SYP

3. The State and Taiwan's Economic Development

Alice H. Amsden

Two features of Taiwan's post-World War II history are striking. First, it is one of the few nonsocialist economies since Japan to rise from the grossest poverty and to enter the world of the "developed."1 Second, the state in Taiwan has played a leading role in the process of capital accumulation. It has positioned itself to prevail on key economic parameters such as the size of the surplus extracted from agriculture and the rate of profit in in- dustry. To understand Taiwan's economic growth, therefore, it is neces- sary to understand its potent state.

The challenge of understanding the role of the state in Taiwan's eco- nomic development is increased by the fact that the state's initial aims were so clearly military and geopolitical rather than economic. When Taiwan was occupied by the vanquished Nationalist government in 1949, the Guomindang was obsessed with one objective: military buildup in order to retake the Mainland. As Edwin Winckler bluntly put it, "The Jiang Jie-Shi forces, if they had had their own way, wouldn't have spent one penny on economic development."2 Given that militarism and economic develop- ment must to some extent operate at cross-purposes, competing for the same scarce resources, Taiwan's success must seem somewhat paradoxi- cal.

If the role of the state is critical to economic development, why should an economy under the heel of the military end up with a "good claim to be ranked as the most successful of the developing countries"?3 One of the obvious factors mitigating the negative consequences of militarism in the Taiwan case was American aid, which diminished the extent of resource competition. I shall try to show, however, that this was not the only, or necessarily even the most important, factor in allowing economic growth to arise out of militarism. Rather, I shall argue that the reality of economic development itself both seduced the military away from its initial orienta- tion and changed its position within the state apparatus, which then freed up the process of capital accumulation still further.

The State and Taiwan's Economic Development 79

Although the Guomindang state in Taiwan's "economic miracle" is our central focus, it is necessary to take into account other factors that favored the island's economic development after World War II. One of the most important of these is the legacy of the Japanese colonial period. We shall begin, then, with a discussion of the colonial period and from there look at the Guomindang state itself, its role in agriculture and industry, and its gradual transformation from a state in which the preeminence of military aims was overwhelming to one in which military aims came to coexist with an evermore absorbing interest in economic growth.

The Colonial Heritage4

It is a misconception that the Taiwan miracle commenced with the export of labor-intensive manufactures and a reduction of government manage- ment of trade and monetary matters in the decade of the 1960s. Taiwan already enjoyed a relatively fast rising real gross domestic product (GDP) in the 1950s (Table 3.1). Agriculture was then the dominant sector, and the economic regime in industry, as in many other underdeveloped countries, was one of protection of infant industries. Growth was also rapid during the years of Japanese domination (1895-1945). Excluding the war years of 1941-45, the per capita income of the agricultural sector almost doubled in half a century. This is a rather impressive figure given that the population rose by approximately 43%.5

The economy that the Japanese fashioned in Taiwan was achieved by means of deliberate planning and government ownership of major re- sources (in partnership with private Japanese capitalists). The dominance of the Japanese colonial administration in Taiwan's economy mirrored the dominant role of the Meiji government in Japan proper, which distin- guished it in important respects from the colonial offices of England and France.6 The Jiang Jie-Shi forces benefited enormously from their inheri- tance of Japanese state monopolies, and the whole interventionist ap- proach taken by the Japanese to the development of an occupied territory was not lost to the Guomindang.

From the start, Taiwan was regarded as an agricultural appendage to be developed as a complement to Japan. A two-crop economy (sugar and rice) was encouraged much in the classical imperial pattern. But one aspect that sets Taiwan's colonial experience apart from the rest is that primary pro- duction was not confined to a foreign enclave with limited spillover on subsistence agriculture. Many farmers with access to arable land produced rice for market to meet the ever-escalating needs of Japanese consumers. Although sugarcane is frequently cultivated on large plantations in some Third World countries, in Taiwan it was grown by small owner-operators and tenants as well as on large land tracts owned by Japanese sugar man- ufacturers. Thus, agriculture in Taiwan was quickly and generally com- mercialized.7

Table 3.1. Indicators of Taiwan's Economic Performance-Average Annual Growth Rates

Consumer Wholesale Gross Per prices in prices in

Popu- national capita Agricultural Industrial Transportation & Taiwan Taiwan Period lation product GNP production production communications area area Exports Imports

1953-62 1963-72 1973-80

3.5 2.9 1.9

7.5 10.8 8.7

4.1 8.1 6.6

4.8 4.0 2.4

11.7 18.6 11.9

10.4 15.7 19.0

8.7 2.9 12.4

7.6 1.8 12.0

19.5 29.9 25.0

17.0 23.5 27.7

Note: Growth rates in the third through seventh columns are in real terms. Source: Taiwan Statistical Data Book (Taibei: Council for Economic Planning and Development, 1981).

The State and Taiwan's Economic Development 81

The Japanese remodeled the archaic "three-tier" tenancy system, expro- priating the "great" landlords and making the second tier of tenant land- lords the legal owners of the land and directly responsible for taxes. A flat tax on land replaced a proportional tax on output, giving landlords incen- tive to squeeze more production out of the tenants below them. Ground rents remained very high (commonly 50% or more), but the new structure allowed greater scope for the assimilation of new farming practices. The peasants who produced, on rented land, 65-70% of Taiwan's total crop8

applied new seed strains and other technological advances proffered by state-supported research agencies. Thus, the colonial state set in motion the application of science to farming, which characterizes the rural econ- omy of Taiwan today.9

An elaborate network of agricultural associations, under the aegis of the government and rich landlords, provided peasants with extension educa- tion, the cooperative purchase of fertilizers, warehousing, and other ser- vices. When persuasion failed, the police were employed to force modern techniques onto rural communities that resisted change.10 The experience that small tenants gained in experimenting with new seed strains and their familiarization with scientific farming would also prove to be of immense usefulness to the later land reform efforts of the Chinese Nationalists. The extensive network of agricultural associations that the Japanese introduced was created to facilitate police surveillance and control over the local pop- ulation. Today, these associations persist and are an important element in the government's management of agriculture.

In the 1930s, Japan reshaped its policy of transforming Taiwan into a source of food supply for the home market. The shift in policy can be understood only in the context of Japan's increasing militarism and expan- sionism in the Pacific. Belatedly and frantically, Japan sought to refashion Taiwan as an industrial adjunct to its own war preparations and ambitions in Southeast Asia and South China.

From a few industries with strong locational advantages before 1930 (e.g., sugar and cement), industry in Taiwan expanded in the 1930s to include the beginnings of chemical and metallurgical sectors, and as World War II cut off the flow of duty-free goods, some import substitution began. Japa- nese hopes of building Taiwan into an industrial bridgehead to Southeast Asia and South China, however, never materialized.11 The policy was in effect for too short a time before it was halted by World War II. Although the last-minute efforts to construct transport and harbor facilities suited to military and industrial needs proved highly beneficial in postwar years, many projects remained on the drawing board when war erupted.

Thus, economic growth in Taiwan under Japanese rule went about as far as it could go, given the internal contradictions of imperialism. Growth included a rise in per capita income; indeed, the welfare of Taiwanese peasants in the first half of the twentieth century may have exceeded that of Japanese peasants - according to such welfare indices as type of wearing

82 Alice H. Amsden

apparel, housing, local bank deposits, and the like.12 The most enduring legacy of the Japanese occupation, however, was less the betterment of living standards than a relatively well educated population and the build- ing of a foundation for subsequent development. Whereas much of the gain in per capita income was lost as a consequence of war and an influx of Mainlanders following the Communist victory in China (and was not regained until the 1950s and 1960s), a relatively high level of literacy and the economic structure implanted by the Japanese survived. The major les- son of the Japanese interlude, however, was that to exploit the economic potential of Taiwan required much more than a reliance on inexorable mar- ket forces. It required deliberate state policies, something that the invading force from the Mainland seemed very unlikely to be able to provide.

The Nature of the Guomindang State

The state that took over Taiwan from the Japanese was a highly militaristic bureaucracy dominated by a single leader, Jiang Jie-Shi. On this there is much agreement. But the internal structure of the state, the relative power of different groups within it, and even the extent to which it was a real bureaucracy in the positive Weberian sense are open to debate. The ab- sence of descriptive material that would allow one to characterize the Tai- wan state with confidence is well summed up by Winckler as follows:

The basic nature of the political system remains undefined. If the island has been a military and political client of the United States, we know little about the inter- national and inter-bureaucratic workings of the relationship. . . . If the island has been a dictatorship ruled by Chiang Kai-Shek [Jiang Jie-Shi] and his son . . . we do not have political biographies of either for their Taiwanese periods. . . . If the island has been a police state dominated by military interests, we lack institutional descriptions and political history of its internal and external security agencies. . . . If the island has been successful in managing its economic development, we do not have a political account of the persons, agencies, and interests involved. If the island has been ruled by the Kuomintang [Guomindang], we know little more about the party's politics and administration than its own glossy brochures tell us. . . . Finally, we need to know the relative weights of setting, personality, security, eco- nomics, and ideology, and how these elements fit together.13

Despite the accuracy of Winckler's laments, those who would try to ex- amine the state's role have, of course, no choice but to sift through the available evidence and make the inferences that seem the most reasonable. The problem we are presently interested in exploring is the relationship between militarism and economic development. Hence, we are interested in the Taiwan state's military and technocratic dimensions. Concerning the latter, one view holds that Taiwan's new rulers were a competent group. Another view, rooted in an earlier historical period, is far less flattering. A quote from Simon Kuznets provides a summary of the posi- tive view:

The State and Taiwan's Economic Development 83

The governing group were largely newcomers to the island and had no substantial interest roots, no clear affiliation with any of the various Taiwanese interest groups. . . . They could, therefore, act as independent arbitrators attempting to achieve a long-term consensus and calling for sacrifices by some groups for the benefit of all. If we assume that there was substantial agreement among the decision-making groups and the larger groups that were being served, both islanders and mainland- ers; that the historical and cultural community among all groups was an adequate basis for consensus; and that the decision-making groups had enough experience and hu- man capital for generating the required decisions promptly and efficiently, we can argue that this combination of the islanders and the mainlanders was favorable for a poor, developing country like Taiwan [italics added].14

As Barbara Tuchman relates, a more sour impression of the "experience and human capital" of the Guomindang (KMT) was held by Joe Stilwell, U.S. general of the Pacific theater, who had the displeasure of dealing with Jiang Jie-Shi before and during World War II:

By keeping rivals off balance through a technique of "fear and favor/' in Stil- weirs phrase, [Jiang Jie-Shi] appeared strong and indispensable but he did not know how to make a government. Though long on experience, his mind was nar- row and his education limited. His most serious handicap was the lack of compe- tent government servants. He never allowed a really able man to reach an impor- tant post lest he become too strong. Because he made loyalty rather than ability the criterion of service, he was surrounded by mediocrities. His brother-in-law, . . . who as vice-president of the Executive Yuan headed the civil government and usu- ally served as Finance Minister, was described by [the] representative of the Bank of England in China as having "the mentality of a child of 12. If I were to record his conversations with me about banking and play it back, nobody would ever take [J]iang's government seriously again."15

History, however, does not necessarily repeat itself, and it has been con- tended that old dogs do learn new tricks:

[J]iang's authority as a leader derived not only from his skill at political maneuver, but also from his selection of able officials for key positions. He was far more effec- tive in this respect in Taiwan than he had been on the mainland, where a glaring weakness of the Nationalist government was his tendency to value loyalty above ability. Loyalty was important in Taiwan, too. . . . His military and civilian officials in Taiwan, therefore, had to be loyal, but many were also able.16

The U.S. Aid Mission to Taiwan, moreover, needed competent technocrats with whom to work, and the mission used its clout to shield the technoc- racy and to help it compete politically.17

Thus, we can speak of the existence fairly early on of an economic tech- nocracy in Taiwan - albeit an embryonic one - and not make fools of our- selves, for the technocrats need not be equated with the rascals they once were in prewar Mainland China.

Nevertheless, the development-oriented technocracy was overshad-

84 Alice H. Amsden

owed in the early postwar period by the military. Most of the Nationalist cliques of the prewar period had disintegrated with the Communist vic- tory, save one: the Whampoa military cadets. Studies of the Taiwan state in the early and late 1960s discuss the power plays centered around the then security administrator (Jiang Jie-Shi's son). For the purposes of un- derstanding the policy orientation of the Taiwan state in the immediate postwar period, however, it is sufficient to note a consensus in the litera- ture about Jiang Jie-Shi's unchallenged political supremacy and the fact that he, "more than any other person . . . was responsible for asserting and perpetuating the concept of mainland recovery. . . . He insisted on tighter authoritarian controls and a greater diversion of resources to mili- tary and security purposes."1 8

On the basis of this admittedly sketchy picture of the Guomindang state apparatus, the possibilities of the state serving as an effective instrument of economic development would seem bleak. Yet from the beginning of its reign, the Guomindang bureaucracy embarked on a set of policies that, though they appear to have been chosen for political as much as, or more than, for economic reasons, were crucial to the island's eventual economic growth. Nowhere is this more evident than in agriculture.

Agriculture

When the Guomindang regime arrived on Taiwan, agriculture was by far the most important sector economically. Its share of GDP was twice that of industry, and it accounted for 90% of exports. It was, in addition, impor- tant politically. The potential threat of an impoverished peasantry had been driven home to the Nationalists on the Mainland, and they were concerned with restructuring agriculture accordingly. From the beginning, Guomin- dang policy toward agriculture had two races. On the one hand, state ac- tion was the key to increasing agricultural output. On the other hand, ag- riculture was consistently squeezed to provide the surplus necessary to finance the growth of other sectors. The cornerstone of both sides of state policy was the land reform initiated in 1949 and completed in 1953.

Agriculture was reformed in three stages. First, farm rent was limited to a maximum of 37.5% of the total main crop yield. Second, public land for- merly owned by Japanese nationals was distributed on easy terms, with preference given to the tenant claimants. Third, landlords were obliged to divest themselves of their holdings above a minimal size and to sell out to their tenants under the Land-to-the-Tiller Act.19 This end to landlordism and the creation of a class of small holders was the inspiration of Dr. Sun Yat-Sen. The Guomindang's Land-to-the-Tiller Program amounted to sheer rhetoric in China during the 1930s and 1940s because would-be expropri- ated landlords were stalwarts of the Nationalists. In Taiwan, by contrast, the Mainlander government was under no obligation to the rural Tai-

The State and Taiwan's Economic Development 85

wanese elite. Although both were of Chinese origin, they were as different ethnically and socially as the French and the Americans. Landlords were given land bonds in kind and stocks in public enterprise in exchange for the compulsory divestiture of their holdings. Some landlords profited from their stock ownership and became successful industrialists. Others went into bankruptcy.20 The landlord class, however, sank into social oblivion, as the great landlord class had done half a century earlier.

Thus, almost overnight the countryside in Taiwan ceased to be op- pressed by a small class of large landlords and became characterized by a large number of owner-operators with extremely small holdings. By 1973, almost 80% of the agricultural population consisted of owner-cultivators and another tenth of part owners.21 Only 6% of farm income accrued to landlords and money lenders.22 This undoubtedly underscores the fact that income distribution (by household) in Taiwan is far less inequitable than in most other Third World countries and is more like the pattern in ad- vanced capitalist countries, which is not to say, however, that income dis- tribution is equitable.23

The years 1953-68 witnessed annual growth rates in agricultural output that were impressive by any standard. Equally impressive was the spill- over effect on industry, for however tight the squeeze on agriculture under Japanese rule, it was even tighter under the Jiang Jie-Shi administration. Whereas net real capital outflow from agriculture had increased at a rate of 3.8% annually between 1911 and 1940, it rose on average by 10% annually between 1951 and I960.24 Fast growth and a transfer of agricultural re- sources to the towns, however, were neither the outcome of free market forces nor the automatic result of purely technical phenomena - the green revolution. Rather, they reflected the structure of ownership in the coun- tryside and state management of almost every conceivable economic activ- ity.

It is well known that in developing countries there have been substantial gains in income among the few (i.e., the bigger farmers) when the new technology associated with the green revolution has been introduced. By contrast, the green revolution in Taiwan has transformed the life of almost every peasant. Furthermore, such an extensive application of science ap- pears to hinge on government control over capital accumulation. The state distributes resources equally among all peasants, as the market mechanism might not do. Hence, there have been large gains among the many. A small class of big landowners has not yet resurfaced (nor, consequently, has a potentially cohesive source of opposition to the state). It is, then, a defining characteristic of Taiwan's agriculture that a multiplicity of small peasant proprietors exists in conformity with the bourgeois model of indi- vidualistic family farming, whereas directing this drama is a highly cen- tralized government bureaucracy.

This point was recognized by two anthropologists in a study of rice farm- ing in three Taiwan villages:

86 Alice H. Amsden

In this small island with its geographically mobile population, the arm of the state reaches down to virtually every farmer - outside the mountainous regions. This is a basic social and administrative characteristic of agriculture in Taiwan that has been long in the making.

