Definitions direct analytic attention and confine the scope of an empirical study; conducting a study requires an analytical framework. An analytical framework is required to reveal the conditions under which a particular institution is effective in generating a particular behavior, to expose causal relationships, to generate predictions, and to evaluate arguments. An analytical framework is particularly important in studying institutions, because beliefs and norms are unobservable.

Central to the analytical framework used here is classical game theory. Because the proof is in the pudding, this book contains five empirical studies that attest to its usefulness in studying institutions as defined here. These studies use an empirical method that combines detailed contexual knowledge of the situation and its history with explicit, context-specific modeling. This case study method uses contexual knowledge to develop a conjecture regarding the relevance of a particular intertransactional linkage and the related institution. It then uses an explicit (and in this work) game-theoretic model to evaluate this conjecture. This empirical method is employed in all the historical analyses presented in this book. It is elaborated on in Part IV.

The empirical studies presented here consider the institutional foundations of markets. Doing so departs from a long tradition in institutional analysis that goes back to Adam Smith and considers markets as primitives that need not be explained. According to this view, markets emerge spontaneously when and where there is an opportunity for profitable exchange. As O. Williamson put it, “in the beginning there were markets” (1975, p. 20).

Societal organization – complexes of economic, legal, political, social, and moral institutions – is highly correlated with per capita income in contemporary societies: most developing countries are “collectivist,” whereas the developed West is “individualist.” In collectivist societies the social structure is “segregated,” in the sense that each individual interacts socially and economically mainly with members of a particular religious, ethnic, or familial group. Within these groups, contract enforcement is achieved through informal economic and social institutions. Little cooperation exists between members of different groups, but members of collectivist societies feel involved in the lives of other members of their group.

In individualistic societies, the social structure is “integrated,” in the sense that economic transactions are conducted among people from different groups, and individuals frequently shift from one group to another. Contract enforcement is achieved mainly through specialized organizations, such as courts. Self-reliance is highly valued.

Sociologists and anthropologists believe that the organization of society reflects its culture, an important component of which is cultural beliefs. Cultural beliefs are the shared ideas and thoughts that govern interactions among individuals and between them, their gods, and other groups. Cultural beliefs differ from knowledge in that they are not empirically discovered or analytically proved. Cultural beliefs become identical and commonly known through the socialization process, by which culture is unified, maintained, and communicated.

That cultural beliefs influence outcomes is intuitive, but formal examination of the relations between cultural beliefs and societal organization is subtle.

In writing this book, I have relied heavily on primary sources. The most important source was extensive interview data from local governments and firms over an eight-year period (September 1996 to August 1997, June–August 1998, July 2000, June–July 2002, January 2003, and July–August 2004). In each municipality, interviews were conducted with officials in the auto office of the municipal government, the general managers (in some cases) and the purchasing managers (in all cases) of the JV assembly project (both Chinese and foreign), managers in the head office of the Chinese partner, the general managers of individual supply firms (at both JVs and Chinese SOEs), and in some cases the managers in charge of manufacturing.

Interviews were structured (according to the category of interviewee), but were open ended. Questions would start off with general background, proceed to the nuts and bolts of operations, and then finally move on to broader questions (biggest challenges, prospects for the future). To give a flavor of the types of questions that were asked, local government officials were asked general background questions about the structure of the local auto sector (number of firms, ownership of firms), the organization of the corporate group, and its bureaucratic, financial and personnel relationships with local government, reasons for supporting auto sector development, specific means of support (investment figures), and so on.

Despite the dramatic twists and turns that have typified China's economic history during the last century, the city of Shanghai has more often than not symbolized each particular era. In 1987, writing in the party mouthpiece Hongqi, Mayor Jiang Zemin put forth the official CCP view of Shanghai history: “Before Liberation it was a microcosm of old China. It reflected in a concentrated form the major contradictions and various pathologies of the colonial, semi-colonial, and semi-feudal society. After Liberation, as the important economic centre and industrial base of our nation, its development fully manifested the vitality and superiority of the socialist system.” Three decades of rigid socialist planning removed many of the outward signs of bourgeois capitalism, but – despite official pronouncements to the contrary – also removed much of the city's former economic vitality. Losses in the state sector, which dominated industrial output and urban employment in the city, continued to rise during the 1980s; the symbol of socialist development became the symbol of socialist stagnation. Shanghai's leaders sought once again to transform the city, this time through reform and development.

Shanghai municipal leaders, like leaders in other cities, chose the auto sector as a prime candidate for development because the extensive linkages of the industry created the potential for a broad impact. These same linkages, however, increased the challenge of coordinating development in the sector.

