A Tale of Two Empires in the Developing World

1

From Heaven to Hell:

A Tale of Two Empires in the Developing World

Alice H. Amsden

Chapter 6

Divine Discipline

Auto assemblers in Brazil had to meet an extremely ambitious domestic-content schedule to be eligible for the full range of financial subsidies. Each year their vehicles had to contain an increased percentage of domestically purchased components. By July 1, 1960, trucks and utility vehicles were to contain 90 percent domestic content, and jeeps and cars, 95 percent.

Helen Shapiro, Engines of Growth, 1994.

The First American Empire wasn’t exactly beaming from the sidelines, but it allowed developing countries to substitute domestic production for imports, not infrequently imports from the United States. Import substitution was one of the most learning-intensive activities. It required picking winners, raising productivity, and out-smarting 300 pound line-back incumbents on the market. Without excellence in project execution and production engineering, scale economies were hard to achieve and brand name recognition was beyond hope. Costly subsidies could not be shifted to other products in order to start the next generation of import substitutes. Most important, without productivity gains, exports remained in the closet.

How did such a problematic process work?

Import substitution did not involve leaping into the technological unknown (most mid-tech industries had mature technologies) but rather, jumping into the economic abyss. The more complex an industry, the more market power incumbents had, and the more difficult the entry and export of newcomers became. When South Korea was building its first petrochemical plant, it was striving to reach international scale but its information was imperfect---world scale increased in the middle of its construction phase. When construction was over, and Korea tried to export, the big incumbents began dumping, driving prices down below average cost. This created a loss-making time interval that required government support. When Korea’s Hyundai shipyard finished building its first ships, with state-built infrastructure right to the docks, it missed its deadline and world prices fell. Hyundai was stuck with a large inventory of unsold tankers. It therefore established the Hyundai Merchant Marine Company and bought its unsold ships from itself. The Korean government then decreed that all crude oil delivered to Korea had to be carried in Korean-made vessels.

Government intervention in import substitution was pervasive, not least of all in terms of rigging prices. The World Bank, in its East Asian Miracle Report (1993), confessed to the mountain of price-“distorting” policies in East Asia. The most worn policies included:

targeting and subsidizing credit to selected industries, keeping deposit rates low and maintaining ceilings on borrowing rates to increase profits and retained earnings, protecting domestic import substitutes, subsidizing declining industries (?), establishing and financially supporting government banks, making public investments in applied research, establishing firm- and industry-specific export targets, developing export marketing institutions, and sharing information widely between public and private sectors.

Governments also bailed out infants in trouble (like Korea’s petrochemical plant) and set performance standards (such as local content for Brazilian automobiles), which made price interventions workable. Import substitution involved an economy’s total capabilities, but it was bitterly criticized because of “rent seeking.” True enough, rent seeking was a reality, although impossible to measure. But rent seeking didn’t run rampant, particularly in countries with manufacturing experience. New institutions kept it under control and growth rates were phenomenal.

Import Substitution

The “rent-seekers” favorite story about the horrors of ISI was the Chilean automobile industry.

Chile was a country with less than 10 million people in the 1960s, whereas the manufacture (as distinct from assembly) of autos had huge scale economies---at the time, more than 1,400,000 million cars rolled off Ford Motor Company’s assembly lines in a single year. Thanks to government incentives (mostly tax-related), Chile attracted around 20 automobile assemblers, each operating on an infinitesimally small scale. The government considered restricting entry to only one or two firms, but didn’t know what criterion to use. There was no weeding-out process by the market, which might have charitably left only one or two players to compete. Soon Chile’s automobile industry collapsed altogether.

Chile obviously had not picked a winner, or hadn’t stipulated rules to create a winner, although an industry as diverse as automobiles gave latecomers much leeway to win or lose. China, when it had about as many cars on the road as Chile, decided to license only the assembly of luxury cars (Peugeot, Toyota) for business executives and top civil servants, which is where demand was greatest. These cars were then heavily taxed. Some haughty civil servants in the Ministry of Machinery derided the idea of a “people’s car” for years to come. But it came practically overnight, outside the Ministry’s window, and the Ministry lost power to regulate the building of “people’s cars” at the provincial level. Taiwan, with a population the size of Chile’s in thee 1960s, licensed a few auto assemblers, including nationally owned firms. These assemblers never did too well but they were encouraged to import substitute certain parts. These large-scale parts reached a high level of excellence and were soon exported. Taiwan became a major exporter of automobile parts and components worldwide.