As a result of [land tenure reforms], the rural landlord social class in the villages disappeared. The power of the state could reach then direct to every villager. The tenants of the past pay land tax now to the state and water fees to the government directly.25

In 1965, government agencies or related credit institutions supplied 65% of all agricultural loans. Before land reform, private moneylenders accounted for 82% of credit.26 With respect to such activities as agricultural education and marketing, the government exerts its control through the elaborate network of agricultural associations laid down by the Japanese.27

The state monopoly on fertilizers was perhaps the most important ele- ment both in stimulating production and in extracting the surplus from agriculture. The positive effects of the fertilizer monopoly on the peasantry are well summarized by Falcon:

It permitted all farmers to obtain the key modern input. It provided a source of credit that was an alternative to rural moneylenders. And it reduced price risks to farmers. (Widespread emphasis on risk-reduction is evident in Taiwan's agricul- tural policies and seems to be one of its important lessons.)28

At the same time the fertilizer monopoly was the key to extracting surplus from agriculture. Fertilizer was bartered for rice, and the barter ratio was highly unfavorable to farmers. The price that Taiwanese farmers paid for 100 kilograms of ammonium sulfate in 1964-65 was higher by almost 40% than the price that Japanese, Dutch, Belgian, American, or Indian farmers paid.29

Other mechanisms were also used to transfer real net surplus out of agriculture: land taxes, compulsory rice purchases by the government, loan repayments, and repayment for land resold to tenants under the Land-to- the-Tiller Program. (See Table 3.3 for a comparison of the tax burden of farm and nonfarm families.) All such collections were made in kind. All amounted to "hidden rice taxes/' because the government's purchase prices were considerably lower than implicit market prices. The government's gains through rice collection were enormous. The hidden rice tax exceeded total income tax revenue every year before 1963.30

Three classic problems typically prevent small peasant production from becoming a solid basis for capital accumulation. First, peasant production is generally unproductive because it is unscientific. Second (according to an argument popularized by Stalin in defense of collectivization), peasant production frustrates the extraction of a surplus by the state because, at a low level of per capita income, farmers are said to consume their incremen- tal output rather than market it (i.e., they may be more resistant to exploi- tation). In land-scarce Taiwan, the Guomindang state managed to over-

The State and Taiwan's Economic Development 87

come both of these problems. State provision of educational and scientific infrastructure helped to resolve the first problem. The second problem re- solved itself as scientific agriculture raised per capita income and forced the peasantry to part with its crop in order to obtain fertilizer and socially necessary items of consumption.

The third problem historically encountered in peasant production is de- scribed by Hla Myint. When peasants become full-time producers for the market,

[they] cease to be self-financing and have to borrow from the chief source available to them - the money-lenders who charge them high rates of interest. With their ignorance of the rapidly changing market conditions, they tend to get heavily into debt, and where land is alienable, they lose their land in default of loans and get reduced to the status of tenants.31

Economic history in Taiwan, by contrast, saw the state effectively preserve an agrarian structure of small peasant holdings by stabilizing prices and by making credit generally available (i.e., simulating a perfect credit market by having no market at all). The Jiang Jie-Shi government also dispensed with foreign middlemen, who typically exercise monopoly power in rural areas of other economies, by itself buying cash crops cheaply from the peasantry and selling them at high prices.32

Thus, a self-exploitative peasantry, working long hours to maximize pro- duction per hectare, and a superexploitative state, ticking along effectively to exact the fruits of the peasantry's labor, operated hand in hand in Tai- wan to great advantage until the late 1960s.

The only question that remains is, to whose advantage in particular? For Taiwan is not a classless entity, and the state acted in the interests of an elite when it squeezed the countryside. Unfortunately, whereas a volumi- nous amount of statistical information is available about Taiwan, very little class analysis has been published. Clearly, however, the historical roots of the Guomindang's etatisme and its class affiliations are traceable not only to Japanese colonialism, but also to events on the Mainland. We may hy- pothesize that the system of "bureaucratic capitalism" of late imperial China, with its total interpenetration of public and private interests, was trans- planted into Taiwan, along with the Mainlanders. Although historical con- ditions were unpropitious for economic development under bureaucratic capitalism in China, they were favorable in Taiwan. The 1953 land reform and subsequent agricultural development breathed new life into the Guomindang apparatus, and the bureaucratic capitalism of the Guomin- dang regime sustained the life of the reform and small-scale farming.

In summary, agriculture in Taiwan gave industrial capital a labor force, a surplus, and foreign exchange. Even during the immediate postwar years of economic chaos and a world record rate of population growth, agricul- ture managed to produce a food supply sufficient to meet minimum do- mestic consumption requirements as well as a residual for export.33 Good

88 Alice H. Amsden

rice harvests have been a major factor behind Taiwan's stunning price (and real wage) stability. The foreign exchange saved as a result of high produc- tivity in agriculture has been equally important.34 Agriculture also man- aged to provide an important source of demand for Taiwan's industrial output, particularly chemicals and tools, and a mass market for consump- tion goods. The agrarian structure provided a degree of political stability sufficient to draw the most timid of foreign firms to the island. Agriculture has even been sufficiently productive to set a floor on industrial wages. Factory women who returned home to the farm during the sharp depres- sion of 1974-75 subsequently refused to return to wage employment at prevailing rates.35 A labor shortage symbolizes Taiwan's introduction to the problems of capitalist development rather than underdevelopment, and it is to industrialization that attention is now turned.

Industrialization

Taiwan's industrialization has often been falsely characterized as exhibit- ing the efficacy of a "laissez faire" strategy. It has been argued that Tai- wan's success is due to the fact that it, unlike most Third World countries, resisted the temptations of infant industry protection. A study published by the Organization for Economic Cooperation and Development (OECD) in 1970, comparing industrialization in Brazil, Mexico, Argentina, India, Pakistan, the Philippines, and Taiwan, made a start toward dispelling this illusion.36 The study showed that a regime of import substitution preceded the export of labor-intensive manufactures in Taiwan. Nor was infant in- dustry protection a trivial episode in Taiwan's economic history. The pro- tection afforded to sales on the home market in 1966, the year under ex- amination in the OECD report, far exceeded that which prevailed in Mexico, a country reputed to be highly protectionist.37

In the period from 1956 to 1961, the government introduced a package of reforms to reorient the Taiwan economy toward export-led growth. Monetary and fiscal policies were redesigned, the exchange rate was de- valued and unified, inflation was brought under control, and exports were made highly profitable. These changes have earned the title "liberaliza- tion" and have been responsible for Taiwan's reputation for successful de- velopment with sound formulas. Nevertheless, what is not appreciated is that protection in Taiwan of key import substitutes never appears to have abated.38 Whereas exporters were allowed to import their inputs duty free after "liberalization," critical second-stage import-substitute items con- tinue to be shielded from foreign competition. These include intermediate inputs, consumer durables, and transportation equipment. Although ma- chinery has received little tariff protection, it has not yet needed much. The level of sophistication of Taiwan machinery has been such that the competitive niche it occupies is different from that of American or Euro- pean equipment. Indicative of the spirit of Taiwan's tariff system is the fact

The State and Taiwan's Economic Developme 89

that machinery is technically freely importable - but, in the case of machin- ery that is locally available, only by machinery users and not distributors and except in the case of a handful of countries, which just happen to be those that constitute a genuine threat to local machinery builders, such as Japan and South Korea. Still another indication of the protectionist procliv- ities of the Taiwan government is the fact that most import tariffs are re- dundant, that is, higher than necessary.39 This is the case despite the vig- orous efforts of the liberalization lobby, to be discussed shortly, to wipe the tariff slate clean.

Buried in a dense amount of data in a World Bank study on the virtues of export-led growth is the revelation that governments in Third World countries not only have begun to subsidize exports, but have also contin- ued to protect import substitutes, Taiwan being no exception.40 Thus, av- erage net effective subsidy rates for the manufacturing sector were 38% in Argentina, 10% in Colombia, 7% in Israel, - 1 1 % in Korea, —4% in Sin- gapore, and 10% in Taiwan; such rates in Taiwan not only are relatively high by this reckoning, but also show the traditional pattern of escalation as one moves from lower to higher levels of transformation. Thus, not only in Taiwan, but also in other semiindustrialized countries, higher levels of transformation (read second-generation import substitutes) have far higher average net effective subsidy rates than those just reported. Quantitative import controls have also played an important role in the protection of manufacturing industries.41

Inward-oriented growth in Taiwan was introduced in 1949 partly by de- fault (traditional agricultural exports no longer found protected or prefer- ential markets in Japan and China) and partly by design (it was politically expedient to aid the class of small capitalists that had acquired a portion of the old Japanese facilities). Small enterprises were in serious trouble by 1949 as a result of the loss of the Mainland market and the reappearance of competitive Japanese goods.42 Import, foreign exchange, and licensing controls were introduced by the government to salvage small establish- ments from extinction and to ease the critical balance of payments situa- tion.

