Georgia State University ScholarWorks @ Georgia State University Economics Dissertations Spring 5-6-2019 Essays on Migration, Energy Use, Emissions, and School Assignment Cody Reinhardt Follow this and additional works at: https://scholarworks.edu/econ_diss Recommended Citation Reinhardt, Cody, "Essays on Migration, Energy Use, Emissions, and School Assignment." Dissertation, Georgia State University, 2019.edu/econ_diss/161 This Dissertation is brought to you for free and open access by ScholarWorks @ Georgia State University. It has been accepted for inclusion in Economics Dissertations by an authorized administrator of ScholarWorks @ Georgia State University. For more information, please contact scholarworks@gsu. ABSTRACT ESSAYS ON MIGRATION, ENERGY USE, EMISSIONS, AND SCHOOL ASSIGNMENT By CODY KARL REINHARDT 4/1/2018 Committee Chair: Dr.
Spencer Banzhaf Major Department: Economics This dissertation has two essays. The first studies migration patterns in the U. and the relationship between migration patterns and energy use and carbon emissions. It uses a two-city model of energy use and household migration to analyze emission implications from city level green policies.
Per-household emissions are calculated for the largest 49 MSA’s in the U. and data on migration patterns used to assign substitute locations to mi- grating households. Results show large differences in net carbon emissions from migration, which has implications for a wide range of policies affecting migration decisions. The second essay studies how school quality is assigned to properties through various methods.
It first replicates methods in the literature, such as assignment by distance and district means, and adds new methods to assign measures of school quality to census blocks. Next, these assignments are compared to a new dataset of school assignment to determine accuracy. Both distance matching and assignment by district means are shown to be relatively inaccurate methods of assignment. The accuracy also varies over space and district size.
ESSAYS ON MIGRATION, ENERGY USE, EMISSIONS, AND SCHOOL ASSIGNMENT BY CODY KARL REINHARDT A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Andrew Young School of Policy Studies of Georgia State University GEORGIA STATE UNIVERSITY 2019 ACCEPTANCE This dissertation was prepared under the direction of the candidate’s Dissertation Committee. It has been approved and accepted by all members of that committee, and it has been accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Economics in the Andrew Young School of Policy Studies of Georgia State University. Dissertation Chair: Dr. Spencer Banzhaf Committee: Dr.
Kyle Mangum Dr. Daniel Kreisman Dr. John Winters Electronic Version Approved: Dr. Sally Wallace, Dean Andrew Young School of Policy Studies Georgia State University May, 2019 Contents 1 U.
Internal Migration Networks, Energy Use, and Emis- sions 1 1.2 The Two-City Model. 32 2 Accurate Assignment of School Quality to Properties 49 2.2 School Quality Assignment Methods. 77 iv List of Figures 1 Representative Migration Cities. 11 2 Wharton Regulation Index and Carbon from In-Migration.
19 3 Atlanta Representative In-Migration City 2008. 21 4 Atlanta In-Migration Carbon Differentials, 2008. Representative In-Migration City 2008. 24 6 Washington DC In-Migration Carbon Differentials, 2008.
25 7 San Antonio Representative In-Migration City 2008. 27 8 San Antonio In-Migration Carbon Differentials, 2008. Out-Migration City 2008. Out-Migration Carbon Differentials, 2008.
In-Representative Carbon Differential, 2000 43 12 Washington D. In-Representative Carbon Differential, 1992 45 13 San Antonio In-Representative Carbon Differential, 2000. 45 14 San Antonio In-Representative Carbon Differential, 1992. 46 15 Atlanta In-Representative Carbon Differential, 2000.
46 16 Atlanta In-Representative Carbon Differential, 1992. 47 17 SABINS Data Coverage. 56 18 SABS Data Coverage. 58 19 SABS Boundaries and Schools in Detail.
Unified School District. 65 21 Hershey Public School District. 66 22 Kernal Density for Assignment Error, Math Scores. 76 23 Kernal Density for Assignment Error, Student-Teacher Ratio.
80 24 Kernal Density for Assignment Error, Free-Reduced Lunch. 81 v 25 Kernal Density for Assignment Error, Reading Scores. 82 List of Tables 1 Representative Out-Migration Cities 2008. 36 2 Representative Out-Migration Cities 2000.
37 3 Representative Out-Migration Cities 1992. 38 4 Representative In-Migration Cities 2008. 40 5 Representative In-Migration Cities 2000. 41 6 Representative In-Migration Cities 1992.
42 7 Regression Results for Wharton Regulation Index. 43 8 Summary of Assigned Carbon Per Household. 44 9 Summary Statistics for School Quality Data, School Level. 60 10 SABINS 2009 Summary Statistics, School Districts.
61 11 Match Rates for Blocks to Schools. 62 12 Regression Results for District Block Match. 68 13 Correlation for Student-Faculty Ratio, Sample 1. 71 14 Correlation for Free and Reduced Lunch Proportion, Sample 1 72 15 Correlation for Math Scores, Sample 1.
73 16 Correlation for Reading Scores, Sample 1. 74 17 Correlation for Student-Faculty Ratio, Sample 2. 78 18 Correlation for Free and Reduced Lunch Proportion, Sample 2 78 19 Correlation for Math Scores, Sample 2. 78 20 Correlation for Reading Scores, Sample 2.
