Title: Economic Development as Self-Discovery
Author: Ricardo Hausmann and Dani Rodrik
Scope: 3 stars
Readability: 2 stars
My personal rating: 4 stars
See more on my book rating system.

Topic of Book

Researchers apply the concept of economic complexity to give developing nations advice on how they can enrichen themselves.

Key Take-aways

• Developing nations need more than just foreign technology and good institutions.
• Fast-growing nations, such as South Korea, Japan, China and India often departed greatly from what economist say is the ideal.
• It takes trial-and-error learning for developing nations to identify what products they are good at producing.
• Industrial success entails concentration in a relatively narrow range of high-productivity activities.
• The specific product lines that eventually prove to be hits are typically highly uncertain and unpredictable.

Important Quotes from Book

The theory and practice of economic development have converged in the last two decades on a remarkably simple view of growth fundamentals. Stated in its starkest form, this view is that economic growth requires two things: foreign technology and good institutions. This perspective is well grounded in the neoclassical model of economic growth, which predicts that poor countries will experience rapid convergence with advanced economies once they have access to state-of-the-art technologies and their governments respect property rights. From this perspective, failure to grow can be attributed two one or both of two pathologies. One is the “closed-economy” pathology, in which governments retard technological progress by reducing access to foreign investment and imported capital equipment and intermediate goods. The other is the “corruption” pathology, in which political leaders fail to respect property rights and screw things up deliberately in order to enrich themselves and their cronies. The natural remedies for these pathologies are economic openness and improved governance. With these remedies in place, economic growth should follow naturally. In the words of a recent paper on growth: “Once a developing country government establishes the rules to a fair game and ensures their enforcement, it would be well advised to stand back and enjoy the self-generating growth” (Roll and Tallbott 2001). Reforms in the areas of governance and openness have accordingly become the cornerstones of development strategy in virtually every country during the last fifteen years.

Actual development experience presents at best an awkward fit with this conception of growth basics. The first of these relates to the economic performance of Latin American countries during the 1990s. By the standards of the consensus view, the quality of policymaking in Latin America has been unmistakably and significantly better in the 1990s than it was two or three decades before… Yet these economies’ response to the reforms has been extremely disappointing. Economic growth in the 1990s has been on average much lower than in the decades before 1980, even though the region was closed to trade and had poorer institutions by most benchmarks in the earlier period. In fact, only three Latin American countries (Chile, Uruguay, Argentina) have outperformed in the 1990s their record during 1950-1980. Of these three, only Chile remains a clear success.

The other side of the coin, and the second strand of ill-fitting evidence, is presented by the experience of countries that have had greater success. Some of the most important among these countries—South Korea and Taiwan since the early 1960s, China since the late 1970s, and India since the early 1980s—have done extremely well under quite heterodox arrangements. All these countries have emphasized exports and none grossly violated property rights. But their strategies bear only passing similarity to today’s consensus precepts. South Korea and Taiwan retained high levels of protection for a long time, and made active use of industrial policies… China achieved phenomenal growth rates without formally enacting private property rights—something that would have seemed impossible to many economists had the Chinese miracle not taken place. India barely reformed its incredibly cumbersome trade and industrial regime before its economy took off in the 1980s. And even after more ambitious reforms were enacted in the early 1990s, the Indian economy remained among the world’s most protected.

Our focus is on a particular type of learning that we believe has not received enough attention in the literature on economic development: learning what one is good at producing.  We emphasize that this a key challenge in the process of transformation into a modern economy. Neither economic theory nor management science is of much help in helping entrepreneurs (or the state) choose appropriate investments among the full range of modern-sector activities, of which there could be tens of thousands.

If learning what a country is good at producing requires an investment and the returns to that investment cannot be fully appropriated, the problem faced by potential entrepreneurs in developing countries is identical to the problem faced by innovators in the advanced industrial countries. However, the policy environments facing the “innovators” in the two settings are quite different. Typically, the intellectual property regime protects discoverers of new goods through the issuance of temporary monopolies, i.e., patents. But the investor in the developing country who figures out that an existing good can be produced profitably at home does not normally get such protection, no matter how high the social return.

Laissez-faire cannot be the optimal solution under these circumstances, just as it is not in the case of R&D in new products.

With regard to Asia, our framework helps us understand why the provision of rents by governments (through trade protection, temporary monopolies, subsidized credits, and tax incentives) has gone hand in hand with industrial growth and diversification. These rents may have been needed to stimulate the cost discovery process. At the same time, our framework highlights how rents can backfire if governments do not complement them with policies that rationalize industries and discipline firms that end up with high costs. We would hypothesize that the absence of such discipline was a hallmark of import substitution policies (ISI) in Latin America.

The key policy recommendation that comes out of our framework is that governments need to play a dual role in fostering industrial growth and transformation. They need to encourage entrepreneurship and investment in new activities ex ante, but push out unproductive firms and sectors ex post. This is of course easier said than done. The specifics of how this can be managed is likely to differ considerably from country to country.

