Trade and Technological Progress in Endogenous Growth Models

Alexis Murphy – Senior Sophister

Solow's Growth Model, which for years dominated thinking on economic growth, is criticised for the exogeneity of growth in the model. This lead to the emergence of endogenous growth models. In this essay Alexis Murphy examines this approach and in particular the role of trade and technological progress in the model.

This paper reviews some of the recent literature that deals with the effect of trade on the ‘residual' or technological progress element in endogenous growth models, and uses this to highlight some inherent difficulties in addressing this issue.

The two central issues that are to be discussed in this essay are trade and growth. Both issues are highly topical for a variety of reasons. Growth theory is at the cornerstone of modern macroeconomic analysis. The reason is self evident: we are preoccupied with how our economies can grow in order to guarantee a better standard of living on aggregate for the agents in these economies. Trade theory is viewed as important because it analyses the best means for each country to develop its comparative advantage. Both of these branches of modern economics suggest that economic activity far from being a zero-sum game, can be beneficial to all participants.

An analysis of the effects of trade is furthermore important because of the rapid speed at which the creed of open-economy liberalisation and globalisation is being spread, the growing trend towards regional integration, and the movement towards world economic integration through the World Trade Organisation (WTO). Improvements in IT and transport have also contributed significantly to promoting global integration.

Many developing and transition economies have followed the trend towards liberalisation either voluntarily, in a desire to replicate the successes of economies such as Hong-Kong or Taiwan, or under the compulsion of the IMF and the Structural Adjustment Programs (SAPs) of the World Bank. This is perhaps the clearest example we have to date of the paradigm shift from relatively closed planned economies to free-market open economies.

The role of the IMF and World Bank in encouraging trade liberalisation implicitly imposes a very US style perspective on the political economy of the developing world. The sovereignty and ability of many developing countries to choose for or against trade liberalisation is constrained by their economic weakness relative to the developed world. There may be a short-run bandwagon effect where one developing country follows its neighbour in order not to miss out on the potential upside consequences of economic liberalisation. By discussing the issue of trade and technological progress, it may appear that the ones who have the most to gain from trade liberalisation are the high-tech countries such as the U.S. who have significant comparative advantages in terms of R & D and human capital relative to transition and developing economies.

However, even in the developed world, the gains from trade and integration are far from clear. The benefits of technology spillovers are significant in Ireland, where the country has benefited from Foreign Direct Investment (FDI) by a variety of high-tech multinationals. Is it the case that the current supremacy of Silicon Valley is beneficial to world economic growth because of positive spillovers of new technologies, or is it the case that high R & D in the U.S. and their apparent comparative advantage in new technologies is giving the U.S. a comparative advantage which may be detrimental to resource re-allocation and long-run growth in the rest of the world?

The relationship between the residual of endogenous growth models and trade is important in view of certain recent trends which have affected developed economies in particular. The growing importance of the services industry in developed economies indicates the need to develop models in which growth accounts for more sophisticated inputs than simply labour and capital. The importance of the services industry and its corollary, human capital, show the way in which the inputs of production are now much more difficult to categorise in the simplistic way of traditional growth models. This, in part, has lead to an in-depth investigation of the ‘residual' in growth models.

By ‘residual', we understand any factor that affects output, other than labour or capital. The definition is suitably vague as to be able to include anything from Research and Development to climatic conditions. However in the recent literature, the term: ‘residual' has been replaced by technology, technological progress, or technical progress. The vagueness of the definition also introduces the theme of my essay which is that the taxonomic difficulties involved in defining the residual (and in parallel, openness, in the trade sense) lead to inherent problems in both the theoretical and empirical material relating to the relationship between trade and endogenous growth. Particularly, the surreptitious replacement of ‘residual' which has the connotations of being the ‘black box' of economic growth, by ‘technological progress' has narrowed some of the scope for research. This is not to say that the existing literature on technological progress and growth should be condemned, but points to the importance of broadening the ambit of the issues under discussion in the field today.

There are two main constraints faced by researchers in this field. From the theoretical perspective, there is a need to present a relatively simple model that isolates the key factors underlying the growth in technology and the manner in which it is dissipated through international trade. Linked to this is the main constraint of empirical work, which involves finding a measurable proxy or proxies of the determinant(s) of technological progress, and similarly measures of openness.

Several theoretical models have been developed to analyse the effects of trade on technology and thereby on long-term economic growth. The three most important contributions build on Romer's short article entitled ‘Endogenous Technological Change'. The model focuses on the importance of human capital as the catalyst of R & D, which in turn stimulates the residual of the endogenous growth model. The benefits of R & D are in terms of the new innovations themselves which guarantee monopolistic rents for the entrepreneurs through patenting, and because new innovations add to the stock of non-excludable knowledge, hence raising the productivity or reducing the marginal cost of new innovations. He then suggests that trade would be beneficial to technological innovation, superficially by providing access to new markets, but more fundamentally by integrating the market for human capital. Romer cites the example of US counties that had access to navigable waterways during the 19th century which had higher rates of patenting than more isolated counties.

