2.3Purpose of the Study

The previous sections revealed several issues related to the role of information technology in the United States. They can be grouped into three distinct issues: the productivity paradox, the spatial dimension of IT and the relationship between IT and income inequality. Here are some interesting questions related to the issues: How can we explain the productivity paradox? Do computers and other IT equipment really enhance productivity? If so, then how do they do it, i.e. what is the mechanism by which IT affects productivity? Are the returns to IT capital higher than the returns to “traditional” capital? What is the contribution of IT capital growth to output and productivity growth? Is the productivity paradox only a concern at the aggregated national level? Are the relationships between IT and productivity also paradoxical at the industry level? In particular, are there differences in the returns to IT capital among industries that could partially explain the national paradox? At the state level, are there differences in the returns to IT capital across states that could also explain the national paradox? These are the questions being addressed in the first part of this dissertation (chapters 3, 4 and 5). Its purpose is to shed some light on the productivity paradox debate. In order to do so, I will build a dataset disaggregated at the detailed industries level by state, and analyze econometrically the relationships between IT capital stock and productivity. I will base my analysis on the framework constructed by Lehr and Lichtenberg (1999). I expect to find a positive and significant contribution of IT capital to productivity growth across industries and states. I intend to show why the aggregation at the national level has led to a paradox.

The second part is aimed at answering questions about the spatial aspect of information technology in the United States. For instance, why do the returns to IT capital differ across states? How to evaluate the externality effects associated with employment location? Are the agglomeration or congestion effects important? What is the role of the density of IT (namely the number of IT employees by county)? Does the location of IT activity affect productvitity? I intend to answer these questions based on my dataset at the county level in the United States. My work is inspired by the study of Hall and Ciccone (1996) who analyzed the effects of the density of economic activity on productivity.

Finally, I will focus on the ambiguous relationships between information technology and income inequality. Have computers and other information technology increased inequality? If so, by which mechanism has this occurred? What are the variables affecting income inequality? Is there a positive and significant relationship between the information technology intensity in a given state and its level of income inequality? To answer these questions, I will use an original framework based on the regressions of states Gini coefficients on different independent variables representing the importance of IT in that state (such as the level and intensity of IT capital stock).

Each of these three essays presents the same structure composed of three levels. A first chapter (3, 6 and 9) discusses the literature relevant to the subject and previous findings. A second chapter (4, 7 and 10) describes the methodology adopted to tackle the problem: which model(s) will be used? What variables? How the dataset is constructed? Then, a final chapter (5, 8 and 11) presents the results and discusses the outcomes. Finally, chapter 12 will summarize and interpret the main findings of this study, indicate fruitful directions for future research and conclude the dissertation.