The dissertation develops and demonstrates a new Marxist approach to the epistemological problem of cognitive modernism, the problem of knowing the true laws of economic reality. This new approach is an antiessentialist and postmodernist critique of versions of Deweyan pragmatism. In American economics, versions of Deweyan pragmatism provide epistemological justification for the verity and primacy of two different economic theories of the world: the American Institutionalism of Thorstein Veblen and the Chicago School of Milton Friedman. Each school uses Deweyan pragmatism to ground its claim to be a science, and each uses Deweyan pragmatism to prove its contention that it offers the correct scientific analysis and view of the fundamental laws of operation of the economy. The dissertation demonstrates that Deweyan pragmatism cannot provide such justification. The primary reason is that Deweyan pragmatism, like all other philosophies of science, is subject to the epistemological problem of cognitive modernism. It is thus unable to provide objective, transdiscursive, and essential knowledge of economic reality. Chapter 1 is an introduction to modernist methodology in economics. It situates Deweyan pragmatism within the tradition of economic modernism. Chapter 2 examines the Deweyan pragmatism of Veblen's American Institutionalism. Chapter 3 examines the Deweyan pragmatism of Friedman's Chicago School. Both schools offer Deweyan pragmatisms as theories of knowledge which prove the truth of each's theory of society. Chapter 4 offers a postmodern critique of both modernist versions of Deweyan pragmatism. The analysis suggests several conclusions. First, for such different and directly opposed theories to claim a common affiliation to Deweyan pragmatism must mean that they understand that affiliation in fundamentally different ways. Second, by presenting different versions of pragmatism it becomes clear that it is not possible to discover the real Dewey, nor is it possible to evade the partiality of all readings of Dewey's philosophy. Third, by contesting pragmatism itself, I demonstrate that the cognitive modernist quest for certain foundations is a failed one, and that all knowledge products in economics are bound by the cultural conditions and discursive fields in which they are produced.

This dissertation studies the influence of the level of the real exchange rate on economic development. It first provides an econometric evaluation of this relationship. The main finding is that competitive real exchange rates tend to be associated with higher economic growth. The association is especially strong and statistically significant for developing countries. It then develops a historical narrative and episode analyses of several Latin American countries since the post-Second World War, in which the relationship is further investigated. There are three main findings. First, Latin American countries have experienced external and financial crises with immediate and/or long-lasting negative effects on growth that resulted from persistent real exchange rate overvaluation. Second, the most successful episodes of growth acceleration in the region have occurred in periods during which the governments aimed at maintaining stable and competitive real exchange rates. Finally, the analysis shows that currency over and undervaluation emerged as a result of explicit economic policies. This finding suggests that governments can use economic policy to manage the level and volatility of the real exchange rate to promote economic development. How macroeconomic policy needs to be managed to that end is evaluated with a formal model. The model shows that exchange rate policy targeting a competitive currency would more likely accelerate growth if it is implemented in coordination with domestic demand management policies that prevent non-tradable price inflation and wage management policies that coordinate the pace of wage increases with tradable productivity growth. An empirical illustration of these results is carried out through a comparative study of Argentina during the 2000s and Chile between the mid 1980s and the mid 1990s.

In view of the prevalence of information asymmetry and incentive incompatibility problems, the market for capital and the employment relationship are shown to be prey to pervasive agency problems that make economic transactions subject to enforcement costs. The presence of these costs subverts the Walrasian presumption of efficiency for decentralized markets, and the separation between efficiency and ownership concerns, resulting in forms of capital rationing and productive inefficiencies. Under these conditions, government intervention might help to attenuate the coordination failures and their associated efficiency losses for the economy by facilitating ownership of assets by those economic agents for whom information is relatively costless, and who have the capacity to take into account the consequences for all stakeholders. In this dissertation, we test the hypothesis that a union-controlled investment fund, such as the Fonds de solidarite des travailleurs du Quebec (FSTQ) might constitute an institutional innovation attenuating problems of information asymmetry and incentive incompatibilities. Through a case-study, we first analyze the original institutional features of the FSTQ that could help alleviate and/or compound the said agency problems. We then turn to an analysis of a segment of the FSTQ investment portfolio to verify whether these FSTQ-specific institutional features generate gains or losses for stakeholders, the economy and the government. On the basis of the evidence gathered, it is shown that the FSTQ effectively helps relieve problems of capital rationing; and fosters greater labor-management cooperation that results--under certain conditions--in measurable labor productivity gains, thus engendering a net social gain. In view of these gains and of the concentration of the FSTQ investments in the riskier end of the investment spectrum, it is argued that government support of the FSTQ in the form of tax expenditures is justified and engenders a net fiscal surplus.