Brazil was the inventor of local content rules, but the contentious issue was over final assemblers. After Brazilian-owned assemblers failed, all remaining assemblers were foreign, and for years, all of them refused to export. The Mexican government wanted to merge two of its national assemblers. But GM and Ford objected. They complained to the American Embassy and the Mexican government dropped the idea of a merger, and exports stalled until the 1990s.

With an eye on Japan in the 1950s, Korea targeted automobiles as a leading sector. The government allowed only two companies, a wholly owned subsidiary of the Hyundai group and a joint venture between GM (now in financial trouble) and Daewoo, also a member of a group (now bankrupt). GM was present only because the government couldn’t say no to it for political reasons. After the introduction of a new model, the government allowed prices to exceed world prices, to cover costs, but then prices had to rapidly fall. For 25 years, no foreign cars were to be seen on Korean roads and no Korean cars were to be seen on foreign roads. Then Hyundai Motors exported to Saudi Arabia. Then it exported to the US and bombed; the car was rated as one of the lowest, possibly because it was from a developing country (after all, it had a Mitsubishi engine, and Mitsubishi rated high). Then Hyundai exported to the US again and became one of the fastest growing small cars on the world market.

Developing countries with low purchasing power, many in Sub-Saharan Africa, did a nice job picking winners. “Winners” tended to be labor intensive, and aimed at a broad, popular market. Poor countries achieved a light dusting of manufacturing industry in the 1960s and 1970s. They produced beer, flour and other foodstuffs, building materials like bricks, steel tubes and cement, and miscellaneous manufactures like matches, metal boxes, and the assembly of consumer durables, including cars. The indentors responsible for imports (Ford’s agent, for instance) were also responsible for assembling cars. What brought these industries to their knees was inadequate domestic demand, due to declining terms of trade in agriculture, or home-made products like beer. Exports were also tough to generate since poor countries tended to have the same set of industries. After the debt crises of the early 1980s and a steep fall in income, most industries collapsed. Workers returned to their traditional jobs on the docks and in the mines, with a knowledge of few transferable technologies.The complexity of import substitution, and the urgency of drawing down balance of payments deficits, didn’t mean that rules for picking winners were a mishmash. Starting with Japan, rules were simple and straightforward. Two rules applied to picking industries and firms: the growth rate of output, and the growth rate of productivity (both worldwide). These rules would have been easy for the Chilean automobile industry to follow, limiting entrants to two, as in Korea. The problem was not import substitution but badly executed policy.

The flagship of every “developmental state” was a development bank. This bank was unique; it is hard to find an equivalent in past history. Not only did it lend money at below market interest rates. It attached performance standards to the loans it made, it monitored them, and gave advice technically and economically to borrowers. Part of the advice was picking winners, and many of the banks in the 12 countries with prewar manufacturing experience converged on the same broadly defined industries. Great minds---of top bureaucrats---worked alike.

Development Banking

Development banks took a shotgun rather than rifle approach to kick-start industrialization, lending to many would-be winners. The only obvious investment criterion that did not figure explicitly in credit allocation was 'comparative advantage'---industries with static comparative advantage already tended to exist while industries with dynamic comparative advantage could not be identified as such ex ante. Import substitution was the dominant form of investment, but development banks also funded export activity per se if they could keep afloat in world markets.[i]