Grossman and Helpman then wrote a series of articles that developed open economy, general equilibrium frameworks where the effects of trade on technology were investigated. This issue is also discussed in some detail in their book Innovation and Growth in the Global Economy. Their main model is a three-good (intermediate goods, final goods, and R & D), two-country economy, with skilled and unskilled labour as the two inputs. The central theme is that even if there is a difference in the relative factor endowments of the two countries (for example, one country may be rich in human capital or skilled labour), trade will have created static gains for both countries. There is a trade-off between the production of final goods and R & D, both of which are human capital intensive. If demand is high for final goods, there may be a re-allocation of resources from R & D to final goods production. This will have the effect of increasing current output at the cost of tomorrow's output. However this increased demand is mitigated by the fact that increased demand for final goods also increases the incentives and rewards of R & D.

Their broader conclusion is that factor price equalisation will eventually ensure that both trading countries enjoy identical growth rates of output and consumption irrespective of their factor endowment. Another conclusion is that "an equiproportionate once-and-for-all increase in the effective labour force accelerates long-run growth" (an increase in the ‘effective labour force' means the productivity of labour, through inter alia new technologies and innovation). This is an extension of the truism postulated by Romer that policies aimed at increasing human capital should be unambiguously positive for long-run growth. Problems in an open economy arise when increases in technology in both countries are not at a common rate, or if after an increase in technology in one country the spending shares on final goods change. This last qualification highlights the problem of steady-state analysis as opposed to dynamic analysis.

The limitations of their model may also help to explain the emergence of a technology-gap between developed countries and less developed countries. A static analysis of such a gap in the context of trade was carried out by Young (1991) where he considers economies without knowledge spillovers but where differences in the potential for learning by doing (which is determined by the stock of technology) enable the LDC to outgrow the DC. However in recent history the technology gap seems to have increased rather than decreased.

Grossman and Helpman themselves point to instances in which trade may not be beneficial to growth in the long run. This occurs when the country that is relatively well endowed in technology is bigger in size than its trading partner. They conclude that because knowledge spillovers are imperfect, factors such as national factor endowments will lead to long-run growth differences between countries.

From the empirical perspective, as alluded to above, there are serious problems in terms of measuring both the technology variables in growth and openness. Although the empirical evidence seems to overwhelmingly show that trade is beneficial to growth, it is difficult to identify the exact mechanisms by which trade affects growth. Does convergence occur through capital accumulation, factor price equalisation, knowledge spillovers or trade-mediated technology transfers?

In the case of many East-Asian economies, the evidence seems to suggest that the benefits from their investment in R & D occurred only because they initially operated as relatively closed, import-substituting economies, and then benefited from trade once their strong high-tech industries had been developed.

Coe and Helpman focus on the effects of R & D among trading partners for a variety of developed open economy countries. This empirical study follows on from the theoretical literature of Helpman and Grossman, which focuses on R & D as the primary source of innovation in the economy. The transmission mechanism can occur directly by one partner learning about "new technologies and materials, production processes, or organisational methods" or indirectly by access to imports that facilitate innovation. As a proxy for foreign R & D capital stocks they use imported weighted sums of trade partners' cumulative R & D spending. Their conclusions suggest that trade does affect Total Factor Productivity (TFP) growth positively. Furthermore they identify the presence of large spillover effects of R & D: "about one quarter of the total benefits of R & D investment in a G7 country accrue to its trade partners."

The study led to two recent articles that criticise some of the econometric techniques used by Coe and Helpman. Lichtenberg et al corrects two faults in the Coe and Helpman paper. The first is the ‘aggregation bias' which occurs in the weighting scheme used to calculate foreign R & D stocks. They also correct an ‘indexation bias'. However their corrections corroborate the conclusions presented by Coe and Helpman.

Keller is more critical of Coe and Helpman. He uses a ‘Monte-Carlo' experiment which estimates international R & D spillovers among randomly matched trade partners. He concludes:

"Considering the complicated with respect to the data generation process underlying the R & D and TFP series, the results in this paper might in fact seem rather unsurprising. However, it is clear from this analysis that the extent to which R & D spillovers are related to the pattern of international trade must be estimated in a model which allows simultaneously for trade-unrelated international technology diffusion".

This econometric evidence highlights the difficulties that arise from trying to analyse the relationship between trade and technology growth. Long-term economic and political histories suggest that openness to trade and foreign influences foster new ideas and innovation which are ultimately beneficial to mankind. The deliberate policies of Ming China to shut itself off from the outward world seem to explain the failure of that country to develop in subsequent centuries. Similarly trade and travel in Europe from the 15th century onwards seems to explain the manner in which a variety of new ideas, innovations, and technologies helped to turn Europe into the dominant world power in subsequent centuries.

