Three Essays on Amenities, Earnings, and City Prices

Abstract

This dissertation examines the interaction between amenities, earnings, and city prices. The first essay uses three different methodologies to investigate whether, given a fixed level of amenities, workers are fully compensated, in terms of higher wages, for higher price levels across cities. In the first, I use the overall city price level and aftertax earnings as endogenous variables in a two-stage least squares (2SLS) system and find that owners and renters are not fully compensated for higher city price levels. In the second, I split city price levels into housing prices and non-housing goods prices and use housing prices and aftertax earnings as endogenous variables in a 2SLS system. Using this method, I find that highly-educated homeowners and renters are fully compensated for higher housing prices while less-educated homeowners and renters are not fully compensated for higher housing prices. I also find that all homeowners and renters of all education levels are fully compensated for higher non-housing good prices. Finally, I assume that homeowners and renters are fully compensated for higher nonhousing goods prices and use housing prices and aftertax earnings as endogenous variables in a 2SLS system. Using this method, I again find that highly-educated homeowners and renters are fully compensated for higher housing prices while less-educated homeowners and renters are not fully compensated for higher housing prices. In addition, I determine implicit prices for 13 different amenities for both owners and renters with different education levels. While I am not the first to calculate these implicit prices, I am the first to differentiate these implicit prices among education levels and ownership status. In the second essay, I use the implicit price results from the first essay to create a city ranking. There are two main approaches in the city ranking literature. The implicit price of amenity approach uses implicit amenity prices as weights that are multiplied by the amount of each amenity in each city. The sum of the "market values" of a city's amenities can then be used to create a city ranking. The real wage approach involves finding the logarithmic difference in nominal wages across areas and subtracting this from the logarithmic difference in housing prices across areas, using either a rent-based or housing-value based index. The idea behind this approach is that, after typical housing characteristics and worker characteristics are accounted for, the differences in rents and wages must reflect differences in local amenities. I use similar methodologies to both of these approaches, but I improve upon what I view as shortcomings in both strands of the literature. First, I include the effect of educational attainment, income inequality, and job growth. Second, I look at how city rankings differ by level of education and by ownership status. Third, I add new components to the existing literature on firm rankings. All of these additions give a richer, more accurate view of how workers and firms view cities. In the third essay, I look at the same issues from the first essay, but I shift the focus to how the results differ by gender, marital status, and the presence of children in the household. The first question deals with how well compensated people are, in terms of higher wages and better amenities, for higher housing prices. I find that single people fare better than married people, and people without children fare better than parents. Also, I find little difference between men and women overall. The second question deals with the implicit price of amenities. I find significant differences in the price people are willing to pay for amenities by gender, marital status, and the presence of children in the household.