Despite the voluminous literature on North-South macroeconomic interactions and the key role of terms of trade variations in growth transmission from one region to another, a significant research gap persists for two reasons. First, there has been very little empirical work on testing of the relationships between growth patterns and terms of trade movements. Second, the empirical studies dedicated to testing the Prebisch-Singer Thesis (PST) focused on testing the long-run tendency for the terms of trade of primary commodities to deteriorate and neglected the joint nature of the predictions arising out of a complete formulation of PST. This dissertation seeks to properly specify the PST, provide a generalization of it to the case of imbalanced trade, and extend it to a three-region framework through a structuralist North-South model. Multiple paths of growth divergence/convergence and terms of trade deterioration/improvement emerge depending on the structural changes influencing the income-elasticity differentials. I carry out two sets of empirical analyses. First, I use aggregate data on North-South terms of trade indices to test the presence and significance of a downward trend. Second, I use panel data analysis and rolling regressions to show the evolution of income-elasticity differentials. The results suggest that the growth rates of developing countries during the 1980s declined in both absolute and relative terms partly as a result of the downward trend in terms of trade and partly as a result of income elasticity differentials reflecting the productive and technological asymmetries between the developed and developing economies. However, these structural asymmetries have not remained constant: the results show that they changed both over time and over cross-sections of different groups of countries. In general the countries that diversified towards manufactured exports had better chances of eliminating the elasticity differentials, and thus attaining relatively higher rates of growth. The cross-country study is complemented by a comparative case study of Turkey and Malaysia. The results show that industrial and trade policies, if carefully designed and effectively implemented, can counter potential costs of external market dynamics while taking advantage of the opportunities for advancing dynamic comparative advantages.

This dissertation looks at the formation of property rights institutions, using the creation of property rights in the radio spectrum in the United States from 1899 to 1934 as a case study that sheds light on both economic theories of institutional change and current policy debates over spectrum allocation. I first develop a theoretical framework for understanding the creation of property rights, recognizing that property rights can take on diverse forms that distribute the costs and benefits associated with resource use in different ways. This diversity is constrained by five economic factors—resource characteristics, technology, product markets, existing institutions, and ideologies. My historical case study provides compelling empirical support for theories developed by Douglas North and other “New Institutional” economists that stress the importance of existing institutions and ideologies in shaping property rights, placing these factors on an equal level with the first three, which have been the basis of neoclassical property rights theory. I argue that serious consideration of the roles played by existing institutions and ideologies in shaping property rights logically points to the need to redefine current theoretical assumptions about individual economic actors and to incorporate concepts of power into economic analysis. The historical component of my dissertation is based on research at six historical archives. I focus on the economic history of spectrum property rights in the United States before World War I, choosing this time period for two reasons. First, as my research shows, the activities of early commercial wireless firms of this period were crucial in shaping a unique system of spectrum property rights in the United States. Second, the importance of this period in shaping the structure and allocation of spectrum rights has gone largely unrecognized both in the existing radio history literature and in the economic literature on property rights in the spectrum. In the concluding chapter I argue that economic analyses of spectrum property rights must incorporate the distinctive enforcement issues associated with resource characteristics of the radio spectrum, and the public goods attributes of spectrum product markets. Such considerations are important in efforts to preserve and extend public and other non-profit rights not only to the spectrum but also in other areas of telecommunications.

The persistence of wage differentials--differentials that are inconsistent with both human capital theory and conventional Marxian theories of the working class--has led economists to reexamine the structure and historical development of labor markets and the labor process in the U.S. The central thesis of this dissertation is that such economic divisions result from differences in (implicit and explicit) bargaining power, and are the basis for the unequal returns to human capital that characterize segmented labor markets. In this dissertation, three forms of bargaining power are identified: (1) implicit individualistic power based on institutional attributes such as firm-specific human capital, (2) implicit collective power based on solidarity along lines of race, ethnicity and gender, and (3) explicit collective power exercised via unions. These forms of bargaining power are closely associated with occupation. Theoretical models are developed that identify the relationship between workers' power and wage determination. The models, which combine efficiency wage theory with non-cooperative game theory, show that the return to human capital is a function of bargaining power. Thus, groups of occupations with different degrees of bargaining power will exhibit unequal returns to human capital, i.e. they exhibit the characteristics of segmented labor markets. These models provide an explanation of labor segmentation that is microfounded in the process of wage determination. The theoretical wage models are tested econometrically, using wage data for individuals in 1970. The results substantiate the hypothesis that power-based occupational groups constitute distinct segments of the labor market. Limitations on U.S. wage data preclude a direct test of this hypothesis in the pre-War period. As an alternative, data from Massachusetts in the 1900s is used to econometrically demonstrate that occupational power attributes are good predictors of occupational earnings. Statistics identifying divisions among U.S. wage earners are constructed for 1870, 1910, 1950 and 1980. The resulting quantitative history of workers' power differs significantly from previous historical works, which posit labor segmentation as a characteristic of the post-World War II period. The results show that power-based divisions among wage earners existed in 1910, and strongly suggest that labor segmentation is a fundamental characteristic of the modern U.S. economy.