A crude, but useful characterization of the policy environments in East Asia and Latin America, as viewed from the perspective of our framework, would be as follows. East Asian governments provided their firms during the 1960s and 1970s with both promotion (the carrot) and discipline (the stick). Against this benchmark, Latin American industrial performance has fallen short because of varying shortcomings. Under ISI, Latin America was marked by plenty of promotion, but too little discipline. In the 1990s, Latin America has considerable discipline (provided through competitive markets and open trade), but too little promotion.

Consider Pakistan, which is not too dissimilar to Bangladesh in its economic circumstances. Pakistan exports a large quantity of bedsheets, but few hats. Since these commodities are fairly standardized and labor-intensive, it is difficult to believe that the resource endowments and cost structures of the two countries predispose them in any predictable way to specialize in one but not the other. More likely, existing patterns of specialization are the consequence of historical accidents and serendipitous choices by entrepreneurs.

The case of hats and bedsheets constitutes not the exception but the rule… In fact, only six products are in the top 25 for both countries.

We repeat the exercise in Tables 2 and 3 for two other pairs of countries in different parts of the world and at different levels of incomes: Honduras versus the Dominican Republic and Taiwan versus South Korea. Once again, these are pairs of countries that would be expected to have fairly similar patterns of specializations. At a sufficiently aggregated level, that is certainly true. Honduras and the Dominican Republic concentrate on garment exports, while Taiwan and Korea focus on computer products. But at the level of individual products, there are again striking differences.

The range of products countries end up becoming good at producing is quite narrow. Exports are typically highly concentrated, even in rich and relatively diversified countries.

The top 4 export commodities account for around 40 percent of South Korea’s and Honduras’ total exports to the U.S. market, and 30 percent of the other countries’. The cumulative share of the top 25 export items ranges from a high of 83 percent for Honduras to a low of 43 percent for Germany, with the figure for all others lying above 60 percent (save for Taiwan).

The message we take from these numbers is twofold. First, for all economies except possibly the most sophisticated, industrial success entails concentration in a relatively narrow range of high-productivity activities. Second, the specific product lines that eventually prove to be hits are typically highly uncertain and unpredictable.

There is really no such thing as off-the-shelf technology.

The Hyundai story is broadly indicative of Korea’s experience with regard to technology imports: “It is a striking fact that formal purchase of technology in complete packages through such means as turnkey plant contracts and licensing, plus their functional equivalent—direct foreign investment—accounts for only a modest share of the technology that has been mastered in Korea…” Similarly, in Taiwan, “activities such as imitating, copying, or limited improvement of the existing foreign product (i.e., various reverse engineering tactics), were the major sources of acquiring foreign technologies”.

Even in technologically lagging countries, significant amount tinkering normally takes place in order to adapt imported techniques. In Argentina, “many of the newly erected firms found themselves needing to gradually develop ‘in house’ technological capabilities in areas such as product design, production engineering, industrial planning, and organization where ‘from the shelf’ technology was not particularly well suited, given the highly idiosyncratic nature of the local production structure”

Consider the garment industry in Bangladesh, which is a well-studied case of industrial take-off (Rhee 1990). At the origin of the Bangladeshi garment “miracle” lies a largely serendipitous investment made by a local entrepreneur in a joint venture with Dawoo of Korea… But once the venture, called Desh, proved successful, imitation was very rapid: “…115 of the initial corps of 130 workers trained by Daewoo in Korea left Desh at different times following the end of the Desh-Daewoo agreement to set up their own, often competing, garment exporting firms” (Rhee 1990, 341). [See the story of the Mohammadi company in the paper.] As a result, the industry grew from a handful of factories in 1979 to more than 700 exporters by 1985 (ibid). The spread of know-how from the workers initially trained by Desh was critical—much more so than the availability of capital and other resources. Once it became common knowledge that Bangladesh had a profitable market niche in garments, the industry took off like a mushroom…  The spillovers to other types of manufacturing seem to have been very limited, and Bangladeshi exports remain extremely undiversified. Or consider the Colombian cut flower industry, the story of which is told in Rhee and Belot (1990). How did Colombia get from nowhere to being the largest supplier of cut flowers to the U.S.? Once again, it was a single entrepreneur that was responsible for it all. Thomas Kehler, an American businessman in Colombia, happened to be looking for a business opportunity. He and some partners did a feasibility study.. David Cheever, one of the founders of Floramerica, left the company after two years to become a consultant to the growing Colombian flower industry. By assisting in the start-up of many new flower companies, he has been a driving force in diffusing know-how accumulated in Floramerica throughout the country. Second, two salesmen at Floramerica’s Miami sales office left the company ….” By 1990, there were about 250 flower export firms in Colombia, the vast majority of which were fully Colombian-owned.  

Related Books and Articles

• “Atlas of Economic Complexity: Mapping Paths to Prosperity” by Hausmann, Hidalgo et al
• Observatory of Economic Complexity (this is a free online website displaying data related to economic complexity)
• “The Product Space Conditions the Development of Nations” by many
• “Dynamic of Economic Complexity over a 42 Year Period” by Cesar Hidalgo
• “Discovering Southern and East Africa’s Industrial Opportunities” by Cesar Hidalgo
• “Shooting High or Low: Do Countries Benefit from Entering Unrelated Activities” by many
• “Linking Economic Complexity, Institutions and Income Inequality” by Hartmann, Hidalgo et al