Inward-oriented growth in Taiwan's small domestic market, however, soon stalled. Although it conferred high profits to some, it conferred infla- tion, monopoly, excess capacity, a reliance on American donations of hard currency, and corruption to all. It was only after manufacturing had made a fair start, however, that the Taiwan government hesitantly charted a new course in the direction of export-led growth. Two points are worth stress- ing in this regard. First, "liberalization" should in no way be interpreted as a restoration in Taiwan of a "market economy." Government manage- ment of capital accumulation has continued, as evidenced, if by nothing else, by tariff protection. Second, although economists have viewed a re- gime of export-led growth as one that is more in keeping with liberal eco- nomic principles, the Taiwan government has not been guided by any the- oretical orthodoxy to turn a profit. A civil servant writes:

90 Alice H. Amsden

Unorganized production and export often led to excessive production and cut- throat competition in foreign markets, which inevitably cause a sharp decline in price, deterioration in quality, and finally loss of the export market. To combat these shortcomings, the government has encouraged unified and joint marketing of exports in foreign markets through limitation of production by means of export quotas, improvement of quality, and unified quotation of export prices.43

Cartels, in whatever variation, have been encouraged by the government and, at one time or another, have covered most of Taiwan's major exports: textiles, canned mushrooms and asparagus, rubber, steel, paper products, and cement.44 The government has tried to get the marketing of all exports into Taiwanese hands because both bureaucrats and businesspersons alike are sensitive to the inroads in overseas marketing made by large Japanese trading companies; this is in spite of the efficacy of these trading compa- nies and the lackluster track record of those of Taiwan.

At issue is not merely the quibble that the government of Taiwan has intervened far more in the Taiwan economy than liberal economists who champion export-led growth acknowledge. The point, rather, is that the government of Taiwan has never been guided by free-market principles as such; so to attribute Taiwan's success to a commitment to such principles, whether in theory or in practice, is misleading. What has obsessed the Guomindang state since its defeat in China is economic stability. In the words of K. T. Li, one of the chief architects of government policy, "During the 1950s . . . the overriding economic consideration was stability."45 And in the 1970s:

Everything possible is done to maintain price stability, even if it requires being less mindful of the growth rate. Our large governmental sector employs numerous military personnel, civil servants, and teachers, and any price increase affects these persons with particular severity46

And again:

I wish to emphasize . . . that we do not seek unduly rapid growth, but prefer a steady, moderate expansion in directions which will be of greatest benefit to our nation.47

To achieve stability, the government of Taiwan appears to have thought it prudent sometimes to shield the economy as much as possible from market forces and at other times to use them, but never to embrace them as a rule of thumb out of conviction.

Foreign Aid, Foreign Capital, and State Enterprises

Flows of U.S. aid in the 1950s and 1960s and, more recently, flows of for- eign direct investment played a critical role in capital formation in Taiwan. These flows, particularly the initial deluge of U.S. aid, obviously had im- plications for policies and politics in addition to providing capital. The

The State and Taiwan's Economic Development 91

magnitude of the aid cannot be denied. In 1955, when aid reached its peak, it amounted to over half of gross investment, and it was not until 1964 that it fell below 20%.48 Politically, aid was critically important: It kept the state in power by helping to bring inflation down from 3400% in 1949 to 9% in 195349 by means of the arrival of large quanties of both consumer goods and producer goods at a desperate moment.

In terms of long-run economic growth, the impact of aid was minor.50

Most aid went for military purposes and the remainder for infrastructure broadly defined. (From this perspective, aid to Taiwan was a success by U.S. standards, for the goals of providing aid tend to be the buildup of infrastructure and the buttress of political stability.) Thus, the effects of aid may be said to have been felt in perpetuity only in terms of a multiplication of civil engineering projects and know-how and improvements in the ad- ministrative capability of the Taiwan technocracy. The extent to which Tai- wan's turn to export-led growth can be attributed to U.S. pressures is dif- ficult to say: Now, all parties concerned are eager to take credit for it. All that is certain is that freer trade and freer enterprise (including freer foreign enterprise) were preached by the American Aid Mission. In view of the preferences of its major benefactor, the extent to which the Guomindang government persisted in its "etatisme" is all the more impressive. To ap- preciate this fully, we must focus on public sector production as well as on policy making.

In 1952, as much as 57% of total industrial production (value added at 1966 prices) and 56.7% of manufacturing output were accounted for by public corporations (Table 3.2). Since then, the government has repeatedly been under pressure either to freeze or to reduce the size of its holdings. Partly under the persuasion of U.S. Agency for International Develop- ment, majority or 100% equity in four public corporations (one highly prof- itable, two others distinctly less so) was transferred to landlords in 1954 as partial compensation for confiscations carried out under the 1953 land re- form. There has also been sporadic pressure for denationalization from local capitalists, who want less crowding out in credit markets or who want a greater share of the action in lucrative state enterprises (in 1982, however, fourteen of fifteen state enterprises were in the red).51 Finally, there has been pressure from what may be termed the liberalization lobby to reduce both tariff protection and the scope of public corporations.

Yet the government has resisted divestment, in the tenacious tradition of Sun Yat Sen, whose criticisms of monopoly capitalism became integral to Guomindang ideology.52 By the early 1980s, the share of the public sec- tor in manufacturing production had fallen to less than 20%. Nevertheless, the government remains dominant in such fields as heavy machinery, steel, aluminum, shipbuilding, petroleum, synthetics, fertilizers, engineering, and, recently, semiconductors. Almost every bank in Taiwan is also wholly or partially owned by the state (foreign banks were not allowed to establish operations until 1969). The lending activities of all financial institutions have

92 Alice H. Amsden

Table 3.2. Selected Indicators in Taiwan's Industrial Sector, 1952-80

Indicator

Industrial output index (1952 = 100) Share of industry in NDPfl Share of industry in total employmentb

Share of public ownership in total industrial production

Share of industrial products in total ex- ports

Selected shares in industrial production Mining Food processing Textiles Chemicals, petroleum, rubber, and

nonmetallic products Metal products, machinery, and

electrical machinery

1952

100 18 9

57

8

1962

303 26

(25)

46

51

1951

10.2 18.4 18.4

10.4

5.6

1975

2,010 39 36

19

84

1971

3.5 7.2

20.8

22.3

17.3

1980

4,108 46 42

18

91

1980

1.9 10.1 11.2

19.1

20.1

Note: Industry includes mining, manufacturing, construction, and utilities. "Net domestic product; all shares are expressed as percentages. bThe figure for 1952 is from Gustav Ranis (see note 49); the figure in parentheses is for 1967. Beginning with 1967, figures are averages per 1,000 workers. Source: Taiwan Statistical Data Book (Taibei: Council for Economic Planning and De- velopment, 1981); National Income of the Republic of China (Taibei: Directorate General of Budgets, Accounts, and Statistics, 1981).

been under strict state supervision.53 Thus, if the government in Taiwan does not quite "control the commanding heights/' it goes a long way to- ward doing so.

The government has been slow to divest itself of its holdings for two basic reasons. From the beginning, public enterprise has served to consol- idate the power of the Mainlander bureaucracy. In recent years, public enterprise has also allowed the Guomindang to buttress its own power vis- a-vis foreign capital. One of the fundamental consequences of public en- terprise has been the control by the state rather than by multinationals of key sectors in the economy. This is not to belittle the power of the multi- nations, nor to suggest the absence of an organic solidarity between the productive activities of the state and foreign investors. Recently, in the case of automobiles, they have tried to ally to form a nucleus of expansion. But the state has held its own in several crucial respects. The government did not abandon its traditionally conservative attitude toward foreign invest- ment until the export boom of the late 1960s had gotten underway. Only then did foreign firms begin arriving in Taiwan in significant numbers.