Internal Migration Networks, Energy Use, and Emissions 1.1 Introduction Climate change as a result of carbon emissions is a highly studied and broad topic in the economics literature. As noted in Glaeser and Kahn (2010), a significant proportion of US carbon emissions come from household energy use, and urban structure plays a prominent role in how much energy house- holds consume. Mangum (2017) and Glaeser and Kahn (2010) have shown that cities vary greatly in per household levels of emissions, with the high- emission U. cities having nearly twice the per-household emissions as the low-emission cities.
Glaeser and Kahn examine differences in urban struc- ture and both within city and between city variation in household energy use. This paper extends this literature by using historic internal migration data to examine the role migration plays in the total emissions for the U. Given the plethora of local policies on housing and zoning, and the popu- larity of local green regulations, it is highly unlikely that emissions will be optimally taxed. As noted by Glaeser and Kahn, even an otherwise perfectly calibrated Piguvian carbon tax is not sufficient for optimal location decisions in the presence of local policies or incentives which restrict development in green areas and subsidized development in less green areas.
In reality, the U. has many such policies and incentives. According to Glaeser, “By re- stricting new development, the cleanest areas are pushing development to areas of higher emissions” (Glaeser and Kahn). So migration will play a key role in how optimal emission decisions are made from a country per- spective, because how the population is distributed and moving among the cities of various emissions levels affects the total country level of emissions.
1 As household migrate between cities, they change their housing consump- tion, carbon content of electricity and heating, and driving patterns as they change locations. Any local policies directly or indirectly taxing carbon emis- sions would have to consider the potential migration effects on emissions an how movement of households to and from their neighbors contributes to the national carbon account. Policies in all of the cities are important, as well as a city’s location in the sense of it’s largest migration neighbors. While Mangum (2017) considers simulations of national level policies, this paper focuses on local policies with migration effects following historic migration patterns.
The purpose of this paper is to examine the role migration plays in the total carbon emissions in the U. This paper extends a two-city model first developed in Glaeser and Kahn (2008). It does this by using city pairs constructed from data on MSA emissions and MSA-to-MSA migration data. This will represent the migration effect of the MSA by weighting its migrants with the per-household emissions of their destination MSA.
Each MSA will thus have different migration effects, for both out- and in- migration, due to their place in the migration network and the greenness of substitute cities in their part of the network. The paper proceeds as follows. Section 2 presents the two-city model and the generation process for the representative migra- tion city. Section 3 describes the data used in the paper.
Section 4 details the results and implications. Section 5 concludes and discusses opportunities for further research.2 The Two-City Model This section expands on the two-region model presented in Glaeser and Kahn (2008). The original model is introduced and then expanded by consid- 2 ering the changes on energy use. The model contains two regions (which will be defined as cities in this paper) where individuals are free to move between them to maximize utility.
They maximize utility by choosing location and energy service consumption. The individual wishes to live in the location where they can get the most utility from energy service consumption, which depends on the price of energy services and that location’s utility function with respect to energy. For example, heating and cooling expenses can be expensive in an area with a very mild climate and total energy service con- sumption could be lower and yield a higher total utility. With income and total population being held constant, the model shows that the distribution of population between regions with different energy prices, energy uses, and external costs of energy service consumption affects total utility.
New zoning or tax policies cause a movement between cities as well as a change in energy service consumption within. The two regions are expanded from abstract areas to constructed empir- ical areas using migration data to represent the migration effect of a city. The model is presented and then followed by the representative migration city construction. The two-city model begins with individuals maximizing a quasi-linear utility function Yi − PiH − (PiE + t)Ei + tÊ + Vi (Ei ; Xi ) − C(N Ê) where Yi is income, PiH and PiE are prices of housing and energy services for city i; t is an energy use tax; E is energy use in city i; Ê is the national average energy consumption; Vi (.) is a function for city-specific benefits from energy services; Xi is a vector of exogenous attributes for location i; C(N Ê) is the external cost of energy use by the whole country, which can be thought of as the national contribution to climate change; and N is pop- 3 ulation.
Note that in modeling energy services, I am looking at the cost of, e., maintaining a given temperature in the home, which will be a function of energy prices but also house size, weather, and so forth. Finally, note that the tax is revenue neutral, since individuals are receiving a lump sum rebate of tÊ. Next, each city i has QFi identical employers, with revenues f (.) increasing and concave in the the number of people hired. Each city has builders QB i , with costs k(.) increasing and convex in buildings constructed.
Now wage income is f 0 ( QNFi ), or the marginal revenue product of labor (MPL), i and housing cost is k 0 ( QNBi ), the marginal cost of supplying housing. Individ- i uals hold equal rights to all business profits. The two equilibrium conditions are as follows: individuals choose privately optimal energy consumption Ei∗ to maximize their utility, so PiE + t = V1 (Ei∗ ; Xi ), with V1 (Ei∗ ; Xi ) being the first derivative of V (.) with respect to E. The next condition is a locational equilibrium, so f 0 ( QNFi )−k 0 ( QNBi )−(t+PiE )Ei∗ +V (Ei∗ ; Zi ) must be equal for all i i cities.
Individuals in this model are identical, and the social welfare function used is additive: X Ni Ni QFi f ( F ) − QB E i k( B ) + Ni (V (Ei ; Xi ) − Pi Ei − C(N Ê)) (1) i Qi Qi So this yields two first order conditions.