موسسات و خطر اخلاقی در اقتصاد باز

I investigate the interaction between international trade and national institutional development in an environment characterized by heterogeneous individuals choosing their education levels to maximize their utilities; and institutions alleviating moral hazard by allowing managers to better observe and verify the productive efforts of workers. Liberalized trade allows institutions to serve as independent sources of comparative advantage. In this setting, I examine the effect of trade liberalization on the distribution of income in institutionally developed and underdeveloped nations. Trade affects income via a direct effect on prices and an indirect effect on the incentives to invest in education.

Many of the consequences of international trade remain a mystery, among them the effects of trade on the growth and distribution of income.1 Trade liberalization seems a necessary though insufficient cause of economic development. Trade allowed the Asian tigers to capitalize on their comparative advantages, helping grow their economies. Yet other countries, even those facing lower trade barriers, have not enjoyed the same success.2 As Birdsall et al. (2005) note, “history and economic and political institutions have trumped other factors in determining economic success.” If a nation's institutions and its access to foreign markets jointly determine its economic development, then we must understand the interaction between institutions and international trade. Such an understanding may also provide insights into the impact of trade on inequality within nations. A growing literature documents the importance of institutional quality on economic growth and development. There is also a long history of theoretical and empirical investigation into the effects of international trade on growth and the distribution of income. However, relatively little work has been done on the effect of institutional quality on the pattern and consequences of international trade. There are at least two reasons to address this gap in the literature. First, not understanding the interaction between institutional quality and international trade may bias empirical studies of the effect of trade on incomes. I find, for example, that liberalizing trade directly influences agents' education decisions, and in this way may indirectly increase income inequality. This may help explain the lack of consensus about the role of international trade in widening America's wage gap. Second, not understanding the interaction between institutional quality and international trade may raise unrealistic expectations about the benefits of liberalized trade. I consider a framework in which institutions alleviate moral hazard. There are two industries: X and Y. An agent works alone in the Y sector, but the X sector requires the joint effort of two workers. The two industries differ in that individual productive effort is observable and verifiable only in the Y sector. Neither individual effort nor output is observable or verifiable in the X sector, requiring managers in that sector to monitor employees to determine individual contributions. I assume that better trained managers are better able to monitor their employees. I also assume that managers in countries with more developed institutions are better able to monitor their employees: managerial training and national institutions mitigate moral hazard. In order to capture the effect of the interaction between institutional quality and international trade, I consider a simple game in which agents choose their sector of employment, level of job training, firm, and level of effort. Agents differ according to “natural ability”: agents with more ability obtain training while incurring a lower utility cost of education. This framework enables me to explore subtle interactions in a realistic but tractable setting. Industries differ in the difficulty of judging individual contributions to firm output. Such judgments are straightforward in settings where units of output can be traced back to individual workers. Certain traditional forms of production–for example, agricultural harvesting, handloom weaving of textiles, and craft production of shoes–best exemplify the circumstances in which an employer can view both an employee's work and his product and draw a correlation between the two. Even industries in which more than one worker contributes to a given unit of output allow such straightforward judgments when each worker contributes to production in such a deliberately prescribed manner that specific units of output can be traced back to individual workers. This case obtains in assembly-line production. But the modern workplace is particularly characterized by another type of production, one in which multiple workers jointly produce each part of each unit of output. When teams share responsibility for output, employers often have no clear and systematic way to assign responsibility. Investment banking exemplifies team production. That workers exert more productive effort when they are rewarded for their efforts is hardly controversial. Individual production industries, like agricultural harvesting, find it easy to provide powerful incentives. A piece rate wage suffices. Farmers pay laborers for each piece of fruit they harvest. All things being equal, a worker choosing to exert less effort earns less than one exerting more. A laborer cannot easily fool the farmer. But an investment bank dependent on team production cannot provide such straightforward incentives. This is particularly true for junior staff members who are not responsible for bringing in or maintaining clients. Piece rate wages lose their effectiveness when employees work collaboratively on their analyses and presentations. The parts of the whole–especially in good work–cannot readily be attributed to individual workers, rendering it difficult to pay workers for their individual contributions. Management is thus forced to rely on imperfect performance measures if it wishes to judge the contributions of individual workers in team-production industries. Absent such performance measures, employers lack meaningful incentives and workers are unable to reap the fruits of their labor. Management in team-production industries thus needs to establish performance metrics based on such criteria as attitude, effort, and quality of ideas to evaluate and reward employees. Management also needs to institute evaluation systems to measure how its employees carry out their work. Good management often institutes overlapping channels of communication to allow the reporting–and evaluation–of multiple perspectives on each worker's interactions with peers and supervisors. But the firm–and whatever performance evaluation system its management embraces–forms only part of the picture when it comes to rewarding and incentivizing a high-performance culture. The broader institutional environment in which a team-production firm operates also influences the degree to which worker contributions can be attributed to specific individuals. This broader picture clearly involves nations and the nature and quality of their institutions. Countries differ, for example, in their accounting systems. Effective accounting systems–particularly those to which the state delegates some authority–regularize the reporting of data through which both the productivity of firms and the contributions of their various elements–down to the level of individual employees–can be assessed. Judicial systems also play important roles. The more competent a country's legal institutions, the more accurately courts can assign credit for specific contributions to production processes when disagreements enter the judicial system. The quality of the legal establishment influences incentives even when no disputes arise. Parties to contracts understand that more competent legal systems can verify a larger fraction of the observable facts. Employees thus exert effort, believing that employers will be legally obligated to reward their work. Better trained managers and better national institutions thus improve the quality of performance measures. This assumption leads, in a very straightforward manner, to the first two results of the paper. First, better trained individuals enter the team production industry where they are rewarded for their ability to minimize moral hazard. Second, all else being equal, trade liberalization encourages countries with better developed national-level institutions to specialize in team production industries. Institutional-quality induced international specialization has important consequences for cross-country comparisons of income and education. The citizens of countries specializing in team production will choose to receive more education than equally able and advantaged citizens of countries with less developed institutions. This conclusion allows us to reconcile two distinct viewpoints within the development and growth literatures. Mankiw et al. (1992), for example, demonstrate starkly and undeniably positive effects of human capital accumulation on growth and income. On the other hand, Acemoglu et al., 2001 and Acemoglu et al., 2002 link reversals in economic fortune with reversals in institutional development. This paper establishes that human capital accumulation and institutional quality are inextricably linked. Individuals endogenously choose their levels of schooling understanding that institutional development determines the benefit to education. The interaction between international trade and institutional quality can also help explain intra-national phenomena. My model predicts that trade tends to widen the distributions of education and income in countries with better developed institutions. It causes the opposite to occur in countries with under-developed institutions. This prediction is similar to typical Heckscher–Ohlin trade theory where the difference in real income between capitalists and laborers compresses in labor-abundant countries and widens in capital-abundant countries. Compared to the literature on moral hazard, the literature on institutions and trade is relatively incomplete. While I consider the role of institutions in alleviating moral hazard in imperfect labor markets, in a related paper, Matsuyama (2005) considers the role of institutions in alleviating problems with imperfect credit markets. Like my paper, Costinot (2004) investigates the interaction between imperfect contract enforcement and international trade. Costinot reaches a conclusion similar to Matsuyama's and mine: institutions can act as independent sources of comparative advantage. The paper that most resembles this one is Grossman (2004): the perfect institution benchmark model from which my work diverges nearly replicates Grossman's benchmark model. Grossman abstracts from institutional quality and focuses on the pattern and consequences of trade when nations differ in their distributions of talent. As with the majority of the trade literature, both Costinot and Grossman exogenously fix the level (Costinot) or distribution (Grossman) of human capital/talent per worker. My paper deviates from the standard trade literature in assuming that factor endowments–particularly the distribution of human capital–are determined endogenously. I show that institutional quality is the common cause of comparative advantage and human capital accumulation (whereas Costinot finds that both institutions and human capital per worker are independent sources of comparative advantage). This insight leads to a set of conclusions relevant to both the international trade and development economics literatures: although increased human capital is certainly important, institutional reform lies at the heart of the development process. Moreover, trade liberalization impacts incentives to accumulate human capital, differentially affecting nations situated at different points within the development process. Such conclusions lie outside of the scope of the Costinot and Grossman frameworks, and all frameworks that treats endowments as exogenous. The remainder of the paper is in six sections. I provide the setup of the model in Section 2. In it, I describe a two-sector economy in which training-augmented labor is the only input, I quantify agent utility, and I provide the game theoretic structure of the model. The Walrasian equilibrium of a world of perfect institutions, the case of the first best, is described in Section 3. This serves as a benchmark for what follows. In Section 4, I characterize the autarky equilibrium-occupational choice, job training decision, and utility of agents of different natural abilities. In Section 5, I consider the effects of institutional change within a small country open to trade. Section 6 deals with the effect of two large countries opening to trade with each other. Section 7 concludes.

