Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.594124
Title: Essays on human capital
Author: Feng, Andy
Awarding Body: London School of Economics and Political Science (University of London)
Current Institution: London School of Economics and Political Science (University of London)
Date of Award: 2013
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
Abstract:
This thesis entitled “Essays on Human Capital” is comprised of three essays on various aspects of human capital and its effects on firms and labor markets. Chapter 1 provides an overview. In Chapter 2 we estimate the effects of human capital on firm-level management practices. We adopt an instrumental variables strategy to overcome the potential endogeneity of human capital. Starting with data on management practices from the World Management Survey, we geocode the locations of more than 6,000 manufacturing plants in 19 countries. Then, we calculate driving times to universities in the World Higher Education Database. Using distance as an instrument for human capital, we estimate that every one standard deviation increase in the share of workers with a university degree leads to 0.5 of a standard deviation improvement in management. These findings are robust to a battery of checks and a placebo instrument using distances to world heritage sites. We show that both managers’ and non-managers’ human capital matter. In Chapter 3 we estimate the effects of university degree class on initial labor market outcomes. We employ a regression discontinuity design which utilizes university rules governing the award of degrees. We find sizeable and significant effects for Upper Second degrees and positive but smaller effects for First Class degrees on wages. A First Class is worth roughly 3 percent in starting wages which translates into $1,000 per annum. An Upper Second is worth more-7 percent in starting wages which is roughly $2,040. We interpret these results as the signaling effects of degree class and provide evidence consistent with this. Finally in Chapter 4 we study the labor market effects of increased automation. We build a model in which firms optimally design machines, train workers, and assign these factors to tasks. Borrowing concepts from computer science and robotics, the model features tasks which are difficult from an engineering perspective but easy for humans to carry out due to innate capacities for functions like vision, movement, and communication. In equilibrium, firms assign low-skill workers to such tasks. High skill workers have a comparative advantage in tasks which require much training and are difficult to automate. Workers in the middle of the skill distribution perform tasks of intermediate difficulty on both dimensions. When the cost of designing machines falls, firms adopt machines mainly in tasks that were previously performed by middle-skill workers. Occupations at both the bottom and the top of the wage distribution experience employment gains. The wage distribution becomes more dispersed near the top but compressed near the bottom. As design costs fall further, only the most skilled workers enjoy rising skill premiums, and an increasing fraction of the labor force is employed in jobs that require little or no training. The model’s implications are consistent with recent evidence of job polarization and a hollowing-out of the wage distribution. In addition, the model yields novel predictions about trends in occupational training requirements that are consistent with evidence we present.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.594124  DOI: Not available
Keywords: HC Economic History and Conditions ; HD Industries. Land use. Labor ; HD28 Management. Industrial Management
Share: