Home » 2013 Categories » 2013 – Literature Review

Category Archives: 2013 – Literature Review

School Quality and Property Values

By David Wang LR_WangDavid


School Quality and Property Values

The American education system spans communities of extremely diverse populations, across many socio-economic and ethnic lines.  It is likely that the successes of students in these communities hinge on the combination of the students’ innate intelligence, living environment, and school quality.  Standard models suggest and many have provided empirical evidence that housing prices have a correlation to this school quality.  However, since these factors could be interrelated, to isolate and identify the actual effects of school quality on property values requires careful consideration of student, house, and neighborhood-specific attributes.  It is also important to note how school quality is measured.  With the passage of the No Child Left Behind Act of 2001 and the shift away from using per pupil expenditures as a proxy for the quality of education, attention has been drawn to the use of standardized test scores as the marker of education quality.

Recent literature reconfirms the positive correlation of school performance on house prices, using basic hedonic models with added controls for test scores of local schools.  However, in older papers, the authors have difficulties controlling for neighborhood characteristics that are correlated with the test scores and house prices.  Black (1999) develops a new method for assessing school quality by using attendance district boundaries to account for neighborhood characteristics.  This method allows her to compare school to school differences in test scores with house prices.  Crone (2006) uses a model on a full unrestricted sample that allows for testing of house price and test score relationships on both a school and district level.  In addition, he adapts Black’s boundary model to allow for this district level analysis.  In contrast to Black, Crone argues that it is a district-wide educational quality, not individual school quality that affects house prices.  Finally, Clapp, Nanda, and Ross (2007) also consider Black’s model, but instead use a time-based fixed effects model over the period from 1994 to 2004 to control for the neighborhood characteristics.  Despite using different methods, all three papers agree that a positive correlation exists between school test scores and housing prices.

Black’s (1999) measurement of differences across attendance district boundaries enables the use of fixed effects in her model.  This district boundary is the line that separates the respective attendance areas of schools.  This line provides a discrete point at which standardized test scores should change.  However, the line may run through continuous neighborhoods, allowing Black to compare any sudden jump in test scores with houses that are situated in similar neighborhoods.  By using dummy variables to account specifically for the districts, Black avoids the omitted variable biases of property taxes, public goods, and neighborhood characteristics.  Using MEAP testing data from Massachusetts elementary schools, Black focuses on the fourth grade level.  Under the basic, unrestricted model, she must control separately for house level characteristics, distance from the CBD, in addition to other school quality characteristics, such as per-pupil expenditures.  She finds that per-pupil expenditure is positively correlated with house prices while higher pupil/teachers ratio is negatively correlated with house prices.  Nevertheless, the crux of the problem involves the unobservable characteristics of a neighborhood.  Black examines different subsets of her data, restricting the samples to houses nearer and nearer the boundary and increasing the probability that the houses on opposite sides of the boundary differ in only the elementary school quality.  Her study reveals that if neighborhood characteristics are not carefully controlled, the marginal value of school quality as measured by test scores on housing prices will be overestimated. Black concludes that parents will pay higher house prices for better schools, but does not examine whether there exists a district level effect of school quality on prices.

Newer researchers incorporate Black’s boundary model and conclusions as supplements to their models.  However, unlike Black, Crone (2006) argues that home buyers actually value local public education at the district level rather than the neighborhood school level.  Using fifth and eleventh grade Pennsylvania System of School Assessment (PSSA) data from Montgomery County, Crone makes findings that differ from Black’s.  While Black argues that differences on an individual school basis affect home prices, Crone claims otherwise.  Crone argues that for fifth grade test scores, differences are only significant on the district-level.  In fact, he finds that fifth grade test scores are better predictors of house prices than eleventh grade scores.  Perhaps this discrepancy could be attributed to the location of families with young children and the subsequent lack of relocation as the children grow up.  Crone’s differing results could also be due to his use of the full sample rather than a boundary restricted sample.  Crone’s more comprehensive dataset allows him to make district level regressions, while Black’s dataset is restricted to individual schools.

Crone’s study also provides additional factors that may affect school quality and thus house prices.  For example, class size is not significant at the elementary school level, but it makes a significant difference at the high school level.  By incorporating this measurement into the main model, Crone reduces the chance of omitted variable bias from Black’s neighborhood fixed effects model.  The neighborhood fixed effects do not account for differences in the schools, such as per-pupil expenditure or class size, of which the latter was not included in Black’s model, which only seeks to explain the impact of school quality differences.  As an additional test, Crone uses Black’s boundary method to estimate the effect of both school and district test scores on housing prices.  He finds that with this smaller sample, there is no significant coefficient on fifth grade scores, further conflicting with the results given by Black.  However, on the high school level, the results become more significant with the smaller sample with boundary dummies.  This result differs from the result when controlling for detailed characteristics in the model with a full sample.  Finally, Crone’s study also finds that per-pupil expenditures do not affect the house prices above their effect on student test scores or achievement. Overall, Crone brings the conclusion that school district quality should be considered over the quality of individual schools when determining the effect on house prices.

Clapp, Nanda, Ross (2007) introduce a twist on the examination of test scores and housing prices by suggesting that the quality of school districts is a function of both test scores and demographic composition.  Since people tend to use the most accessible signals to judge school quality, they often rely on the demographics of a school, which are very visible.  Thus, Clapp examines the significance of the test score and demographic composition effects on house prices.  Like Black and Crone, Clapp also finds a statistically significant, though very small, effect of test scores on house values.  Clapp also agrees with Black’s finding that failing to control for unobservable characteristics in the neighborhood leads to overstatement of the test score effect.  However, Clapp extends this argument by also including the effects of race percentages on home prices.  He finds that an increase in percent African-American and percent Hispanic leads to a decline in property values.  Nevertheless, over this time period, people appear to be placing more importance on test scores and less on demographics when evaluating school quality.

Of the three papers, Clapp’s is the only one to use a time-based fixed effects model.  While the studies of Black and Crone use averages of a single three-year period and district boundaries as fixed effects, Clapp instead exploits the cross time variation in the 1994-2004 panel data to separate school attributes from neighborhood quality.  Clapp also incorporates additional neighborhood fixed effects by comparing sales occurring in different neighborhoods, but the same school district.  This combination of time variation based identification strategy and also neighborhood fixed effects should yield more accurate estimates than either strategy alone.  However, one downside to Clapp’s method that does not appear with Black’s or Crone’s is the possibility of unobservable changes over the sample’s long time period.

