The Misery of the “Dutch Disease and Deindustrialization” Argument


High commodity prices, abundance of liquidity, and sustained growth in Asia: all factors of good luck in the current foreign environment of Latin America, right?

ByOtaviano Canuto

Not quite, according to frequent references to “deindustrialization” and “resource curse” afflicting the region as features of the international labor division that has been consolidating along the current world economic cycle.

This pessimistic view has two components. One relates to the structural downward move of the price of labor-intensive goods and services relative to natural resource-based ones and the associated geographical dislocation of manufacturing activities toward Asia (either in relative or absolute terms), combined with increased natural resource-linked activities in Latin America. On the same direction, a second strand of “bad luck” arguments alludes to a kind of “Dutch disease” contracted by Latin American economies in the current cycle. The substantial increase of value of exports of natural resources, in conjunction with the liquidity glut abroad and huge capital inflows would be generating insurmountable pressures toward overvalued local currencies, and a corresponding second round of deleterious effects upon domestic labor-intensive production.

Those who characterize this double trend as intrinsically negative express a view that the specialization in natural resource-based production is necessarily harmful to development in the long run. In our view, both lines of arguments often convey an ungrounded prejudice in terms of sector analysis, potentially leading to a mistaken policy agenda if taken for face value in the usually superficial manner with which they are formulated.

Let us recall the concept of “Dutch Disease”. The term appeared for the first time in an article in the November 26, 1977 issue of the The Economist magazine. It referred to Holland’s discovery of natural gas in the North Sea in the 1960s, and how the real appreciation of the Royal Dutch as a consequence of increased exports contributed to a decline in the local manufacturing production. Since then, the expression has been used whenever one sees harmful consequences of discoveries – or sudden upward revaluations – of exported natural resources in developed and developing countries alike.

The theoretical argument about the ultimate effects of such discoveries or permanent revaluations of domestic natural resources on the local GDP composition and labor allocation – after full adjustment in both goods and labor markets – is clear-cut. Independently of whether nominal exchange rates are fixed or flexible, as well as of whether or not domestic aggregate demand is continuously fine-tuned to keep the local price index stable, any substantial and prolonged surge in export revenues of a limited set of goods tends to rob labor and capital from other tradable sectors. See Deindustrialization and the Dutch Diseaseregarding the different channels by which the Dutch Disease may inflict the competitiveness of other existing activities in the economy.

In order for such a blessing by nature to be ironically treated as a “disease” in Latin America, amplified by the huge extension of low-cost labor-intensive production in Asia, one has of course to state why it entails negative effects for the region’s income growth, levels or quality of employment, etc. despite improved terms of trade. And at this juncture of History and knowledge one has to go much beyond the outdated generic contraposition of “primary products” versus “manufacturing”, as it was the case at the time of the old-fashioned Prebish-flavored controversies. Furthermore, let me take as reasonably evident that in the long run the system of prices and the corresponding adjustments prevail upon “structural rigidities” in factor and product markets. It does not suffice simply to proclaim that natural resource-based patterns of specialization are intrinsically inferior to manufacturing-based ones!

This falls far short from saying that “patterns of specialization do not matter”, as traditional trade theory has predominantly led us to think. We suggest here two recent papers that take stock of theoretical arguments and empirical evidence favorable to two key propositions that contrast with what most conventional trade models convey: firstly, different patterns of specialization do affect future growth of individual economies (Hausmann, Hwang and Rodrik), and secondly, such patterns are “sticky” and tend to exhibit path-dependence and differing degrees of scope for evolution, diversification, and technological change (Hausmann and Klinger).

Nevertheless, one should not incur in a mistaken full-of-prejudice association between natural resource-based products and low requirements in terms of technology and skilled labor. Take for instance the case of agro-energy (biofuels) or soybeans in Brazil, where the bioengineering of seeds and the diffusion of computer-aided weather and soil analysis have been at the core of astounding high rates of productivity increases. Besides those backward linkages and in-house technological requirements, that progress generated favorable inducements forward (as e.g. new flexible sugar-ethanol mills close to crop sites and new flex-fuel ethanol-user vehicles). 

Think also of Brazilian savannahs (“cerrados”), which have become fertile areas as a result of agricultural research, whereas they were considered as desert-like zones not long ago. The fact is that the “discovery” and incorporation of “cerrados” as natural resources has been made feasible by both the dynamism of agricultural markets abroad and technological change, and it has crowded-in capital accumulation, rather than stolen factors of production from more prosperous activities. One should also keep in mind the indirect positive effects that such a “discovery” brought to other activities, by e.g. helping decrease levels of the country risk premium.

Nor it is appropriate to identify the mere presence of manufacturing activities with automatic progress, even when these activities include unskilled labor-intensive segments of production chains in industries classified as “high-tech”. If human and physical capital accumulation, as well as local mastery of technology, does not evolve accordingly, a country may get stuck in producing low technology, low skilled-labor manufacturing activities.       

The bottom-line is straightforward: instead of playing with protectionist measures to halt what is taken as “Dutch disease” and mischievous deindustrialization forces, Latin American economies might better focus on mitigating the real diseases, i.e. low levels and quality of education, insufficient capillarity of science and technology in the production system, dead-weight inefficiencies and lower private willingness to invest derived from institutional failures in the business environment, lack of fiscal space for public investments in infrastructure and other structural malaises.