The analysis in this book accepts the notion that people tend to respect the socially expected and normative sanctions (Chapter 5). It also rests on a particular notion of rationality, maintaining that when institutions generate behavior, socially articulated and disseminated rules regarding the situation span the domain that people understand and within which they can act rationally. Are these two premises consistent with each other? Is it appropriate to consider individuals as strategic while recognizing that social and normative considerations influence behavior? Or should we model people as homo sociologicus, as passive rule followers? Specifically, is it appropriate to model individuals who have such social and normative inclinations as rational decision makers when they are guided by socially articulated and disseminated rules? Do they have stable preferences regarding outcomes? Are they motivated by the consequences of their actions? In other words, do they act strategically? This appendix presents evidence to support the claim that, although people have social and normative propensities, it is nevertheless appropriate and necessary to consider them as rational in the abovesense.

Experimental game theory is a promising analytical framework to address these questions, particularly because participants share common knowledge of the rules of the game and many experiments were explicitly designed to reveal individuals' social and normative inclinations. These experiments provide three ways to address the foregoing questions: considering whether nonrational explanations better fit the data, testing whether the observed behavior is consistent with some well-behaved preference ordering, and using experimental results to determine whether people are motivated by consequences and behave strategically.

The challenge for Chinese leaders during the first two decades of reform and development – much like in Japan in the 1960s or Korea in the 1970s – was to choose a development path that would lead to the creation of competitive industries. Like its neighbors in previous decades, China faced the classic problems of late development. Rather than accept the division of labor dictated by comparative advantage, China sought to develop industrial sectors that would create a “multidimensional conspiracy” in favor of development: sectors and firms that would foster entrepreneurial activity and create positive spillovers in the economy as a whole. For reasons that had as much to do with national security and pride as economics, China's leaders were not content to build yet another workshop for the developed world, manufacturing whatever products required cheap labor and low skill levels. They wanted to fly airplanes made in Shanghai, use computers built in Beijing, and drive automobiles manufactured in Guangzhou. China was a poor country at the beginning of the reform era, but it was not lacking in ambition.

Although the ability of China to realize these ambitions rested on many factors, none was more important than its capacity to effectively utilize foreign direct investment (FDI) as a means of developing its own industrial base. By definition, a “late” developing nation confronts the challenge of creating strong and independent firms in a context of intense competition from the industrialized world.

This book grew out of an attempt to gain a better understanding of the causal factors underpinning economic and political outcomes during the late medieval period (circa 1050 to 1350). It was during this period that the Muslim (Mediterranean) World reached what many scholars consider to be the zenith of its commercial integration, whereas expansion of markets in Europe was so pronounced that prominent historians dubbed this phenomenon as “the Late Medieval Commercial Revolution.” Gaining a better understanding of this period therefore has the promise of advancing our knowledge regarding why and how effective markets and economically beneficial polities prevailed in some historical episodes but not in others. Although economists have long emphasized the welfare-enhancing implications of market expansion, we know surprisingly little about the source for historical trajectories of market development.

This period is also of interest because it was a point of bifurcation in the histories of the Muslim and European worlds. The Muslim world was probably more advanced economically, technologically, and scientifically than Europe during the late medieval period. Indeed, the Europeans learned a great deal from the Muslim world at the time (e.g., Watt 1987). In subsequent centuries, however, the Muslims developed economically and politically along a different path from the Europeans, and became economically worse off in the long run.

In attempting to understand this period and its implications on subsequent development, I benefited from the training in historical analysis that I received when pursuing an advanced degree at Tel Aviv University.

This chapter illustrates the merit of interactive, theoretically informed, context-specific analysis by examining a central question in economic history and development economics. This question concerns the institutional evolution that enabled increasingly more impersonal exchange in some economies but not in others (see North 1990; Greif 1994a, 1997a, 1998b, 2000, 2004b, 2004c; Rodrik 2003; Shirley 2004). We often assert that such institutional evolution facilitates specialization, efficiency, and growth. Yet we know little about the historical development of the institutional foundations of impersonal exchange.

This historical development is the focus of the chapter. It examines the nature and dynamics of institutions that supported impersonal exchange characterized by separation between the quid and the quo across jurisdictional boundaries in premodern Europe. Commerce expanded particularly during the three hundred years prior to the mid-fourteenth century even though there were no impartial courts with geographically extensive judicial powers to support exchange among traders from various corners of Europe. What were the institutions, if any, that supported interjurisdictional exchange characterized by separation between the quid and the quo over time and space? Specifically, were there institutions that enabled such exchange that was also impersonal, in the sense that transacting did not depend on expectations of future gains from interactions among the current exchange partners, or on knowledge of past conduct, or on the ability to report misconduct to future trading partners?