Through these three papers, we see a wide variety of techniques used to analyze public school test scores and house prices, yet arrive at the conclusion that standardized test scores do impact housing prices.  In these papers, however, we assume that district boundaries are fixed and that students must attend schools in their attendance zone.  With the rise of charter schools, students no longer are limited to the public schools near their homes.  As the population of students going to charter schools increases, we may begin to see a declining importance of neighborhoods under the models discussed in the three papers reviewed here.  It may be worthwhile to examine the effects of charter schools on home prices, both in the area of the school and of the students.



Works Cited


Black, Sandra E. “Do Better Schools Matter? Parental Valuation of Elementary Education*.” Quarterly Journal of Economics 114.2 (1999): 577-99. Web. 7 Feb. 2013.

Clapp, John M., Anupam Nanda, and Stephen L. Ross. “Which School Attributes Matter? The Influence of School District Performance and Demographic Composition on Property Values.” Journal of Urban Economics 63.2 (2008): 451-66. Web. 7 Feb. 2013.

Crone, Theodore M. “Capitalization of the Quality of Public Schools: What Do Home Buyers Value?” Working Paper Series, Federal Reserve Bank of Philadelphia (2006): n. pag. Statistical Insight [ProQuest]. Web. 7 Feb. 2013.

Gastronomic Curiosity and Racial Migration in Durham


            A life-long Durham resident, I seek to learn more about the cultural and historical roots of my home. Durham has a unique reputation as an up-and-coming city worthy of New York Times appraisal and the curiosity of hipsters nationwide. My Duke friends tease that they wouldn’t venture off campus for fear of Durhamite encounter, crime or lackluster dining. I find Durham to be quite the contrary: vibrant, growing and delicious. A self-proclaimed gourmet, I have become more and more infatuated with the Durham dining scene through college. Ethnic restaurants are more prominent than I recall from my childhood, and the quality of dining has unambiguously improved. Downtown Durham more closely resembles Carrboro, with restaurants supporting local farmers and vegan diets, than the grungy urban sprawl it once was. My curiosity is endless. How does a shop like Scratch exist in the same city that scares pastry-loving Duke students? Who created the concept of Watt’s Grocery? Why did Magnolia Grill close? Who moved here to open these eateries, and who are the people keeping them in business today?

My obsession with food aside, the question can be initially expanded to national demographic trends—what factors cause Durham to be an urban hub in the South? Historically speaking, the South is not a particularly attractive destination for business. Since before the civil rights movement, the South has struggled to maintain a constant influx of business. Wright (1987) posits a model of Southern out-migration explained by rocky assimilation with Northern industrial production. A dramatic decline in low-income farm laborers forced a majority of the ethnic population out of the South.

However, Frey and Liaw’s model of racial migration trends suggests the possibility of an established minority network in the South. Their model of out-migration and destination selection suggests that minority groups tend to stay in areas that have large populations of their minority group and move toward areas with similarly established minority networks. Movement into the South has been striking across all minority groups studied, a testament to the growing economy and employment opportunities in the South. Combining Wright’s theory with Frey and Liaw’s model, I see the potential for Durham’s economic success to be explained by a flight to economic growth combined with a return of black laborers to what could be considered an informal ethnic network, a historical home base.

Zhao’s model predicting discrimination in real estate suggests that brokers of the same racial-ethnic status as their client tend to discriminate less, a testament to the cyclical potential of movement into the South. As immigration continues, the ethnic network grows and racial similarity becomes a larger factor in attracting and retaining racial minorities.

These three surveys open a discussion of the causes of Southern economic improvement but do not fully explain my question of why there are so many boutique eateries in Durham. In further investigations, I hope to answer the following additional questions: What is the ethnic makeup of migrants to Durham? Are minorities a majority in Durham? What is the demographic profile of Durham small business owners? Does the desire for racial similarity drive migration and economic growth, or is it the economic growth that fosters migration? I hypothesize that the economic growth in the South combined with its attractiveness to immigrants and minority groups has created an environment that fosters small business and boutique ethnic eateries.

Model Particulars

            Wright outlines a model more qualitative than the econometric models of Zhao and Frey and Liaw. Wright poses the question: why can’t macroeconomic factors explain the homogenization of Northern and Southern US economies, and why did the Southern economy finally become assimilated with the North? Historically, he notes that labor flows have in general occurred in an East to West direction rather than North to South, a trend “…rooted in certain geo-agricultural continuities, such as familiarity with seeds, crops, livestock, and climate.” (Wright, 164) As the North continued to receive international immigrants and technological improvements, the South was geographically isolated from this influx of productivity. The dynamic was compounded by the fact that Northern wage rates far exceeded those in the South, making economic assimilation more of a daunting task. As the North grew, Northern producers had little incentive to expand into the South because it was “…much cheaper to utilize existing channels or expand them incrementally than to lay out the large fixed cost that would have been required to redirect the established lines of the market.” (Wright, 164) The South became stagnant, despite the potential for American unity provided by the booming machine tool industry.

The largest constraint to assimilation was the large gap between Northern and Southern wage rates, and as national wage policies were implemented the gap closed. This was overall productive for the Southern economy—in-migration of educated Northerners increased, out-migration of farm laborers increased and integration began. However, this was harder on low-income workers and “…the majority of the departing farm population had few options other than leaving the South.” (Wright, 172) Particularly hard hit was the black population due to the combined reduction in agriculture and tobacco manufacture—the black share of labor force more than halved in the former Confederacy from 1930 to 1960. On the flip side, Northern migration created opportunity for education, “…yet the same migration channeled other blacks into the high-unemployment ghettos which if anything have worsened with the passage of time.” (Wright, 175)

Frey and Liaw (2005) paint a contrary picture 20 years later. Their extensive study of the effect that racial/ethnic background has on migration patterns concluded that minority migration is occurring in a general Southward direction, particularly within the black population. Frey and Liaw seek to understand the effect of two separate theories of migration. The cultural constraints theory suggests that migration networks are shaped by racial and ethnic attachment. Spatial assimilation, on the other hand, suggests that education and socioeconomic status become larger determinants of migration in the upper-middle class, even within minority populations.

Frey and Liaw consider two components of migration separately—decision to migrate out of a location (“out-migration”), and the choice of destination location. They formulate a two-level nested logit model to show the effect of observable explanatory variables on the probability of out-migration and destination choice at the state level.[1] Study conclusions were focused predominantly on migration patterns into and out of California, but will be omitted in the context of this survey.