However is it important to reconcile such qualitative facts with the more quantitative economic analysis of today. The theoretical research in the field is obliged to make grossly simplifying assumptions. Similarly the empirical literature has problems both in using the correct proxies for the ‘residual' and in terms of identifying transmission mechanisms through which such proxies affect output.

R & D is a convenient proxy for the resources being put into innovations, yet it fails to account for many innovations. The operating system DOS was created with very little R & D costs, but had massive consequences on the world of PC's. Similarly, Viagra was discovered by mistake, rather than as the result of a conscious investment decision. There are many random elements that influence the creation of ideas and therefore the residual in endogenous growth models. However it also seems clear that certain parameters both on a national and international level are conducive to encouraging this form of growth. Investment in human capital is highlighted by Romer. If Grossman and Helpman are partially correct in their approach, international patenting laws should be enforced in order to create the monopoly rents that encourage R & D which may only pay the individual entrepreneur through exploitation of such rents.

Institutional factors such as the enforcement of property rights, the efficiency of legal and fiscal systems, and a variety of other social and political practices are the key to creating the background against which innovations can occur. The stagnation of an economy such as Russia can largely be attributed to the failure of such systems and practices.

This would point to an interesting field of research that would be focused on the effect of trade on such practices. It is illegal for U.S. companies to pay bribes or kickbacks in foreign countries, whereas there are no such legal restrictions for European firms. Perhaps trade could be seen as a means of propagating better standards of transparency, honesty, and fairness. The implicit adoption of such practices would then create the necessary parameter for strong growth in TFP. This would present an alternative picture to that of the foreign firm or multinational exploiting a weaker trading partner.

This is one proposed approach that would be compatible and complementary with some of the current research reviewed above. It is an approach that may provide a certain methodological incommensurability with the existing approach, yet it may also provide a richer understanding of the issue at stake.


Aghion, P. and P. Howitt (1990) "Model of Growth Through Creative Destruction" in NBER Working Paper 3223. Cambridge, MA.

Aghion, P. and P. Howitt (1998) Endogenous Growth Theory. MIT: Cambridge, Mass.

Becker, G. (1964) Human Capital. Columbia University Press: New York.

Borensztein, E., J. De Gregorio, & J.W. Lee, (1998) "How Does Foreign Direct Investment Affect Economic Growth?" in Journal of International Economics, 45, pp. 115-135.

Coe, D. & E. Helpman (1995) "International R & D spillovers" in European Economic Review 39, pp. 859-887.

Greenaway, D., W. Morgan, & P. Wright (1998) "Trade Reform, Adjustment and Growth: What does the evidence tell us?" in Economic Journal 108, pp. 1547-1561.

Grossman, G. & E. Helpman (1990) "Comparative Advantage and Long-Run Growth" in The American Economic Review 80 (4), pp. 796-815.

Grossman, G. & E. Helpman (1991a) ‘Trade, Knowledge Spillovers, and Growth' in European Economic Review 35, pp. 517-526.

Grossman, G. & E. Helpman (1991b) Innovation and Growth in the Global Economy. MIT: Cambridge, MA.

Grossman, G. &E. Helpman (1994) ‘Endogenous innovation in the Theory of Growth in Journal of Economic Perspectives. Winter, 8 (1), pp. 23-44.

Jones, Charles (1998) Introduction to Economic Growth. W. W. Norton: New York

Keller, Wolfgang (1998) "Are International R & D Spillovers Trade-Related? Analyzing Spillovers Among Randomly Matched Trade Partners" in European Economic Review 42, pp.1469-1481.

Kennedy, Paul (1988) The Rise and Fall of the Great Powers. Unwin Hyman: London.

Lichtenberg, F. R., & B. van Pottelsberghe de la Potterie (1998) "International R & D spillovers: A comment" in European Economic Review 42, pp. 1483-1491.

Ocampo, J. & L. Taylor (1998) "Trade Liberalisation in developing economies modest benefits but problems with productivity growth, macro prices, and income distribution" in The Economic Journal, 108 (September), pp. 1523-1546.

Romer, David (1996) Advanced Macroeconomics. McGraw and Hill: New York.

Romer, Paul (1990) "Endogenous Technological Change" in Journal of Political Economy Oct. 98 S71-S102.

Romer, Paul (1994) "The Origins of Endogenous Growth" in Journal of Economic Perspectives. 8 (1), pp. 3-22.

Solow, Robert (1957) ‘Technical Change and the Aggregate Production Function' in Review of Economics and Statistics 39 (August) pp. 312-320.

Weitzman, Martin (1996) ‘Hybridizing Growth Theory' in The American Economic Review 86 (2) pp. 207-212, May.