The criteria for Brazil's development banking emerged out of historical circumstance. According to BNDES, the development bank: "The second administration of President Getulio Vargas, begun in 1950, inherited from the previous administration a nation anxious for change. The favorable balance of trade was being weakened by the importation of heavy industrial products and equipment, the rise in post-war consumption and international fuel prices. Given such a dilemma, the nationalistic middle class emphatically called for funds for development of basic industries." None of this precluded the goal of raising exports: "Between 1958 and 1967, fully one half of BNDES' funds went to steel making, transforming Brazil, at the first stage, into a self-sufficient steel producer and, later, into a major exporter of steel products." Moreover, the policies of the BNDES changed over time: "Beginning in 1974, with the oil crisis that suddenly hit Brazil's balance of payments hard, the government decided to intensify its import substitution program, as set out in the second National Development Plan." BNDES began to finance "principally two major sectors: capital goods and basic raw materials, consisting of minerals and ores, steel and non-ferrous metal products, chemical and petrochemical products, fertilizers, cement, pulpwood and paper."

Taiwan's heavy industries were targeted as early as 1961-64, during the Third Plan: "Heavy industry holds the key to industrialization as it produces capital goods. We must develop heavy industry so as to support the long-term steady growth of the economy" (Ministry of Economic Affairs). At the same time, exportables such as watches and other electronic products were promoted. After most heavy industries were, in fact, developed (steel, shipbuilding, petrochemicals, machinery), and the second energy crisis occurred (1979), goals changed. In 1982, the Taiwan government began to promote "strategic industries" (machinery, automobile parts, electrical machinery, information and electronics) based on six criteria: large linkage effects; high market potential; high technology intensity; high value-added; low energy intensity; and low pollution.

Selection of promoted industries in Thailand, as stated in the 1950s, also had multiple criteria. First, they had to save a lot of foreign exchange. Second, they had to have strong linkages to other industries. Third, they had to utilize domestic raw materials. Yet another reason for promotion, according to the Ministry of Industry, was to gain technological knowledge: "Hopefully, the industries to be promoted such as automobiles, chemicals, shipbuilding, and so forth will transfer technological knowledge from developed countries."

India's development plans listed objectives that were broader and more political than those of other countries, which is maybe why it grew more slowly: 1. A faster expansion of basic industry than light industry, small firms than large firms, and the public sector than the private sector; 2. Protection and promotion of small industries; 3. Reduction in disparities in regional location of industry; and 4. prevention of economic power in private hands.

According to Turkey's Second Five-Year Plan (1968-1972), it was important to promote manufacturing because it was the sector that would 'pull' the economy ahead in the future. Industry priorities were chemicals, commercial fertilizers, iron, steel and metallurgy, paper, petroleum, cement and vehicle tyres. "Intensified investments in these sectors will create to a large extent import substitution effects and lay the necessary foundations for industrialization in the long-run". At the same time Turkey's Plan set targets for a large increase in exports, and the textile industry was heavily promoted.

In the case of Mexico's development bank, the principles that guided it in the early 1960s were to assist those industrial enterprises whose production could improve the balance of payments, achieve a better industrial integration, induce savings or increase the level of employment. By the late 1980s, after a debt crisis, the principles were to "promote the restructuring, modernization and financial rehabilitation of companies as a way of achieving better efficiency and production, which is necessary in order to increase exports and substitute for imports permanently, thereby reaching a level of international competitiveness".

According to the 1969 Annual Report of the Korea Development Bank (KDB), top priority in lending was given to export industries and industries designated in a Bank Act that "improved the industrial structure and balance of payments". These included "import substitute industries." Import substitution and export promotion were not seen as antagonistic; both involved large, long-term capital investments. By 1979, the end of Korea's heavy industry drive, the following factors were emphasized in financial commitments: the economic benefits to the nation; the technical and financial feasibility of a project; its profitability; and the quality of an applicant's management.

The 'hot' industries of development banking---industries that received the largest and second largest shares of credit in various decades, are basic metals (mostly iron and steel), chemicals (primarily petrochemicals), machinery (electrical and non-electrical), transportation equipment (ships, automobiles and automobile parts) and textiles. These are the privileged borrowers.[ii] These industries are broadly defined and comprise a variety of products. While the sub-branches of such industries varied across countries, all of dozen countries with manufacturing experience (data exist for 7 countries) targeted more or less the same basic industries for postwar growth. They saved foreign exchange, they had lots of linkages (some with export potential), they moved technology forward without venturing into the unknown, and they had large, global markets.