Overall US out-migration was found to be reduced by the cultural constraints hypothesis, with education level playing a minimal role—“…these [racial similarity] constraints do not play a stronger role for less educated than more educated members of these [coethnic] groups.” (Frey and Liaw, 236) Destination selection is also affected by racial similarity, with “…positive effects on migrant destination selections for each race-ethnic group” (Frey and Liaw, 241). However, distance from previous location, contiguity and population size along with employment growth rate affect destination selection more strongly than racial similarity.

Two of their findings provide a possible explanation for minority (and black in particular) migration back to the south and are thus most relevant with respect to my topic: first, the finding that “both persons born in a different state and the foreign-born are more likely to out-migrate than persons born in the same state” (Frey and Liaw, 240) and second, the fact that destination selection is positively affected by racial similarity. Aside from cultural constraints and spatial assimilation, the human capital investment theory of migration is supported in the finding that less out-migration occurs as employment growth rate and per capita income increase. This may explain the success of the Southern economy, as it is able to attract and retain laborers.

Zhao (2005) investigates the question of whether number of homes showed by a broker varies with the homeseeker’s race, a proxy for racial discrimination in the housing market. He conducted a paired experiment in which two auditors of similar gender and age but different minority status (one white and one minority) were assigned similar income levels, marital status and parental status and sent to the same real estate agency to be shown available houses. Zhao built off of Page’s 1995 Poisson model with fixed effects, which established a relationship between auditor characteristics and discrimination. This study did not predict discrimination using a more complex model that incorporated auditor and agent characteristics and interaction terms. Zhao expands Page’s basic model to include visiting order, agent characteristics, actual auditor socioeconomic characteristics, as well as interaction terms between race and auditor characteristics, agent characteristics, house value, neighborhood characteristics, month of visit and site of home. He interprets each variable’s coefficient as the effect of the variable on the number of houses shown to the auditor, testing each coefficient for significance in the complete model (including all variables and interaction terms above).

Zhao additionally hypothesizes three potential causes of discrimination. First, broker’s prejudice, summarized as “distaste for minorities.” (Zhao, 135) White customer’s prejudice is defined as discrimination by brokers based on perceived desires of whites in the local community. For instance, if a broker has a large base of white customers he or she might act to satisfy their needs rather than the minority client. Finally, statistical discrimination involves prediction of minority behavior based on the probability of a transaction occurring. If the broker believes a minority buyer would not be interested in a residential area with a large white population or would not be able to pay for a home, he or she may be less likely to show the house.

Focusing on the black/white paired samples, Zhao finds that the customer prejudice hypothesis is true to the extent that discrimination decreases once the share of black residents increases. Additionally, discrimination increases with percent of owner-occupied homes and with value of house. Black homeseekers are shown overall 30% fewer homes than white homeseekers, which reflects an increase in discrimination of 12% since 1989 (Zhao, 144). Though the South is quickly re-establishing itself as an economic hub, there may be evidence of racial discrimination that limits economic development in the real estate market.


Bo Zhao, 2005. “Does the number of houses a broker shows depend on a homeseeker’s race?” Journal of Urban Economics 57(1): 128-147.

Gavin Wright. 1987. “The economic revolution in the American south.”  Journal of Economic Perspectives 1(1): 161-178.

William H. Frey and Kao-Lee Liaw, 2005. “Migration within the United States: role of race-ethnicity,” BWPUA 2005: 207-248.

[1]This includes race, immigrant status, age, immigration rate, employment growth rate, population size and housing value, along with a number of other variables and interaction terms


Literature Survey: The Effect of State Economic Development Incentives on Regional Economic Growth

by Zhe Zhao Zhao_Zhe_Literature Survey_Edited


Many American states have created tax incentive programs to maintain businesses, attract businesses from other states, and stimulate new start-ups. Tax-related incentives for business began in colonial times and have increased over time. According to Chi and Hofmann (2000), the number of states with tax incentives for businesses steadily increased since the 1980s. For example, 24 states offered tax incentives for job creation in 1984; 43 states offered those in 1998. R&D tax incentives were offered to businesses by 9 states in 1974 and by 39 states in 1998 (Chi & Hofmann, 2000). The reason for the increasing use of business incentives is that states see positive effects from these programs.

States believe that tax incentives are good fiscal and economic tools. Public officials create economic development programs to influence firm relocation and expansion, or rescue failing businesses, or protect them against competition, or start-ups (Burnier, 1992). The governments perceive tax incentives as revenue foregone because they are not cash paid-in. States view that benefits of tax incentives outweigh costs in the long-term (Buss, 2001). In spite of predicated benefits and billions spent annually, states do not evaluate tax incentive programs (Bartik, 1991). Buss (2001) argues that most business tax incentives are the outcome of interstate competitions to attract businesses from other states. Are tax incentives good economics or simply politics?

The broad use of incentives has generated interest in the effects of incentives on employment growth and regional economic growth. In this overview, I select several most cited and recent research works to illustrate the effect of tax incentive programs. How do states implement incentives? Do tax incentives induce regional economic growth? How effective are tax incentive programs? By reviewing the literature, I capture some answers to these questions. Researchers have utilized different methods to evaluate the costs and benefits of tax incentives. A good number of literature shows that certain types of tax incentive programs have positive effect on local job growth (Luger & Bae, 2005; Billings, 2008; Douglas & Paulo, 2008; Bartik, 1991). But many studies find an ambiguous relationship between tax incentives and economic growth (Fisher & Peters, 1997; Gabe and Kraybill, 2002). This review synthesizes these findings, methods, and data used. Because of the complications of evaluation and paucity of data, many sub-questions still need to be studied in- depth.

Types of Tax Incentives

States offer incentives for businesses under two main categories: tax exemptions and financial incentives. Within 15 tax exemption programs, the most commonly used tax exemptions are on corporate income, land and capital investments, raw materials and equipment in manufacturing, and creation of jobs. Financial incentives offer cheap financing for 16 kinds of business activities, including bond financing, loans for building constructions and equipment, and financing aid for plant expansion (Chi & Hofmann, 2000). The literature discussed in the review cover a broad range of tax incentives. For example, the case study of North Carolina by Luger and Bae (2005) focuses on tax credit for job creation, machines and equipment, central administrative offices, and R&D. Studies of BMW plant in South Carolina (Douglas & Paulo, 2008) and enterprise zones in Colorado (Billings, 2008) show that targeted tax incentives attract private firms to a new location, stimulating the economy and creating jobs, through multiplier effect.