By 1971, overseas Chinese and other foreign investments amounted to

The State and Taiwan's Economic Development 93

roughly one-seventh of total registered capital (about the same as in Bra- zil), although statistics on foreign investment are problematic.54 Foreign investments, however, have been concentrated in electronics, chemicals, and textiles destined for export. Foreign investments have not mush- roomed with the government's turn to heavy industry. According to one account, between 1973 and 1980, foreign firms were responsible for as much as half of total investment and as much as 20% of total exports in the elec- tronics sector, but they were responsible for only 10% of total investment in manufacturing.55 Finally, Taiwan has not become highly indebted to in- ternational banks to finance its heavy industry. As discussed shortly, a relatively low reliance on foreign credit is due to a high domestic propens- ity to save. Moreover, although Taiwan has been a sizable international borrower by the standards of less developed countries, its international debt service/export ratio is very low - ranging between 5 and 10% - as a consequence of its very rapid growth of exports.56

The Jiang Jie-Shi government, therefore, cannot be said to have deliv- ered Taiwan into foreign hands, either by letting foreign banks dominate credit or by letting foreign firms dominate manufacturing. Nor can the gov- ernment be said to have been overwhelmed by its own technocracy (mostly American-educated), which has lobbied for lower tariffs and a privatization of state enterprises. The voice of the military has remained audible even as its visibility within the state apparatus has diminished (e.g., in the 1950s, an army general served as economics minister). The voice has remained audible to the extent that it appears to speak for continued state control of the economy for purposes of defense and social order, much as in the old Guomindang tradition.

This contrasts with the behavior of another military in another small de- veloping country with a highly educated population: Chile. Here, the mil- itary did buy the economic liberalism of its technocracy, closed the doors to state enterprise and opened them to foreign imports and investments - with disastrous effect.57 The point is that there is no necessary association between militarism and economic nationalism, although in Taiwan, the relationship is positive.

The public sector in Taiwan is also still very important as far as capital formation is concerned. Although the state's share of gross domestic in- vestment has fallen from a high of 62% in 1958, it still amounted in 1980 to as much as 50%.58 State spending has gone largely to finance ten major development projects in infrastructure, integrated steel, shipbuilding, and petrochemicals, which contributes to the fact that manufacturing in Taiwan has progressed in breadth (the percentage of manufacturing in GDP) and in depth (the percentage of "sophisticated" products and processes in total manufacturing output). Whereas in 1952 agriculture accounted for 36% of net total domestic product and manufacturing for only a tenth, agriculture now accounts for only 9% and manufacturing for more than a third.59 By 1980, 42% of all workers (15 and over) were engaged in the industrial sector

94 Alice H. Amsden

(Table 3.2). There has also been a decline in the absolute number of work- ers in the agricultural labor force since the early 1970s. In historical per- spective, this is highly significant. In the United States, absolute declines in the farm population did not begin until the 1930s.60

Nor has manufacturing been confined to "wigs and wallets," as myth once held. In the course of six "plan periods," which incidentally can be described, at most, only as "indicative," important structural changes have occurred within manufacturing. Textiles and food processing were the leading sectors during the first two plan periods (1953-56 and 1957-60).61

During the second plan period, however, the relative contribution of non- durable consumer goods, particularly food processing, declined, whereas that of intermediate goods (cement and paper) expanded. Chemicals (fer- tilizer, soda ash, plastics, and pharmaceuticals) assumed major importance during the third plan period (1961-64). Capital and the production of du- rable goods (electrical and nonelectrical machinery, such as radios and sewing machines, and transport equipment, such as bicycles and ships) as well as petroleum products grew enormously during the fourth and fifth plan pe- riods (1965-68 and 1969-72). The seventh plan period (1976-81) saw large increases in heavy industry. These structural shifts are evident in Table 3.2, although they are not to be exaggerated. The percentage of so-called light industry in manufacturing output was still 48% in 1977, down, however, from 54% in 1970 and 60% in I960.62

Exploiting the World Market

The nineteen-point reform program introduced by the government be- tween 1956 and 1961 made exporting highly profitable. Not only were ex- porters wined on tax and credit subsidies; they were also dined on tariff reductions from prewar China heights. This allowed them to take advan- tage simultaneously of advanced levels of world productivity for their in- puts and exotic Taiwan wage levels for their outputs. In what follows, we discuss the supply-side factors that permitted Taiwan to profit from the world market, directing attention to the role of the state.

Taiwan exports (90% of them manufactures) are highly competitive in world markets for reasons of low costs and rising productivity. There ap- pears to have been a substantial fall in Taiwan's wage costs per unit of output relative to those of competitors in the period spanning 1954-71, although there is a great deal of uncertainty about the figures.63 After the energy crisis in 1973, when the dust had settled, Taiwan could still be seen to have a cost advantage due to lower wage rates, by comparison not only with Japan, but also with South Korea, Hong Kong, and Singapore. Inter- national wage comparisons are difficult to make because of exchange-rate distortions and international variations in data collection. Nevertheless, es- timates of industrywide trade associations tend to corroborate the infor- mation contained in Table 3.3; to wit, wages in Taiwan are a good deal lower in all job categories than wages in competitor countries.

Table 3.3. Comparative Wages and Salaries, Midyear 1978 (Mean Monthly Salarya in U.S. Dollars)

Professional group

Industrial engineer Mechanical engineer Electrical engineer Accountant General manager Production manager Section chief Executive secretary Typist Junior clerk Foreman Skilled worker Semiskilled worker Unskilled worker Tool maker Cleaning worker

Taiwan

358 407 322 482

1,051 729 461 435 184 150 369 167 115 93

245 120

South Korea

639 587 509 930

1,192 1,049

889 590 318 344 493 318 311 146 330 208

Hong Kong

618 627 590 904

2,389 1,203 1,015

729 245 258 425 255 189 158 303 170

Singapore

821 710 803 923

2,097 1,215

518 503 211 198 413 206 133 102 210 108

Philippines

190 191 203 189 809 498 219 175 73

114 198 81 70 52 64 59

Thailand

437 460 492 498

1,784 669 392 425 153 135 203

86 76 51

135 77

Indonesia

752 786 299 733

1,513 851 439 496 214 173 241 137 96 65

482 53

Malaysia

552 677 762 949

2,672 1,050

786 404 133 180 390 181 129 84

304 82

Japan

1,587 1,244 1,025 1,521 3,413 2,488 1,587 1,360

756 785

1,500 1,161

942 698 665 550

Federal Republic of Germany

2,884 2,884 2,884 1,923 5,495 3,745 3,222 1,465 1,099 1,190 1,328 1,282 1,145

915 1,465

824

flIn addition to the monthly base salary, most companies pay regular bonuses ranging from 1 to 12 months of the salary, varying by company and by country. In order to make the monthly base salary information meaningful in terms of actual cost, this survey has increased the monthly base salary to include any bonuses paid. However, extraordinary bonuses, commission payments, etc., have not been included. Source: International Business Review, Hong Kong as cited and adapted by Anton Galli, Economic Facts and Trends (London: Verlag Weltforum for the Institut fur Wirtschaftsforschung, 1980).

96 Alice H. Amsden

Table 3.4. Long-Run Changes in Industrial and Labor Costs (Annual Rate of Change)

Real wages Money wages Labor productivity" Unit labor cost

1954-61

2.5 11.4 4.2 7.2

Wages,

1961-73

6.3 9.7

10.8 -1.2

Productivity,

1973-75

5.2 29.5 6.0

23.5

Note: Data are expressed as percentages. flThe first column is for the years 1953-61. Source: Adapted from Lundberg (note 64).

Apart from the noise of exchange-rate distortions, wage levels remain relatively low because the rate of increase in wages has been modest. What is more, the movement of wages in relation to labor productivity has been favorable for unit labor costs. In both regards, one detects the arm of the state. Labor unions are virtually nonexistent and strikes are prohibited, because Taiwan is still technically at war. One key to the advance in pro- ductivity is the enormous emphasis placed on public education, financed by a tax system that is highly regressive.

Table 3.4 presents data on changes in real and nominal wages, labor productivity, and unit labor costs, which constitute the most proximate causes of Taiwan's export success, as well as its extraordinary price stabil- ity.64 By comparison with other countries, including South Korea, Taiwan's rate of real wage increase has been low, particularly if one juxtaposes it against the rate of change of per capita income, which may be taken as a proxy for demand conditions in the labor market.65 Also striking is the rapid rise in labor productivity and the fall in unit labor costs in the critical period from 1961 to 1973.

The rapid rise in labor productivity has not been autonomous and may be understood as emerging out of the cumulative process of fast growth itself. As growth has accelerated, throughput time has decreased, profita- bility has risen, and investments have skyrocketed in foreign technology (both embodied in machinery imports and disembodied in the form of ser- vices). The scale of production has mounted and, with it, specialization and the division of labor; and, finally, the time required for the accumulation of experience has been telescoped, opportunities to use such experience to improve process and product have multiplied, and firm-level increments in productivity have been realized.66

That Taiwan has assimilated foreign technology as effectively as it has in no small part harks to its highly educated population. It was thought to be the most educated in all of Asia, with the exception of Japan, when the Japanese occupation of Taiwan ended after World War II. High invest- ments in education continued thereafter, increasingly in the technical fields.