In an economy in which managerial training mitigates moral hazard, job training yields greater rewards in team production sectors. More able agents, those for whom acquiring education is less costly, self-select into the team production industry while less-able agents self-select into the individual production industry. When not only agents but also countries differ in their abilities to ameliorate moral hazard, a parallel international self-selection occurs. Countries with developed institutions specialize in the team-production industry while countries with inferior institutions specialize in the individual production industry. The interaction between international trade and institutional quality has important consequences for the distributions of education and income. As a country with developed institutions liberalizes trade with countries that have weaker institutions, the developed country can anticipate a growing polarization in its distribution of education. The most talented agents will obtain comparatively more education while the least talented will obtain comparatively less. Trade liberalization also increases income inequality in developed countries, directly by increasing the relative price of the team production goods and indirectly through the effect of trade liberalization on education. In developing countries, the direct and indirect effects of trade on income inequality are reversed: both tend to decrease income inequality. Agents in institutionally developed nations tend to obtain more education and earn higher incomes than their equally skilled counterparts in nations with less developed institutions. While training levels are lower in underdeveloped nations, this model suggests that the correct policy prescription is not necessarily to increase education. The return to education is lower in South than in North because South's institutions are inferior. Unless institutions improve in South, it is optimal for Southerners to receive relatively less education than Northerners. Institutional development appears to be the key to economic development, even though increased inequality is a natural consequence of this development. Endogenizing institutional development presents an interesting extension to this model. Of particular interest is the interaction between international trade and the incentive to develop institutions. Corollary 1 provides insight into an extreme example. If a country has sufficiently underdeveloped institutions, then it may specialize in the individual production industry if it liberalizes international trade. In this case, there is no marginal benefit to institutional development; with no one working in the team production industry, no one benefits from marginal improvements in institutions. However, if this nation were in autarky, the marginal benefit to institutional development would be positive because the country would produce positive quantities of the team production good. Although this is a highly specialized example, it highlights the fact that trade, by altering the mass of agents who benefit from institutional development, influences the incentive to invest in institutions.