Economic Growth Correlation

The classic theory to demonstrate the economic growth from tax incentives is elasticity. (Buss, 2001). Elasticity is the percentage effect on state business activities resulted from a 1% change in state and local taxes. Bartik (1991) discovers that the mean of 49 tax studies was -0.25. This suggests that a 10% reduction of all taxes from their original level allows the local businesses and employment to have a long-term 2.5% increase above the growth that would occur without the reduction. Many scholars question the correlation between tax and growth and magnitude of the economic importance. Current economic conditions are likely caused by past economic activity. The areas with tax incentives are often economically different than areas without those (Billings, 2008). To better address the relationship, several recent studies try to separate out the non-tax confounding factors related to economic growth. I will mainly focus on the employment effect and its costs and benefits.

Job Creation

Most recent literature suggests that the relationship between employment growth and tax incentives is ambiguous (Gabe & Kraybill, 2002; Luger & Bae, 2005; Billings, 2008). The study of Colorado enterprise zones finds that incentives do not substantially increase the number of job creations by studying firm expansions after two years.
The relationship might even be negative. The businesses that received incentives have a decrease of 10.5 jobs per firm. In contrast, the establishments that did not receive incentives have an increase of 6.6 jobs per establishment. The finding is that enterprise zones, part of tax incentive programs, have no positive impacts for job creation (Billings, 2008). But the support is weak because the data is limited and the positive relationship does not apply to all industries studied. According to the case study of North Carolina, some 262 new jobs are induced from four types of tax credit, which otherwise would not be created, based on 1999 data. The results suggest that machine and equipment tax credit induces most jobs (48%) and job creation tax credit is the second-largest component (Luger & Bae, 2005). The mixed results continue the controversy of effectiveness of tax incentives.

Even if certain tax incentives result in some new jobs, the effect, however, is not the same in each community. Scholars of these studies find that job creation mostly occurs in more distressed areas, based on population growth, unemployment rate, and per capita income. For example, Enterprise zones in Ohio have to meet one of these certain criteria: lower than state average of employment rate, population growth rate, or per capita income (Billings, 2008). North Carolina divides counties into different tiers and gives a different amount of incentives to each tier. Areas with adverse economies normally receive a larger amount of incentives. With all other factors being the same, firms in distressed areas would have more cost-reduction benefits. Thus, more new jobs are likely to be generated (Luger & Bae, 2005).

Cost and Benefits Analysis

If we assume that these targeted programs in needy communities do induce some job growth, who actually benefits? The findings on this issue are skimpy. Peters and Fisher (2004) find that firms in enterprise zones hire from metropolitan areas instead of local labor markets. They look at the commuting patterns of workers in the zones in a number of states. Most workers in the zones do not live there; the majority of those who live in the zones do not work in them. BMW reported that they hired 5,000 employees at its Spartanburg, South Carolina plant (Douglas & Paulo, 2008), but all high-skilled workers were from outside of the community. Therefore the locals did not necessarily benefit from the job gain.

In addition, job creation from tax incentive programs is costly. The cost per job created in North Carolina in 1999 was about $147,463 on average. The high costs are because the job creation from increases in investment on machine and equipment and R&D are smaller than from direct job creation credits (Luger & Bae, 2005). States buy jobs through targeted tax incentives. BMW in South Carolina pays $1 a year to lease a $36 million piece of land. It pays no land tax, and the building and equipment tax on the first phase is 43 percent lower than what other firms pay (Douglas & Paulo, 2008). Regardless of the revenue loss, state still needs to pay for public services and infrastructure to maintain and attract businesses. These incentive programs make it harder for states to finance important functions, such as transportation systems, public education, and utilities programs (Kaye, 2008). By keeping up the tax incentive programs to attract businesses, states would decrease the ability to offer financing for foundations for future economic growth. The distressed areas would be likely to maintain at the bottom in the competition.

However, some scholars argue that several positive externalities are not considered. Based on the study of the BMW plant in South Carolina, the establishment has spread economic benefits through the multiplier effect. Not only have regional suppliers generated more jobs and revenue, but also employees have purchased more at local businesses. The spending leads to more jobs and income in other establishments. The study finds that BMW’s South Carolina plant supports 23,050 jobs through the multiplier effect and that the value of property, such as housing and land, has increased as well. In addition, four counties in South Carolina have received an additional revenue of $2.4 million every year from the increase of property, income and sales taxes (Douglas & Paulo, 2008). Even though this study shows that targeted tax incentives are beneficial, more research needs to be conducted regarding costs and benefits of employment effect.


Scholars have used various methods to study tax incentive effects, such as case study, survey, econometric regression, and simulation. Previous models have adopted regression analysis to evaluate the correlation between state economic growth and tax incentives. But most models have problems in separating out non-tax confounding variables, such as effects of agglomeration economy, firm establishment-levels, and self-selection of enterprise zones (Gabe & Kraybill, 2002). Furthermore, previous static models might produce incomprehensive outcomes, given economies or establishments are dynamic.

The recent studies discussed in this review address these issues by modifying previous aggression or simulation models. Gabe & Kraybill (2002) develop a two-stage regression model to control non- incentive growth factors, such as industry, size and age of a firm, regional growth, and infrastructure. This approach helps evaluate effects of incentives after business expansions. The North Carolina case study is conducted through a simulation, which projects the employment growth with firm-level data. This approach demonstrates how firms respond to state tax incentives regarding changing in employment (Luger & Bae, 2005). To eliminate bias of self-selection of enterprise zones, Billings (2008) employs a border-matching methodology, which matches only enterprise zone and non- enterprise zone areas in close geographical proximity. This could control time-varying unobservables and economic conditions, generating more accurate relationship between tax incentives and employment growth.

Further Studies

As most related literature suggests, relationship between tax incentives and regional economic growth might be weak at the state level. Two possible reasons could lead to the findings. First, tax effects are likely to be small in larger economies, especially at the state level. Secondly, it is possible that incentives do induce significant new growth in poor and needy communities. But little effects of more developed areas in the state cancel out the growth. Therefore, cases of different states or areas should be studied separately. In addition, more time-varying and micro-level data are needed to produce more in-depth results. Because states are unwilling to conduct evaluations, this issue might be difficult to address. But further studies can be done with available data.