The State and Taiwan's Economic Development 97

By the early 1970s, Taiwan had more engineers per 1,000 persons engaged in manufacturing than all other developing countries for which data are available, with the exception of Singapore.67 Since then, investments in education have soared in tandem with the rate of growth of output; the number of engineering students studying abroad almost doubled between 1975 and 1976, and the number of engineering students studying in Tai- wan doubled between 1968-69 and 1972-73 and then doubled again be- tween 1973 and 1982-83.68 Taiwan trains 50% more engineers in propor- tion to its population than the United States.69

Although there has been no shortage of profitable investment outlets for Taiwan firms to exploit, the opposite has been true of savings. Over the years, however, as a proportion of gross national product (GNP), savings have risen phenomenally. This has spared Taiwan the mutilation of over- extended international indebtedness, as discussed earlier. Savings as a per- centage of national income were less than 5% in 1952, 6.5% in 1962, 26.8% in 1972, and almost one-third a decade later.70 The composition of savings has also altered. The share of foreign savings has nose-dived, from around 40% in 1952-60, the heyday of U.S. aid, to a negative percentage in the period since 1972, whereas the shares of both public and private domestic savings have risen.71

According to Lundberg,72 the phenomenal rise in the savings ratio can to a considerable extent be explained by the high and increasing share of gross profits in national income. As for the government, it has been able to save as much as it has, despite a fivefold increase in real expenditures, by holding back on certain allocations and by raising revenues. On the payments side, the proportion allocated to defense expenditures - the larg- est single item in the government budget - has been halved, and social, health, and pension payments have been kept remarkably low by Western standards. On the revenue side, the tax structure in Taiwan has remained regressive. Initially, agriculture provided a massive investment fund for industry. Now, tax receipts accrue largely from commodities; income and corporate taxation is either negligible or evaded. In short, a "favorable cli- mate for capital formation exists in Taiwan, with a minimum concern for the distribution of income and wealth."73

A few remarks on the relationship between the Nationalist government and Taiwan firms are in order, lest it be thought that the interests of capital have been served by the state unstintingly. Taiwan's ability, by compari- son with, say, a Latin American country, to situate its economy in the lap of the international market has rested in part on the clout that the state has used against private producers. Because the governance of private produc- ers is nowhere clearly defined, capital everywhere imagines incursions into its domain by the state, but it is understandable why such complaints are so vocal in Taiwan.

We have already noted how "the arm of the state reaches down to vir- tually every farmer."74 Similarly, "the government is deeply involved in all

98 Alice H. Amsden

aspects of industry."75 Even after most import and foreign exchange con- trols were lifted, the government exercised its will through the myriad li- censes necessary for a firm to operate; the requirement of prior approval for foreign loans and technology agreements; public ownership of the banking system, which held interest rates much higher than in most Third World countries; vagueness in tax laws such that politically uncooperative firms could be threatened with audits; and so on. All businesspersons agree that they could be put out of business at the caprice of the state in a matter of months. Whereas a large percentage of output in, say, South Korea is realized by giant conglomerates, Taiwan is still a land of small farmers and firms.76 Recently, government policy has shifted toward the encourage- ment of consolidation in agriculture and merger in industry. Not only has this policy met resistance by the parties concerned; one might also imagine the government's ambivalence toward it, for small entities tend to be less threatening and more tractable than large ones.

Rentiers and capitalists alike in Taiwan have, in the past, experienced the dismembering or disabling effects of state power: The former were ex- propriated by a land reform, and the latter succumbed to the liberalization reforms that ushered in export-led growth. When a government charts a new economic course, entrenched interests are threatened. In the case of a turn to export-led growth, class conflict is stirred up by devaluation and deflation. A liberalization of imports hurts import-substituting firms and the banks that are financing them. When these are foreign, they can bring extraeconomic power to bear against new policy directives. These conse- quences of the turn to export-led growth were relatively mild in Taiwan. For one thing, the economy was both far less industrialized and far less inflationary than, say, some of the Latin American economies, so that fewer entrenched interests were upset. The state also controlled many sectors that would otherwise have been hurt, and foreign firms and foreign banks were absent. Nevertheless, private domestic firms that were not satisfied with the sweeteners of export subsidies and protection for second-genera- tion import substitutes were swept away. In short, Taiwan was better able to turn to the international market than other poor countries because the balance of power between the state and both labor and capital was weighted far more to the state's advantage.

To a certain degree Taiwan enjoyed an edge over other developing coun- tries in its export effort because of historical and geopolitical specificities: Taiwan's careful study of its erstwhile colonizer enabled it to absorb the Japanese economic experience and to appreciate that export-led growth was a viable strategy. In addition, the way was paved for the acceptance of Taiwan products in the U.S. market first by Japanese exports and then by those of U.S. multinationals, which had chosen Taiwan as an export platform after U.S. aid had secured it for democracy. Clearly, however, these explanations of Taiwan's ability to exploit the international market - whereas most developing countries appear instead to have been exploited

The State and Taiwan's Economic Development 99

by it - are dwarfed by those explanations that focus on the forceful manip- ulation of Taiwan's political economy by the state.

Economic Growth and the Changing Nature of the State

The initial puzzle of the role of the Taiwan state was not just a question of magnitude. In terms of scope, Taiwan is simply a particularly striking ex- ample of the positive association between state intervention and the accel- eration of economic growth that is now generally accepted to prevail in cases of Third World capitalist development.77 What made Taiwan partic- ularly interesting was that the state was so effective despite the clear prior- ity the military placed on defense over development and the clear domi- nance of the military within the state apparatus. The discussion so far has not resolved this puzzle. On the contrary, the turn to export-led growth seems to make the relation between apparent military dominance of the state and successful economic policy even more perplexing.

The important role of public enterprises, like the early economic self- sufficiency promoted by the state, is quite consistent with what can be presumed to be the interests of the military. Import-substitution industrial- ization, low reliance on foreign investors, and a focus on greater output of basic foodstuffs would all square with the military's presumed preference for autarky. Export-led growth, on the other hand, appears to contradict the policies the military could be expected to favor. Why should a regime fanatically committed to national security tolerate a policy change that made the Taiwan economy highly vulnerable to foreign supply and demand as well as more dependent on foreign-owned firms?

The answer to this question is complex. First, it must be emphasized that the increased reliance on international markets and capital that character- ized the civilian economy did not extend to defense production. The mili- tary ran its own production facilities, supplied with basic inputs by state enterprises. In addition to state enterprises, there are many special-status companies that obtain favors and/or incentives from the state. Among these are the enterprises owned or invested in by the Guomindang or by the Vocational Assistance Committee for Retired Soldiers, a complex of more than forty firms reputed to be the largest single enterprise on the island.78

By the 1960s, military arsenals produced much of the equipment and less sophisticated weaponry and ammunition needed by the armed forces. In 1969, a loan was made to Taiwan to build a factory to co-produce military helicopters (with Bell Helicopter Co.). By that year, Taiwan was producing M-14 rifles, machine guns, artillery shells, mortars, and other defense ma- terials.79 Soon, another small step toward self-sufficiency was taken with the initiation of an agreement with Northrop Aircraft Co. to co-produce F- 5E fighter planes.80 This mushrooming military economy may have made the generals less concerned with the diminishing autarky of the civilian sector.

100 Alice H. Amsden

Export-led growth, then, was consistent with the maintenance, and even expansion, of previous levels of autarky in the sectors that were most crit- ical for warfare. Nevertheless, there is abundant evidence, beginning at the time of Taiwan's turn to export-led growth, that the influence of the mili- tary over economic affairs began to wane. For instance, the government began to support family planning, whereas it was once deterred by the insistence of party elders that such a policy would reduce the number of soldiers available for retaking the Mainland.81 Taiwan ceased being self- sufficient in basic foodstuffs because it was more profitable to specialize in cash crops for export. Defense spending as a percentage of GNP and of government expenditures also fell (until recently). In 1960, when export- led growth was just getting underway, military expenditure as a per- centage of GNP was roughly 13% and, as a percentage of government expenditure, 65%.82 By 1978, these figures had fallen to 8 and 34%, respectively.83

Part of the answer to the apparent contradiction between the turn to export-led growth and the military's interests is that these interests were no longer reflected in state policy as monolithically as they had been in the period immediately following the takeover of Taiwan. And even if the mil- itary had held complete sway, it would have had little option other than to turn to export-led growth. Given the exhaustion of import-substitution in- dustrialization and foreign reserves, the continued pursuit of increased au- tarky would have been economically suicidal, although reliance should not be placed on the functionalist argument that because a decision was eco- nomically necessary the military would endorse it even at the expense of its own geopolitical ends.

