education and economic growth

Economic Growth and Government Policy, HM Treasury, April 2001.

This note provides a brief overview of the effects of education on productivity, drawing on both theoretical work and empirical evidence. The main focus is on the available evidence, and in particular what it tells us about the private and social returns to education. The focus on these returns is motivated by an observation central to economics, namely that it is where private and social returns diverge that there may be a case for using government taxes and subsidies, or other policy instruments, to realign incentives.

Equipment investment and the Solow model

Temple, Jonathan (1998). Equipment investment and the Solow model. Oxford Economic Papers, January, 50(1), 39-62.

This paper investigates the correlation between equipment investment and economic growth, and its compatibility with the Solow growth model. The paper improves on previous work by starting from an explicit theoretical model, taking a rigorous approach to outliers, using instrumental variables, and taking unobserved heterogeneity into account. Rates of return to investment, and their precision, are estimated. The main finding is that the implied returns to equipment investment are very high in developing countries.

JEL classification: O11, O16, O57

To download the data set and documentation, hold down the left shift key and click on the link, or alternatively press the right mouse button and choose ‘save link as’.

equipsol.txt	Documentation
equipsol.csv	Data set for Tables 1-4

Generalizations that aren’t? Evidence on education and growth

Temple, Jonathan R. W. (2001). Generalizations that aren’t? Evidence on education and growth. European Economic Review, May, 45(4-6), 905-918. Reprinted in Belfield, C. (ed.) Modern classics in the economics of education, Edward Elgar, 2006.

Several papers have suggested that the relationship between changes in average schooling and growth is weak in the cross-country data. This might call into question the relevance of micro estimates of returns to schooling, at least for developing countries. This paper examines the reliability of some of the aggregate evidence, and presents an alternative framework for analysing these questions.

GMM estimation of empirical growth models

with Stephen Bond and Anke Hoeffler

Bond, S., Hoeffler, A. and Temple, J. (2001). GMM estimation of empirical growth models. CEPR discussion paper no. 3048.

This paper highlights a problem in using the first-differenced GMM panel data estimator to estimate cross-country growth regressions. When the time series are persistent, the first-differenced GMM estimator can be poorly behaved, since lagged levels of the series provide only weak instruments for subsequent first-differences. Revisiting the work of Caselli, Esquivel and Lefort (1996), we show that this problem may be serious in practice. We suggest using a more efficient GMM estimator that exploits stationarity restrictions, and this approach is shown to give more reasonable results than first-differenced GMM in our estimation of an empirical growth model.

Download the paper: bht10.pdf

Growing into trouble: Indonesia after 1966

Temple, Jonathan (2003). Growing into trouble: Indonesia after 1966. In Dani Rodrik (ed.) In search of prosperity: analytic narratives on economic growth. Princeton University Press, Princeton, 2003.

This paper analyzes the remarkable growth experience of Indonesia since 1966. Over a thirty year period, GDP per capita rose more than fourfold, despite unfavourable initial conditions, some weak institutions, and flawed microeconomic policies. The paper attributes this strong performance to a mutually reinforcing combination of political stability and competent macroeconomic policy, and some important instances of good fortune. It explores the origins of good policy and analyzes three of the main external shocks. The paper also argues that rapid growth interacted with weak institutions in a way that contributed to the severity of the crisis of 1997-98.

Download the final (August 2001) version of the paper: indodp4.pdf

Growth effects of education and social capital in the OECD countries

Temple, J. R. W. (2001). Growth effects of education and social capital in the OECD countries. OECD Economic Studies, No. 33, 2001/2, 57-101.

This paper surveys the empirical literature on the growth effects of education and social capital. The main focus is on the cross-country evidence for the OECD, but the paper also briefly reviews evidence from labour economics, to clarify where empirical work on education using macro data may be relatively useful. It is argued that on balance, the recent cross-country evidence points to productivity benefits of education that are at least as large as those identified by labour economists. The paper also discusses the implications of this finding. Finally, the paper reviews the emerging literature on the benefits of social capital. Since this literature is still in its early days, policy conclusions are accordingly harder to find.

A French translation is available, in Revue economique de l’OCDE, 33, under the title Effets de l’education et du capital humain sur la croissance dans les pays de l’OCDE. A much earlier version of the paper appears in Helliwell, John F. (ed.) The contribution of human and social capital to sustained economic growth and well-being. OECD and HRDC, Canada, 2001.

