The greatest improvement in the productive power of labour, and the greater part of the skill, dexterity, and judgement with which it is any where directed,
or applied, seem to have been the eff ects of the division of labour . . . This divi- sion of labour, from which so many advantages are derived, is not originally the eff ect of any human wisdom, which foresees and intends that general opulence to which it gives occasion. It is the necessary, though very slow and gradual, conse- quence of a certain propensity in human nature which has in view no such exten- sive utility; the propensity to truck, barter, and exchange one thing for another.
(Adam Smith (1776 [1976]), Wealth of Nations, Chapters I & II, pp. 7 and 17) The openness view of development goes back at least to Adam Smith.
Smith (1776 [1976]) argues that openness to trade increases the size of the market, which raises the possibility of greater division of labour. Division of labour in turn improves productivity and productivity improvement induces faster economic growth.7
The neo- classical theory, however, holds a slightly diff erent view.
According to this theory, reduction in trade barriers opens up the possibil- ity of a more effi cient exploitation of comparative advantage through real- location of factors. Labour and capital move towards their highest valued uses improving overall productivity and the welfare of the economy.
Growth takes place in the economy due to transitional dynamics. In other words, growth lasts only for the duration of the transitional period and stops after the economy reaches its new steady- state levels of capital and output per worker.
Another signifi cant theory in the openness and growth literature is the technology transfer view. According to the closed economy neo- classical growth models, given the technology level at a particular point in time, a country accumulating more physical and human capital grows faster than a country accumulating less of the same, but all of them converge to the same long- run steady- state equilibrium rate of growth (Solow, 1956;
Swan, 1956; Mankiw et al., 1992). In this model, income per capita can grow in the long run only when there is exogenous technological progress.
In real life, we hardly observe convergence of income. In contrast, what we notice is that the rich economies are growing faster than the poor ones and the gap between the rich and the poor is widening over time.8 Coe and Helpman (1995) point out that this gap is due to the variable rate of technological progress. They fi nd evidence that technology catch up or R&D spillover bridges this gap. Dowrick and Rogers (2002), on the other hand, report that the gap is due to variable rates of capital accumulation as well as technological progress. Howitt (2000) builds a theoretical model to show that because of technology transfer, R&D performing countries converge to parallel growth paths, whereas others stagnate. Several others, including Coe and Helpman (1995), identify trade openness as a medium of technology transfer. In other words, what these studies argue is that the follower economies adopt technology and knowledge developed in
the advanced economies and the transfer takes place largely through international trade. Therefore, eff ective participation in international trade and opening up by reducing trade restrictions exposes economies to new knowledge and new technology that helps them to grow in the long run.
Other infl uential studies in the openness and growth literature are Sachs and Warner (1995a) and Frankel and Romer (1999).
Sachs and Warner (1995a) use the growth accounting approach to show that trade openness is the key determinant of growth. They calculate trade openness using information on average tariff levels, non- tariff barriers, black market premiums, state monopoly on major exports and so forth.
Their regression results suggest, on average, ‘open’ economies grow some- where between 2 to 3 percentage points faster than ‘closed’ economies.9 Their study, however, is not free from criticism. Rodrik and Rodriguez (2000) show that the Sachs and Warner measure of openness suff ers from over-reliance on the black market premium and state monopoly on major exports criteria.
Frankel and Romer (1999), on the other hand, resort to level accounting using instrumental variable regressions to conclude that trade openness is a key factor infl uencing economic development. They construct an instru- ment by separating out the infl uences of income, population size and land area, which they call the constructed openness. This variable they claim yields more accurate estimates of the relationship between openness and per capita income. They report a strong and positive eff ect of openness on income per capita.
Even though there has been a fair bit of agreement about the associa- tion of trade openness with growth for at least the last couple of decades, recently these results have come under fi re in a series of papers. These papers argue that trade openness loses importance once institutional quality is introduced as a control in a regression framework. Rodrik et al. (2004) reach a similar conclusion, which we discussed previously. One possible explanation is openness enhancing institutional development.
Therefore, a regression model with openness and institutions as controls is not picking up any individual eff ect of openness on income as it is operating through the institution channel. This view is supported by Wei (2000) who suggests that more open countries face greater losses from corruption than less open ones as the loss from corruption is higher in the case of foreign transactions. Therefore, open countries have a higher incentive to develop better institutions. Another explanation comes from Dollar and Kraay (2003). They argue that cross- country regressions of log- level per capita gross domestic product (GDP) on instruments of institutions and openness do not refl ect the relative importance of these
variables in the long run. The high correlation between institutions and openness does not allow regression models to estimate the actual rela- tionship. According to their analysis, regression of changes in decadal growth rates on instrumented changes in trade and changes in institutional quality show a signifi cant eff ect of trade openness on growth. They report that trade openness and institutions are important for economic growth in the very long run, but trade has a relatively larger role over institu- tions in the short run. They do not control for geography, religion and culture. Also, their study explains changes in growth rates, rather than growth itself, and growth accelerations are hard to observe in the long run (Jones, 1995). This also brings about the question of whether the spurt in growth rate that comes out of changes in institutions and openness is a mere impulse which tapers off in the long run or something that is sustainable.
In a recent study, Chang (2008) argues that overtly strong commit- ment to free trade is not good for growth. As an alternative strategy, he promotes ‘heterodox capitalism’, which includes a combination of protectionism, government promotion of favoured industries, strong state- owned enterprise and heavy regulation of foreign direct invest- ments. Chang’s main piece of evidence in support of this argument is as follows. He shows that the developing countries grew at a rate two times faster during the ‘heterodox’ era of the 1960s and 1970s than during the 1980s onwards ‘free trade’ era. The average growth rate for developing countries during the 1960s and 1970s, according to Chang, was 3 per cent.
In contrast, the average growth rate during the period 1980 to 2002 for the same country group was 1.7 per cent. Easterly (2009), however, fi nds several inconsistencies in the fi gures reported by Chang if the 1980 turning point is shifted to 1983 or 1982. He also shows that the average growth rate for developing countries for the period 1983 to 2008 (2.7 per cent) is in fact higher than the same for the period 1960 to 1982 (2.6 per cent).
Therefore, Chang’s argument heavily relies on picking 1980 as the turning point. Easterly (2009) also shows that picking 1980 as the turning point in itself is misleading as trade liberalization was fi rst pushed by the IMF and the World Bank in the aftermath of the Third World debt crisis in 1982.
In sum, even though Chang (2008) perhaps appropriately criticizes the one size fi ts all free trade policy as a development strategy for all develop- ing countries, he also appears to be guilty of moving to the other extreme of making strong claims in favour of protectionism as an alternative.