More understanding is gained, however, by a closer examination of the changing relationship over time of economic and geopolitical factors. On the one hand, the dream of retaking the Mainland grew dimmer as the Communist regime there proved its durability. The repossession of China "had begun as a fierce resolve; it became an aspiration, then a myth, then a liturgy."84 At the same time, it became less clear that Taiwan could rely simply on its anti-Communist credentials to ensure support from its major ally, the United States. From this perspective, increased reliance on foreign investment, which went along with export-led growth, might be seen to be as valuable in securing powerful political allies as in reaping economic benefits.85

The key to the changing logic of Taiwan's international position, how- ever, was bread and butter. Taiwan, as well as Communist China, was proving itself to be a viable economic entity, as evidenced by its control over inflation and its fast growth. The economic viability of Taiwan meant that the population of Mainlanders as a social collectivity no longer needed to retake China to enrich itself. This was all the more true when it became clearer in the 1960s that economic gains in Taiwan were being realized not on the ephemeral basis of plunder, as they had been between 1945 and

The State and Taiwan's Economic Development 101

1949, or of buying cheap and selling dear, as they had been to some extent during the earliest years of import-substitution industrialization, but rather on the sustained basis of capital accumulation. I think that this, more than anything else, is what underlay the military's acceptance of export-led growth and its exit from the center of the stage of Taiwan's political economy.

In the last analysis, despite the continued power of the military within the state, Taiwan provides an interesting commentary on Engels's classic dictum that when "the internal public force of a country stands in opposi- tion to its economic development. . . . [ijnexorably and without exception, the economic evolution has forced its way through/'

86 Engels's concern was with the open contest between an economically reactionary elite's at- tempt to maintain its rule by political means and the transformative power of economic change. Taiwan does not exactly fit this description, but one can observe in Taiwan the powerful effect of economic change on the ori- entation of political force. The military was not overcome by the emergence of successful capitalist development, but it does seem that it found the expanding opportunities such development offered to be more attractive then the shrinking ones that were promised by a continued fixation on geopolitical struggle.

Taiwan, then, is more than a case in which the essential contribution of state intervention to economic development can be observed. It is a case that demonstrates the reciprocal interaction between the structure of the state apparatus and the process of economic growth. The Taiwan state, which appeared on its arrival from the Mainland to be an unlikely instru- ment for the promotion of development, proved to be a most effective one. At the same time, changes in the nature of the state itself appear to have been an important by-product of economic development. The state, in short, can be said both to have transformed Taiwan's economic structure and to have been transformed by it.

Notes

1. By "developed" is meant a fully employed economy wherein output rises not merely because inputs of land, labor, and capital rise, but also because produc- tivity rises.

2. Edwin A. Winckler, Remarks to a conference on Taiwan (Columbia University, 1982).

3. I. M. D. Little, "An Economic Reconnaissance," in Economic Growth and Struc- tural Change in Taiwan: The Postwar Experience of the ROC, ed. Walter Galenson (Ithaca, N. Y.: Cornell University Press, 1979), pp. 448-507.

4. The sections of this chapter on Taiwan's colonial heritage and agriculture bor- row heavily from Alice H. Amsden, "Taiwan's Economic History: A Case of Etatisme and a Challenge to Dependency Theory," Modern China, July 1979: 341-79.

5. A. Y. C. Koo, The Role of Land Reform in Economic Development (New York: Prae- ger, 1968).

102 Alice H. Amsden

6. H. Y. Chang and R. H. Myers, "Japanese Colonial Development Policy in Tai- wan, 1895-1906: A Case of Bureaucratic Entrepreneurship," Journal of Asian Studies 22 (1963): 433-49; S. P. S. Ho, 'The Development of Japanese Colonial Government in Taiwan, 1895-1945," in Government and Economic Development, ed. Gustav Ranis (New Haven, Conn.: Yale University Press, 1971), pp. 287- 327.

7. R. H. Myers, "The Commercialization of Agriculture in Modern China," in Economic Organization in Chinese Society, ed. W. E. Willmott (Stanford, Calif.: Stanford University Press, 1972), pp. 173-91.

8. A. J. Gradanzev, Formosa Today (New York: Institute for Pacific Relations, 1942). 9. S. P. S. Ho, "Agricultural Transformation under Colonialism: The Case of Tai-

wan," Journal of Economic History 28 (1968): 313-40; R. H. Myers and A. Ching, "Agricultural Development in Taiwan under Japanese Colonial Rule," Journal of Asian Studies 23 (1964): 555-70; R. P. Christensen, Taiwan's Agricultural Devel- opment, prepared for the Department of Agriculture, Economic Research Stud- ies, Foreign Agricultural Economic Report no. 39 (Washington, D.C.: U. S. Government Printing Office, 1968); S. C. Hsieh and T. H. Lee, Agricultural De- velopment and Its Contributions to Economic Growth in Taiwan: Input-Output and Productivity Analysis of Taiwan Agricultural Development, Chinese-American Joint Commission on Rural Reconstruction, Economic Digest Series, no. 17, Taibei, 1966. Teng-Hui Lee, Inter-sectional Capital Flows in the Development of Taiwan: 1895-1960 (Ithaca, N. Y.: Cornell University Press, 1971).

10. Myers and Ching, "Agricultural Development in Taiwan." 11. Ho, "Development of Japanese Colonial Government." 12. T. Ouchi, "Agricultural Depression and Japanese Villages," The Developing

Economies 5 (1967): 597-627. 13. Edwin A. Winckler, "National, Regional, and Local Politics," in The Anthropol-

ogy of Taiwanese Society, ed. Emily A. Ahem and Hill Gates (Stanford, Calif.: Stanford University Press, 1981), pp. 13-37.

14. Simon Kuznets, "Growth and Structural Shifts," in Economic Growth and Struc- tural Change in Taiwan, ed. Galenson, pp. 15-131.

15. Barbara W. Tuchman, Stilwell and the American Experience in China, 1911-1945 (New York: Bantam Books, 1974).

16. Ralph N. Clough, Island China (Cambridge: Harvard University Press, 1978). 17. N. Jacoby, U.S. Aid to Taiwan (New York: Praeger, 1966). 18. Clough, Island China. 19. See Christensen, Taiwan's Agricultural Development. 20. Koo, Role of Land Reform. 21. Taiwan Agricultural Yearbook (Taibei: Department of Agriculture and Forestry,

Provincial Government, 1974). 22. Lee, Inter-sectional Capital Flows. 23. John C. H. Fei, Gustav Ranis, and Shirley W. Y. Kno^-Growth with Equity:

The Taiwan Case (New York: Oxford University Press for the World Bank, 1979).

24. Lee, Inter-sectional Capital Flows, p. 28. 25. Sung Hsing Wang and R. Apthorpe, Rice Farming in Taiwan: Three Village Stud-

ies, monograph ser. B, no. 5 (Taibei: Academia Sinica, 1974), pp. 10-11. 26. Christensen, Taiwan's Agricultural Development. 27. A cruder form of state power with a purpose altogether unrelated to economic

The State and Taiwan's Economic Development 103

planning was also remarked upon by Wang and Apthorpe (Rice Farming in Taiwan):

At least for as long as relations between island and continental China continue in their present form, presumably a justification will be found for continuing a form of reliance on the kind of police methods which have now become part and parcel of everyday life. The Minister of the Interior in the Nationalist government used very often in 1971, for instance, to the astonishment of persons familiar with a very different tradition, to speak of the policeman as the most important resource person of all for community develop- ment in the island. Villagers, too, speak of the intimacy of police participation in parts of their daily life (p. 10).

28. W. P. Falcon, "Key Issues in Taiwan's Agricultural Development/' Industry of Free China 41(4) (1974): 2-7.

29. Christensen, Taiwan's Agricultural Development. 30. Shirley W. Y. Kuo, "Income Distribution by Size in Taiwan Area: Changes and

Causes," Industry of Free China 45, nos. 1-3 (1976): 9-38, 9-21, 20-34, respec- tively. It is interesting that, in spite of the high degree of commercialization of Taiwanese agriculture, the government placed minimal reliance on market forces to extract a surplus from the countryside. Rice collections were made in kind, and rice was bartered for fertilizer. Indicative of the government's avoidance of the market mechanism were its attempts (albeit unsuccessful) to barter rice not only for fertilizer but also for cotton cloth, bicycles, soybean cakes, and the like. See Kuo, "Income Distribution."

31. Hla Myint, The Economics of the Developing Countries (New York: Praeger, 1964), p. 48.

32. T. H. Shen, "A New Agricultural Policy," in Agriculture's Place in the Strategy of Development: The Taiwan Experience, ed. T. H. Shen (Taibei: Joint Commission on Rural Reconstruction, 1974), pp. 38-58.

33. Hsieh and Lee, "Agricultural Development." 34. Even after the export of labor-intensive manufactures got underway, Taiwan

ran a trade deficit. The trade balance remained negative until 1969. It became negative again in 1974 and 1975 (Taiwan Statistical Data Book [Taibei: Council for Economic Planning and Development, 1981]). This is a consequence of the fact that per capita income has been growing rapidly and so, too, have imports, especially since 1973 (see Table 3.1). Much exporting also relies on imported inputs. Had Taiwan's agriculture not been so productive, the strain on the balance of payments would have been greater.

35. Free China Review (Taibei, March 1976). 36. I. M. D. Little, T. Scitovsky, and M. Scott, Industry and Trade in Some Developing

Countries (London: Oxford University Press, 1970). 37. The OECD's study of industrialization in seven developing countries compared

effective rates of protection and found them to be lowest in Mexico and next lowest in Taiwan (the effective rate of protection measures the percentage by which import restrictions make it possible for the price of the value added in production to exceed what it would be in their absence). The study cautions, however, that, although "the average levels of protection for all manufacturing industry . . . in Taiwan were moderate . . . these moderate levels were due to zero or negative protection for exporting, while protection given to production for sale on the home market was much higher than in Mexico" (Little, Scitov- sky, and Scott, Industry and Trade).

104 Alice H. Amsden

38. Maurice Scott, "Foreign Trade/7 in Economic Growth and Structural Change in Taiwan, ed. Galenson, pp. 308-83.

39. Teng-Hui Lee and Kuo-shu Liang, "Taiwan/' in Development Strategies in Semi- industrial Economies, ed. Bela Balassa (Baltimore, Md.: Johns Hopkins Press, 1982), pp. 310-50.

40. Balassa, ed., Development Strategies. 41. Robert Wade, "Dirigisme Taiwan-Style," IDS Bulletin 15 (2) (April 1984): 65-70

(Institute of Development Studies, Sussex, England). 42. Ching-Yuan Lin, Industrialization in Taiwan, 1946-1972: Trade and Import Substi-

tution Policies for Developing Countries (New York: Praeger, 1973). 43. H. D. Fong, "Taiwan's Industry, with Special Reference to Policies and Con-

trol," Journal ofNanyang University 11 (1968). 44. Mo-Huan Hsing, Taiwan and the Philippines: Industry and Trade Relations (Lon-

don: Oxford University Press, 1971). 45. K. T. Li, My Views on Taiwan's Economic Development: A Collection of Essays from

1975-1980 (Taibei, 1980), p. 10. 46. Ibid., p. 15. 47. Ibid., p. 3. 48. See Little, "An Economic Reconnaissance." 49. Gustav Ranis, "Industrial Development," in Economic Growth and Structural Change

in Taiwan, ed. Galenson, pp. 206-62. 50. Economists of various ideologies have debated the economic importance of U.S.

aid. Economic nationalists, advocates of export-led growth, and advocates of Third World self-reliance all tend to disparage the consequences of aid, though their reasons are very different (see Scott, "Foreign Trade," for a discussion of different views, including his own). In contrast, Jacoby, U.S. Aid to Taiwan, author of the most thorough study of aid to Taiwan, argues that without it per capita income would have stagnated, assuming no cutback in military outlays. Sifting through these debates and with the hindsight of history, I tend to agree with the disparagers.

51. Specifics of the relationship between the Taiwan state and Taiwan business- persons are still something of a mystery, but see Robert Silin, Leadership and Values: The Organization of Large-Scale Taiwanese Enterprises (Cambridge: Harvard University Press, 1976), who also discusses managerial practices in Taiwan. There can be no doubt that these have also played an important role in the development process.

52. P. M. A. Linebarger, The Political Doctrines of Sun Yat-Sen (Baltimore, Md.: Johns Hopkins Press, 1937).

53. Hsing, Taiwan and the Philippines. 54. "Manufacturing (Taiwan Area)," in Industrial and Commercial Census of Taiwan

and Fukien, vol. 3, Table 16 (Taibei, 1971). E. L. Bacha, "Issues and Evidence on Recent Brazilian Economic Growth," World Development 6 (1977): 46-47.

55. This information was compiled by Professor Chi Shive of Taiwan National Uni- versity and cited by The Economist, July 31, 1982, pp. 1-14.

56. Morgan Guarantee Trust, World Financial Markets (New York, June 1982). 57. Alejandro Foxley, Neoliberal Experiments in Latin America (Berkeley: University

of California Press, 1983); see also Alfred Stepan, Chapter 10, this volume. 58. Taiwan Statistical Data Book (1981). 59. Ibid. 60. S. Lebergott, "Labor Force and Employment, 1800-1960," in Output, Employ-

The State and Taiwan's Economic Development 105

merit and Productivity in the U.S. after 1800, vol. 30 of Studies in Income and Wealth (National Bureau of Economic Research, Conference on Research on Income and Wealth, New York, 1966), pp. 126-31.

61. The analysis of early industrialization that follows is based on Teng-Hui Lee, "Development of Industry," in Agriculture's Place in the Strategy of Development, ed. Shen, pp. 66-70, and on Industrial and Commercial Census of Taiwan and Fu- kien (1971).

62. Li, My Views on Taiwan's Economic Development. 63. Scott, "Foreign Trade." 64. The rate of inflation in Taiwan was 12.3% between 1952 and 1961 and an un-

believable 2.9% between 1962 and 1972. This compares with 33.9 and 13.6%, respectively, for Korea in the same periods; see Erik Lundberg, "Fiscal and Monetary Policies," in Economic Growth and Structural Change in Taiwan, Galen- son, pp. 263-307. Taiwan's price performance, however, has deteriorated since the energy crisis. The GNP deflator was 32.3% in 1974 alone and averaged 10.4% for the period 1971-81. Economists have been unable to explain this deterioration except as a general consequence of both demand pull and cost push factors (Shirley W. Y. Kuo, The Taiwan Economy in Transition [Boulder, Colo.: Westview Press, 1983]).

65. Richard Webb, "Wage Policy and Income Distribution in Developing Coun- tries," in Income Distribution in the Less-Developed Countries, ed. by Charles R. Frank, Jr., and Richard C. Webb (Washington, D.C.: Brookings Institution, 1977), pp. 215-58.

66. Alice H. Amsden, "The Division of Labor Is Limited by the Rate of Growth of the Market: The Taiwan Machine Tool Industry Revisited" (Harvard Univer- sity, Graduate School of Business Administration, 1983, Mimeographed).

67. Manuel Zymelman, Occupational Structures of Industries (Washington, D.C.: In- ternational Bank for Reconstruction and Development, 1980).

68. Taiwan Statistical Data Book (1983). 69. The Economist, July 31, 1982, pp. 1-14. 70. Gross capital formation has also risen, from 15.4% of GDP in 1952 to 36.3% in

1980. The last figure is something of a world record; see Council for Economic Planning and Development, Taiwan Statistical Data Book (1981).

71. Taiwan Statistical Data Book (1982). 72. Lundberg, "Fiscal and Monetary Policies." 73. Ibid., p. 300. 74. Wang and Apthorpe, Rice Farming in Taiwan, pp. 10-11. 75. Silin, Leadership and Values, p. 18. 76. For information on the structure of Korean industry, see L. P. Jones and II.

SaKong, Government, Business, and Entrepreneurship in Economic Development: The Korean Case (Cambridge: Council on East Asian Studies, Harvard University Press, 1980); for information on both Korea and Taiwan, see S. P. S. Ho, "Small- Scale Enterprise in Taiwan," Staff Working Paper no. 384 (Washington, D.C.: World Bank, 1980).

77. See Rueschemeyer and Evans, Chapter 2, this volume. 78. Silin, Leadership and Values. 79. Clough, Island China. 80. Stockholm International Peace Research Institute, World Armaments and Disar-

mament Yearbook (London: Taylor & Frances, various years). 81. Clough, Island China.

106 Alice H. Amsden

82. Ho, "Development of Japanese Colonial Government/' 83. United States Arms Control and Disarmament Agency, World Military Expen-

ditures and Arms Transfers, 1969-78 (Washington, D.C.: U.S. Government Print- ing Office, 1980).

84. Brian Crozier, The Man Who Lost China (New York: Scribner, 1976). 85. Denis Fred Simon, 'Taiwan, Technology Transfer and Trans-Nationalism: The

Political Management of Dependency" (Ph.D. diss., University of California, Berkeley, 1980).

86. Frederick Engels, Anti-Duhring (New York: International Publishers, 1939), pp. 202-203.