Have autocrats governed for the long term?

with Emanuele Millemaci and Fabio Monteforte

The short answer is: probably not. We infer the priorities of national governments from observed outcomes, constructing a statistical proxy for long-term benevolence. We show that, on average, democracies score more highly on our measure. We then investigate whether variation in long-term benevolence can explain the ‘autocratic gamble’ – the well-known tendency for growth rates to vary more widely across autocracies than across democracies. We show that the distribution of long-term benevolence under democracy first-order stochastically dominates its distribution under autocracy. The rapid growth seen in some autocracies probably originates in regime self-interest rather than unusually benevolent leadership.

Initial conditions, social capital, and growth in Africa

Temple, Jonathan (1998). Initial conditions, social capital, and growth in Africa. Journal of African Economies, October, 7(3), 309-347.

Observable variables capturing initial conditions can account for well over half the variation in developing country growth rates. This paper investigates their role in explaining Africa’s recent economic history. Should the origins of slow growth be traced to Africa’s social arrangements, high inequality, and ethnic diversity? Based on cross-country empirical work, this paper argues that the best answers are yes, no, and maybe.

The long-run implications of growth theories

Temple, Jonathan (2003). The long-run implications of growth theories. Journal of Economic Surveys, July, 17(3), 497-510.

This paper draws attention to some possible misunderstandings concerning the place of long-run outcomes in growth theory. It argues that the traditional emphasis on these outcomes is often misguided. As a result of this emphasis, too much attention is paid to the role of knife-edge assumptions, and researchers may be led to interpretations of the evidence, or models of the growth process, that are not wholly sensible.

Political institutions and output collapses

Imam, Patrick and Temple, Jonathan R. W. (2024). Political institutions and output collapses. European Journal of Political Economy, 85, December, 102573.

This paper examines whether major output collapses are more likely under autocracy. Using data on 123 developing countries over 1971-2016, we model the joint evolution of output growth and political institutions as a finite state Markov chain with a two-dimensional state space. We study how countries move between states. We find that growth is more likely to be sustained under democracy than under autocracy; output collapses are more likely to deepen under autocracy; and stagnation under autocracy can give way to outright collapse. Democratic countries appear to be more resilient.

Download CEPR discussion paper no. 19074.

Robustness tests of the augmented Solow model

Temple, Jonathan R. W. (1998). Robustness tests of the augmented Solow model. Journal of Applied Econometrics, July-August, 13(4), 361-375.

This paper demonstrates some techniques for testing the robustness of cross-section and panel data regressions, and applies them to the influential augmented Solow growth model. The paper focuses on robust estimation and analysis of sensitivity to measurement error. In particular, it is shown that estimated technology parameters and convergence rates are highly sensitive to measurement error.

The data set and computer software are available from the JAE’s data archive

Social capability and economic growth

Temple, Jonathan R. W. and Johnson, Paul A. (1998). Social capability and economic growth. Quarterly Journal of Economics, August, 113(3), 965-990.

The conventional wisdom is that post-war economic growth has been unpredictable. In the 1960s, few observers accurately forecast which countries would grow quickly. In this paper, we show that indices of social development constructed in the early 1960s have considerable predictive power. These results indicate the importance of “social capability” for economic growth. We emphasise that social arrangements matter for reasons beyond those discussed in recent work on trust and social capital. However, we are also able to show that one of the indices may be a useful proxy for social capital in developing countries.

To download the Adelman-Morris index and documentation, hold down the left shift key and click on the link, or alternatively press the right mouse button and choose ‘save link as’.

socdev.csv	Data set
socdev.txt	Documentation for the data set (text)

St. Adam and the Dragons: Neo-classical economics and the East Asian Miracle

Temple, Jonathan (1997). St. Adam and the Dragons: Neo-classical economics and the East Asian Miracle. Oxford Development Studies, October 1997, 25(3), 279-300.

This paper addresses recent explanations for the East Asian miracle. The argument that growth is entirely due to the accumulation of inputs is assessed and found wanting. There is still a place for the view that attributes success to activist policy. This is especially so, since the other explanations of East Asian success are rarely wholly convincing. The paper demonstrates that popular cross-country models of growth usually fail to explain the East Asian experience, but presents evidence supporting emphasis on favourable initial conditions, including early specialization in manufacturing.