Shanghai succeeded in the initial stage of auto sector development because the local government took on many of the roles of a developmental state. Firms needed nurturing, and local officials provided it. Even though competition was limited during this period, no other municipality in China did this as well; the institutional infrastructure of Shanghai was particularly well suited for the development of capital- and scale-intensive manufacturing industries. The problem, however, was that Shanghai was succeeding under a very distinct set of circumstances – tariffs were high and products were stable – and just as the municipality was beginning to reap the gains of having mastered this stage of development, the rules of the game began to change. A dramatic rise in FDI and impending accession to the WTO resulted in increasing domestic competition, and cutting costs and increasing technical sophistication replaced the creation of basic manufacturing capability as the primary objective of the local industry.

How effectively would the various municipal auto sectors adjust as the level of global integration increased? None of them were close to being internationally competitive at the end of the 1990s. In 1996, after a decade of development and immediately prior to the dramatic increase in competitiveness, the Development Research Centre of China's State Council commissioned a report on the overall competitiveness of the Chinese auto industry.

For a developing country intent on transforming moribund state-owned firms into proud national champions, the auto sector is an obvious target. “As in the highly industrialized countries, the automotive industry … is becoming and will be without doubt the leading sector of the entire economy, by force of its magnitude, complexity, and dynamism.” Although these words were spoken by a forward-looking official in Brazil in the 1930s, they very well might have been uttered by an official in Tokyo in the 1950s, Mexico City in the 1960s, Seoul in the 1970s, or Beijing in the 1980s. The automobile industry has an enduring appeal for developing countries because the broad supply network creates extensive linkages and because it is often seen as a symbol of a modern industrialized country. Sheer numbers are in part responsible for its importance – currently nearly 50 million new vehicles are produced a year – but given the scope and complexity of the manufacturing processes in the auto industry, it has also been an important testing ground for new ways of organizing economic activity.

China has also used the auto sector as a testing ground, but as I explained in the previous chapter, rather than the central government dictating the development approach, it has been local governments that have experimented with new forms of organizing economic activity.

As my taxi made its way through the streets of Guangzhou, a city in southern China, the evidence of two decades of unprecedented economic growth flashed by the window. The darkened windows of a Mercedes-Benz sped by on the left; a family of four on a motorbike puttered along on the right; kamikaze taxi drivers veered in and out; a minibus trolled for passengers. My driver, at the wheel of a Chinese-made Volkswagen Jetta, was oblivious to the surrounding chaos, and as he maneuvered for position, I asked him how his car stacked up against the competition. “Better than most Chinese cars,” he replied. “It's made in China, but under the hood it's almost all foreign.” He proceeded to rank the quality of the cars on the road based on their percentage of foreign content: imports at the top, joint venture cars in the middle, pure local at the bottom. Although hardly a charitable assessment – the quality of domestically manufactured cars was improving rapidly – it was nevertheless truthful. It was 1997, my first year of field research for this project, and although China had come a long way from the lumbering two-ton “Liberation” truck in the breakdown lane, it was still a long way from the sleek Mercedes in the passing lane. The gap has been narrowing quickly.

In premodern trade, merchants had to organize the supply of the services required for the handling of their goods abroad, because goods were sold abroad only after being shipped to their destination (De Roover 1965; Gras 1939). A merchant could either travel with his merchandise or hire overseas agents to handle his affairs abroad. Employing agents was efficient, because it enabled merchants to avoid the time and risk associated with traveling and to diversify their sales across trade centers. Despite their efficiency, however, agency relations are not likely to be established unless supporting institutions are in place, because agents can act opportunistically and embezzle the merchants' goods.

This chapter examines the reputation-based economic institution – which can be referred to as a coalition – that enabled the Maghribi traders, a group of Jewish traders in the Mediterranean in the eleventh century, to deal with the contractual problem inherent in the merchant-agent transaction. In reputation-based institutions, future rewards or penalties in (auxiliary) economic or social transactions are made conditional on conduct in a central transaction. When effective, this intertransactional linkage enables an individual to credibly commit himself ex ante not to behave opportunistically ex post. In the case of agency relationships, the agent can commit to honesty and hence be trusted. Examining the operation of such institutions requires studying the institutional elements that create the intertransactional linkages and allow future utility to be conditioned on past conduct.