The product cycle: Concept and evidence
The idea that there is a logical progression under which newer, more innovative goods are produced in and exported from high-income economies, and later produced in and exported from lower-income economies, is of long standing (Vernon, 1966; see also Posner, 1961). In its most idealized form, a new good would be innovated and produced in the most advanced large economy (in the 1960s, the United States), because that economy had the most innovative capacity and because of “demand-push” innovation to satisfy the tastes of high- income consumers. The good would diffuse, eventually being produced in and exported from other economies than the original innovator. When the technology of production became sufficiently mature, the good would be produced in low-wage economies (in our terminology, downstreaming). This pattern was dubbed the “product cycle” by Raymond Vernon.
These informal theories developed at a time when there was not a lot of formal theory about the dynamics of comparative advantage, and when empirical work in international trade still faced challenges in testing the static implications of the Heckscher-Ohlin model. As it turned out, available tests of the product cycle have shown that it is not the typical pattern for all goods. In fact, patterns of long-run comparative advantage have shown a good deal of persistence, with only occasional downstreaming.
For example, Gagnon and Rose (1995) examine exports of six economies disaggregated to SITC4 from 1965 to 1989. They divide products into three categories — surplus, deficit, and balanced trade — using dividing lines at one standard deviation from the mean. Over their period, only about 1% of products switch between surplus and deficit, implying only a limited role for product cycles. Similarly, Proudman and Redding (2000) measure revealed comparative advantage (RCA) in a study of 22 broad ISIC-defined manufacturing sectors from 1970-74 to 1990-93. For France, Germany, the United Kingdom, and the United States, only a couple of categories switched from RCA 1 > to RCA < 1 over the period in question. Japan, which was still experiencing convergence in per capita income during the period, was the most dynamic, losing RCA in rubber and plastic, textiles and clothing, and other manufacturing and gaining RCA in non-electrical machinery, electrical machinery, motor vehicles, and computers. Even for Japan, the other 15 industries did not change their status with respect to comparative advantage (see also Chapter 3).
It follows that an appropriate theory of the product cycle should account for the prevalence of such stickiness or persistence of comparative advantage in the usual case, and provide for some criterion for deciding when diffusion and downstreaming in the product cycle are actually being observed.