Three Essays In Applied Regional Welfare Analysis: The Effect Of Education Attainment And Local Interaction On Wealth Inequality In A Spatial Economy, Time Inclusion In Structural Path Analysis, And An Econometric Examination Of Impacts Of Monetary Policy On The Welfare Of Different Income Groups
The first essay (Chapter 1) explores the importance of spatial interactions in an overlapping generation model using an Agent-Based Model (ABM) approach. Incorporating geography into the original Galor-Zeira's model (1993) and allowing heterogeneous agents to interact locally, we show that social interactions play an important role in determining the long-run welfare and wealth distribution. This model shows the impact of endogenous wages and local interactions on agents' decision to invest in human capital. Our simulation reveals that neighborhood interactions lead to changes in the system's steady-state behaviors. In particular different strength of agents' interaction produces different outcomes in terms of the number of educated people and wealth inequality. This might explain persistent inequality in agents' wealth across locations. Policy implications are discussed. The second essay (Chapter 2) introduces the time elements of structural path analysis (SPA). Structural path analysis, which was first introduced by Defourny and Thorbecke (1984), broke down the global multipliers of the social accounting matrix (SAM) into direct influence and total influence. The introduction of time elements into SPA has enabled policymakers to estimate the range of time required for a shock to travel from its origin to its destination. Using the 2008 Indonesian SAM with a focus on the agricultural and manufacturing sectors, this study successfully introduces time into the SPA framework and estimates the possible time range within which a shock from the agricultural and manufacturing sectors will impact different households. The third essay (Chapter 3) explores impacts of monetary policy on the welfare of people in different income groups in Indonesia with a dynamic demand system. In the model, income groups adjust their expenditures shares in response to changes in commodity prices and aggregate expenditure levels. These adjustments are taken to be functions of the rate of change in the flow of financial services, which is affected by the rate of growth of the money supply (M2). Results of model estimation and deterministic numerical simulations conducted with the estimated model suggest that the welfare of the low-income group is affected more by monetary policy than is the welfare of the high-income group.
Agent-based model; wealth inequality; dynamic econometrics; monetary policy; structural path analysis; SAM
Donaghy, Kieran Patrick
Thorbecke, Erik; Mansury, Yuri Surtadi
Ph.D. of Regional Science
Doctor of Philosophy
dissertation or thesis