Scholars could develop a multilateral simulation to evaluate the effectiveness of state tax incentives because one state could generate tax exportation and incidence to businesses in other states. Further analysis of costs and benefits should be done, given little work on cost and benefit analysis of tax incentive programs. I would like to look closer at effectiveness of distressed areas from the perspective of the locals. The cost and benefit analysis should take into account the multiplier effect and intangible effect, such as benefits for future generations. But the analysis should not include benefits or costs of workers coming from outside of the area. Hopefully the more detailed and quantitative analysis can contribute to the study of the relationship between tax incentives and regional economic growth.


Bartik, T. J. (1991). Who benefits from state and local economic development policies?. Kalamazoo, Mich: W.E. Upjohn Institute for Employment Research.

Billings, S. (2009). Do enterprise zones work? an analysis at the borders. Public Finance Review, 37(1), 68-93.

Burnier, D. (1992). Becoming competitive: How policymakers view incentive-based development policy.

Economic Development Quarterly, 6(1), 14-24. doi: 10.1177/089124249200600102

Buss, T. F. (2001). The effect of state tax incentives on economic growth and firm location decisions: An overview of the literature. Economic Development Quarterly, 15(1), 90-105. doi: 10.1177/089124240101500108

Chi, K. S., & Hofmann, D. J. (2000). State business incentives: Trends and options for the future Council of

State Governments.

Douglas P Woodward, & Paulo Guimarães. (2009). BMW in south carolina: An update. Business and Economic Review, 55(2), 13.

Fisher, P., & Peters, A. (1997). Tax and spending incentives and enterprise zones. New England Economic Review, 109-109.

Gabe, T. M., & Kraybill, D. S. (2002). The effect of state economic development incentives on employment
growth of establishments. Journal of Regional Science, 42(4), 703-730.

Kaye, T. A. (2008). The gentle art of corporate seduction: Tax incentives in the united states and the
european union. University of Kansas Law Review, 57(1), 93.

Luger, M. I., & Bae, S. (2005). The effectiveness of state business tax incentive programs: The case of north carolina. Economic Development Quarterly, 19(4), 327-345.

Peters, A., & Fisher, P. (2004). The failures of economic development incentives. Journal of the American Planning Association, 70(1), 27-37. doi: 10.1080/01944360408976336

Literature Survey

by Ingrid Zhuang ZHUANG_INGRID_2013_Lit_Survey_Revised



China’s rapid economic growth has fueled a housing market boom for more than a decade. Property market amounts to a large sector in the Chinese economy, with real estate investment accounting for 13 % of China’s GDP[1] and a quarter of total fixed asset investment (FAI) in 2011.[2] China’s real estate market is primarily driven by population growth, urbanization, and government interventions. Although China is transitioning from a planned socialist economy to a market-oriented economy, the central government continues to intervene the housing market on a regular basis. Therefore, as the country rises in importance in global economic growth, policy makers, investors, and scholars demand a better understanding of the government interventions’ impact on China’s booming housing market.

Literature Review

There are many academic papers that explore monetary and fiscal policies’ impact on the property market. Most studies focus on the supply-side interventions, which result in rising demand and housing prices. In the United States, such regulations include land-use restrictions and other zoning rules, whereas in China, a stimulus package is employed to spur real estate investment and house-ownership. Few studies have been done on the demand-side regulations since, beginning in April 2010, the Chinese government suddenly and for the first time changed direction from a period of encouraging home purchase and real estate investment to cooling the housing market. In this brief literature review, I follow the literature on Chinese housing market in a chronological order, with specific focus on housing price movement influenced by macro trends and the recent policy change. I will first discuss literature surrounding the macro-factors (urbanization and migration) that led to the booming Chinese housing market, particularly relevant to major cities such as Beijing. Next, it is important to examine the research that has been done on the Chinese expansionary policy and study its effect on housing prices. Lastly, I will review a working paper that analyzes the new housing regulation, in order to shed light on my research of the policy’s impact on housing demand and prices, using Beijing as a study case.

Since the economic reform of 1978, China’s urbanization started to pick up its pace, the urban population grew from 185 million to 607 million and the urbanization rate increased from 19.0% in 1979 to 45.7% in 2008[3]. Major cities are going through dramatic expansion in conjunction with the transition to a market economy. As more and more workers migrate from rural areas to cities, housing policy becomes an important issue that demands extensive research. There are many comprehensive discussions on the formation and development of the Chinese housing market, such as Zheng and Kahn (2007), Deng et al. (2011), and Ding (2012) among many.

Zheng and Kahn (2007) choose Beijing as the focus of their research. Using two geocoded data sets, they present new evidence on Beijing’s recent free real estate market, including housing price gradient, land price gradient, population densities, and building densities. They conclude that the monocentric model predictions are largely upheld in Beijing: land and property prices decline with distance from the Central Business District (CBD), as does population density. Additionally, they find that local public goods, such as clean air, access to public transportations, and proximity to high-quality schools, are significantly capitalized into real estate prices (2007). Although much has changed in Beijing since 2007, the monocentric model presented by Zheng and Kahn (2007) still holds. As urban population growth continues to exert great pressure on the city’s already overloaded infrastructure, the urban fringe is quickly expanding and being converted from agricultural to urban uses. After new infrastructure took form for the 2008 Summer Olympics, Beijing’s city congestion and air pollution problems became increasingly pressing. A large percentage of residents, especially families with children, who used to live near the city center, begin to move towards the outskirts of the city, substituting rising transportation costs for a better living environment.

Having reviewed the general macro-trends of urbanization and the housing boom, it is important to fit the Chinese government’s interventions into the framework and examine the impact on housing demand and prices. Since the central government issued the stimulus package in November 2008, China’s real GDP boosted an annualized 5.7% (from 6.2% to 11.9%) from 2009 to 2010, as reported in Deng et al. (2011). Amidst this phenomenal response, real estate market in major cities boomed exponentially, with housing prices soaring. Deng et al. (2011) is a comprehensive research that analyzes how 2008 monetary and fiscal stimuli have affected the housing prices in major Chinese cities. Employing land parcel auction data from eight major Chinese cities, Deng et al. (2011) uses a pooled hedonic land pricing model estimated by OLS, a sandwich estimator allowing for clustering by city, and random effect regression. The dependent variable in the model is “the transaction price for each parcel in the logarithmic form measured as the price per square meter of the permitted housing floor space” (2011). The results show that residential land auction prices in eight major cities rose about 100% in 2009, controlling for quality variation (2011). Therefore Deng et al. argue that much of the stimulus investment involved highly leveraged purchases of real estate, which benefited the centrally controlled state-owned enterprises (SOEs), fueling a real estate bubble. The model in Deng et al. (2011) is relevant to my research, as I will be using Beijing’s land parcel and residential housing transaction data to examine the housing price movement following the policy reversal in 2010.

Since the end of 2008, the government’s expansionary package has led to an unsustainable housing boom with the housing prices rising at an astonishing rate. Many studies have suggested that a credit expansion in such a large scale encouraged not only home purchases for consumption use, but also home investment by speculators (Ding, 2012). On April 17, 2010, the central government announced a new Chinese housing policy, known as the State Council “Document Number 10”. The specifics of the “Document Number 10” include: the minimum down payment ratio for second-home purchases was raised from 40% to 50%, and non-residents can no longer obtain mortgages to buy homes in a city unless they have paid taxes in that city for at least a year (see Figure 1, Ding, 2012). Clearly, the new housing policy’s aim is to curb excessive growth in housing prices and to tame housing speculations, with a central focus on tightening mortgage supply for housing speculators.

Researches dealing with the Chinese new housing policy are very limited. Wenjie Ding’s “Evaluating Housing Policy Interventions in China: Using Stock Market Data (2012)” is one among the few working papers. Ding’s work is a comprehensive event study that analyzes the new Chinese housing policy’s immediate impact on publicly traded real estate firms when the policy was first announced on April 17, 2010. Given a lack of appropriate and reliable data, Ding proposes a different strategy in approaching her research question. She collects high frequency financial market data from all publicly traded real estate firms on the Hong Kong Stock Exchange (HKEx), the Shanghai Stock Exchange (SSE), and the Shenzhen Stock Exchange (SZSE), and employs a multivariate regression in which abnormal return is parameterized in an individual stock return equation (2012). Ding detects “cumulative abnormal returns of about -15%, which is consistent with a 2%-4% decline in average house price across the nation. In major markets with inelastic supply, such as Beijing, equilibrium house prices are expected to decline by as much as 6%-10% with a 10 percentage point increase in required minimum down payments for housing speculators” (see figure 4, 2012). Taking into consideration of possible pre-trends bias, Ding extends the event window to 20 trading days before the policy announcement date and uses baseline estimates and other test of robustness to lend support to the estimated impact.

The results of Ding’s event study statistically attest to the intention of the “Document Number 10”. Ding concludes that the new housing policy to curb housing price growth results in negative stock returns for real estate firms. She finds that real estate firms with a diversified portfolio are hedged against housing-specific policy risks, while firms with an affiliation with the central government suffer greater downturns as the government withdraws its support. My proposed research focuses specifically on Beijing’s housing price movement, and will contribute to the existing literature as empirical evidence of the endogenous effects of the “Document Number 10”. Ding’s study is limited to the negative shock after the first announcement of the “Document 10” in April 2010, and ignored the impacts of subsequent amendments in late 2010, and 2011. Instead of using data from all publicly traded real estate firms in China, I will gather Beijing’s individual house purchasing transaction data post 2008, and present Beijing’s real estate market as a case study in analyzing the new policy’s effect on housing prices. With all the background information in mind, I will proceed to present my model in the following section.








Ding, Wenjie.
2012. “Evaluating Housing Policy Interventions in China: Using Stock Market Data.” Job Market Paper. The Wharton School, University of Pennsylvania.

Deng, Yongheng, Randall Morck, Jing Wu, and Bernard Yeung.
2011. “Monetary and Fiscal Stimuli, Ownership Structure, and China’s Housing Market.” National Bureau of Economic Research. NBER Working Paper No.

Kahn, Matthew E., and Siqi Q. Zheng.
2007. “Land and Residential Property Markets in a Booming Economy: New Evidence from Beijing,” Journal of Urban Economics.

Source: National Bureau of Statistics (NBS) of China.

Source: CEIC Data. Macroeconomic, Industry and Financial Time-Series Databases for Global Emerging and Developed Markets.

Source: China Population and Employment Statistical Yearbook 2009

From Homeowners’ Assurance to Public Policy: Origins and Modern Applications of Zoning

 by Stephanie Xu Xu_Stephanie_2013_Literature_review_revised


From Homeowners’ Assurance to Public Policy: Origins and Modern Applications of Zoning


The practice of zoning as a method of urban planning and land use regulation has been surrounded by equity and segregation debates over the past few decades.  A wide body of literature about the topic exists, with a range of focuses, from its early implementation in the early 1900s to its efficiency to its effect upon race and class segregation.  This paper examines the history and progression of zoning practices and policies, as they are crucial to understanding the associated controversies.  We then look into the impact of zoning upon income distribution and equity, first through a more dated but resourceful paper by Fernandez and Rogerson (1997), and then through a 2006 paper by Calabrese, Epple, and Romano.  Lastly, we explore literature concerning the relationship between zoning laws and public health initiatives.


Fischel (2004) offered a comprehensive overview of zoning from its early 1900s origins to modern day policy reforms to address the exclusionary products of such regulations.  Fischel posited that zoning was not a necessary practice until the advent of buses and trucks that allowed businesses to locate farther from streetcar tracks and stations.  Until then, workers were able to buy houses in the suburbs without fear of encroaching business.  However, companies’ newfound capability to transport resources threatened the residential environment that workers found so valuable.  Thus, zoning was the only way to give homeowners security and assurance that their neighboring land wouldn’t become a noisy business district.  Several decades later, the problem was exacerbated by a sudden increase in highway construction in the 1960s.  Firms could now move out of high-density urban areas even more easily.  Failing or struggling cities and suburbs welcomed the influx of businesses for the accompanying fiscal benefits.  Meanwhile, only well-off communities could afford to keep businesses out.  Fischel argued that it was in this environment that income distribution began to play such a strong role.  In this paper, he sought to find a way to keep home values stable, as zoning was initially created to accomplish, without causing the exclusion with which we are faced today.  To resolve this, he proposed selective home equity insurance, despite his acknowledgement of the administrative problems and risks of moral hazard and adverse selection.

Fernandez and Rogerson (1997) and Calabrese et al. (2006) investigated the impact of zoning upon equity, but differed in their approaches in a critical way.  Both used a two-community model with two different income distributions.  Both also made a point of stratification as a natural occurrence, with or without zoning laws.  But the former required in their model that individuals committed to one community or another before any voting took place, and before they bought any property.  Calabrese et al., on the other hand, based their argument upon the Tiebout rationale that people could “vote with their feet” and choose to reside in towns that align with their personal preferences.


Thus they arrived at different conclusions.  Fernandez and Rogerson found that in general, zoning made the richest better off and the poorest worse off, but left a great deal of ambiguity about those in the middle.  Poor individuals who move from a more affluent community to a lesser one because of zoning policies tend to be better off, while individuals who move to a richer community tend to be worse off.  Calabrese et al. finds an overall increase in efficiency and welfare as a result of zoning, due to their assumptions under the Tiebout model.  They conclude that because residents can move after a community votes on tax rates and local public goods, “aggregate welfare gains arise from better Tiebout matching of preferences to levels of public-good provision” (Calabrese et al. 2006, 4).  Comparing these two scholarly works highlights a decisive factor in zoning theory: that the order in which individuals vote on taxes and public goods, buy property, and choose their community affects the predictions of the ensuing welfare gains or losses.


Finally, we examine more recent literature relating zoning codes and public health solutions.  Ransom, Greiner, Kochtitzky, and Major (2011) define the “built environment” as anything man-made or man-modified, from buildings to highways, that act as sources of pollution (94).  They draw linkages between the built environment and health, and cite the disproportionate cases of obesity in densely-packed urban areas as a prime example.


It follows that zoning codes should take public health considerations into their analysis, as the City of Baltimore did in its 2008 TransForm Baltimore campaign (Public Health Work Group 2008).  The planners endeavored to create an environment that allowed people to be healthy through such measures as increasing green spaces, establishing more mixed-use space, and controlling the locations of fast food stores and liquor stores, especially relative to schools and residences.   However, Ransom et al. did identify a number of obstacles to pursuing such zoning policy reform.  First, the research and evidence to support these reforms is complex, both to accumulate and to transform into actionable policies, as the precise causal effects of certain zoning layouts upon health problems is difficult to determine due to confounding factors, such as population demographics, the types of businesses occupying retail and office space, etc.  In other words, the effects of zoning are difficult to isolate from other characteristics of a neighborhood or community.  Second, using zoning to address public health issues is politically difficult, especially when other priorities exist, such as improving walkability and access to healthy food (Ransom et al. 2011, 96).  Third, efforts such as the TransForm Baltimore initiative lacks input from a variety of community members and leaders, which prevents policymakers from fully comprehending the context.  Lastly and perhaps most importantly, business leaders and neighborhood housing associations are well-organized constituency groups that often oppose things such as mixed-use land, mixed-income housing, and reduced parking requirements.


The existing research addresses several arguments in favor of and against zoning, and the diversity of the literature opens many doors for further research.  In the case of Durham, North Carolina, it may be worthy to investigate the impact of Duke Hospital upon the surrounding residential neighborhoods and efforts to preserve a certain environment despite increased traffic.  A similar analysis may look into zoning designations in the Warehouse district downtown, and who made the decisions to make one set of warehouses into the American Tobacco District and another into the West Village apartments.  Durham, as a growing city, stands to see more conflicts in the coming years as businesses grow and residents face the potential intrusion that people feared a century ago.






Calabrese, Stephen, Dennis Epple, and Richard Romano.  “On the Political Economy of Zoning.”  July 2006.


Fernandez, Raquel and Richard Rogerson.  “Keeping People Out: Income Distribution, Zoning, and the Quality of Public Education.”  International Economic Review 38:1 (1997), pp. 23-42.


Fischel, William A.  “An Economic History of Zoning and a Cure for its Exclusionary Effects.”  Urban Studies 41:2 (2004), pp. 317-340.


Public Health Work Group, “Zoning Recommendations.”  Transform Baltimore, City of Baltimore, 17 November 2008.


Ransom, Montrece McNeill, Amelia Greiner, Chris Kochtitzky, and Kristin S. Major.  “Pursuing Health Equity: Zoning Codes and Public Health.”  Journal of Law, Medicine & Ethics (2011), pp. 94-97.



Literature Review

by Katerina Valtcheva Valtcheva_Katerina_RewrittenLiteratureSurvey


I am interested in exploring the efficiencies and inefficiencies of zoning and land use regulation in the U.S. To this end, I will briefly look at the history of zoning and some major events that shaped how it has been exercised. Then I will present the findings of several studies concerning the practice and examine where that leaves us today.

Growth control restrictions and zoning were originally developed as a way to prevent the development of industrial and commercial zones in proximity to residential areas. However, zoning was not largely employed across America until the Standard State Zoning Enabling Act of 1922 was enacted by the U.S. Department of Commerce. Since the adoption of the act, zoning became a widespread practice that has not been significantly affected even by adverse court rulings. As a result, local municipalities have the biggest say in when and where to employ regulations. Since governments have recognized the importance of property value maximization as one of their objectives, zoning has become a way to exclude lower income residents from communities. Theoretically, zoning and growth control restrictions imply that taxes paid for local services are in one to one correspondence with the demand for those services and ensure that no member of society gets services than they have not paid for. (Fishel 1992; Quigley and Rosenthal 2005)

In his paper on property value maximization, Brueckner (1983) builds on Tiebout’s broad idea of governments achieving efficiency in a community by maintaining its population subject to the minimum individual cost of public goods and services. Brueckner’s analysis of governments that engage in property value maximization while allowing for private land and property markets (the effects of this behavior on land rent and house size are ignored), proposes a more concrete application of Tiebout’s original idea.

Brueckner (1983) showed that communities are internally in Pareto-efficient equilibrium when their governments maximize the value of aggregate property by selecting their public good output provided that government revenue is collected by a per-house tax. If the same revenue is being collected with a distortionary tax, such as income tax, the same community will be internally efficient, but not internally Pareto-efficient (it is important to note that the paper did not clearly define an “efficient allocation” as clearly distinct from Pareto efficiency). However, this government strategy of value maximization cannot achieve a globally efficient equilibrium. The reason for this is that in the model the zero-profit utilities between communities will differ depending on the community population and its tastes. The Nash equilibrium of government behavior with respect to maximization of property value ensures the internal Pareto-efficiency mentioned prior; however, it does not always assign community consumers optimally, which leads to global inefficiency. Fischel (1992) runs into similar complications with regards to quantifying zoning efficiency in large areas. However, instead of lack of Pareto-efficiency, he argues that zoning is less effective in large cities rather than in suburbs and rural areas due to the heterogeneity of larger populations’ interests.

Bruckner suggests that a government-applied population control such as zoning could be the key to achieving global efficiency. The reasoning behind his statement is that such restrictions can make the population’s tastes more homogeneous by excluding lower income residents and thus increasing efficiency in communities where taxes are distortionary.  However, the effective introduction of zoning into the model is a problem that has yet to be solved and requires further research. One factor that must be taken into consideration is that studies of zoning and growth regulations have not yet yielded a unanimous and statistically significant proof of such practices’ effectiveness in all areas, or their consequences.

Fischel (1992) claims that even though studying zoning is a challenge given the abilities of governments to amend the laws due to new political environments, it is possible to extract broad conclusions, both because the laws governing the states are similar and because of the effects of the Standard State Zoning Enabling Act of 1922. However, Quigley and Rosenthal (2005) in their paper on the effects of land use regulation argue that the complex interrelationship between demand for housing, availability of public services and conditions for them makes the implementation of an optimal regulatory framework a daunting task for any municipality.

One difficulty that arises when examining the effects of zoning is that numerous purposes of land use and regulatory dimensions exist; however, this is not the only problem with past studies. Methodological problems with various studies include the virtual impossibility of extracting generalizations about large areas. One reason for this is given by differences in standards for measuring land oversight. Other reasons include scarcity of data or not accounting for the fact that regulation and prices are determined endogenously. Moreover, the studies are conducted irregularly, which makes it even more unlikely for a generalization between them to be drawn. (Quigley and Rosenthal, 2005)

Fischel (1992) claims that many studies show that “adoption of more restrictive zoning reduces the value of undeveloped suburban land subject to restrictions and increases the value of already-developed homes.” However, this does not unambiguously prove zoning laws’ effectiveness. Quigley and Rosenthal (2005) examine results from various studies in order to assess what conclusions can be drawn regarding the effectiveness of zoning. Several quoted studies largely failed to demonstrate the effectiveness of monopoly zoning on property values, until an analysis by Thorson (1996) showed promising results. Thorson analyzed reported median home values using more complex models, which included neighborhood and housing quality controls, where his concentration ratio turned out to be significantly related to increased home prices. Other studies restricted to certain areas in the U.S. concluded that zoning has little to no effect on property values. However, some of the later studies that address challenges in existing methodologies confirm the positive effects of zoning on property values. According to Quigley and Rosenthal, a lot of these articles show that regulations of land use raise the value of existing property, but diminish that of developed land, which is the same result at which Fischel arrived. However, Quigley and Rosenthal determine that the overall effect of density regulations on property values is ambiguous. They conclude that it is not possible to make any definite generalizations about the effect of zoning and land use restrictions onto property values, which contrasts with Fischel’s more optimistic view.

It seems that Pareto-efficient equilibrium can be achieved in various communities by the optimization of a public good and thus maximizing property value (Brueckner 1983). This finding provides a partial theoretical solution to one of the problems government have on their agendas, but further research must be done in order to reach a globally Pareto-efficient equilibrium. The way Brueckner proposes this issue to be solved is through the use of zoning. However, there are a lot of unsolved problems that need to be addressed before a global property-value-maximization model can be developed.



Quigley, John, and Larry Rosenthal. “The Effects of Land Use Regulation on the Price of Housing: What Do We Know? What Can We Learn?.” Cityscape. 8.1 (2005): 69-137. Web. 7 Feb. 2013.

Fischel, William. “Property Taxation and the Tiebout Model: Evidence for the Benefit View From Zoning and Voting.” Journal of Economic Literature . 30.1 (1992): 171-177. Web. 7 Feb. 2013.

Brueckner, Jan. “Property Value Maximization and Public Sector Efficiency.” Journal of Urban Economics. 14. (1983): 1-15. Print.

Thorson, James. “An Examination of the Monopoly Zoning Hypothesis.” Land Economics. 72.1 (1996): 43-55. Web. 7 Feb. 2013.


The Economic Impact of Institutions of Higher Learning

 by Michael Rebuck Rebuck_Michael_2013_LiteratureSurvey

The Economic Impact of Institutions of Higher Learning

Institutions of higher learning can greatly affect their surrounding communities and many universities and colleges will undergo studies to state this economic impact. The most obvious manner in which universities and colleges affect their local economy is through employment and through the direct purchase of goods and services. Universities and colleges also act on their neighborhoods in indirect ways, such as by developing real estate. However, many of these economic impact studies exaggerate or incorrectly state the economic impact of their universities. To correct this, several other studies have been performed to identify these problems in order to allow individuals to judge objectively the role these institutions play in their communities. By acknowledging that a university’s economic impact can have both positive and negative effects, it becomes easier to discuss what role a university should take in shaping its community.

J.J. Siegfried et al. seek to address the problems that many university economic impact studies contain. Many of these studies attempt to determine the degree to which an area is better with an institution of higher learning than it would be without it. However, establishing this counterfactual can be difficult. Institutions of higher learning do not appear and disappear quickly and their expansions and contractions occur slowly. Indeed, it is very hard to determine what that local area would be like without the existence of an institution of higher learning. Therefore, it can be beneficial to measure the effects of only incremental investment made by universities (Siegfried 2007). Unfortunately, it is not always easy to say whether an effect is positive and negative. For example, a population that increases due to the new hiring of faculty can be both a good thing and a bad thing. On the one hand, there is the benefit of an expanding economy, but on the other hand there can be an increase in congestion and pollution. Some of the new faculty may bring spouses with them who will take the existing jobs of local residents. Many economic studies performed by institution of higher learning falsely imply that their contributions are always positive and these assumptions need to be challenged.

Determining an area of influence is one of the most challenging factors while creating an economic impact study of universities. The appropriate geographic boundaries depend upon the questions at hand but should always remain constant throughout the analysis. Annette Steinacker attempts to measure the impact colleges have upon their surrounding neighborhoods, which can be very different from their impact upon an entire metropolitan region (2005).  In the past, most US university impact studies were based upon a regional input-output analysis. Until the release of economic data at the zip code level, the smallest geographical unit that the US Bureau of Economic Analysis measured economic multipliers for was the county. By using smaller local area units, it is possible to create a local area unit for only the university’s surrounding neighborhood.

Economic impact is most commonly measured by summing the direct expenditures of the college community created by the existence of the institution. It is important to also apply multipliers in order to account for the interconnectedness of economic activity. Purchases from firms outside the impact area are considered lost to the local economy. The core of the analysis is the expenditures for goods and services by the university in addition to its payroll (Steinacker 2005). Multipliers for university expenditures take into account the additional spending by local companies resulting from each dollar of university purchase. The multiplier for payroll takes into account the propensity for individuals at a certain income level to consume specific goods. Most spending takes place in the area where the individual resides and that is one reason why the act of defining the area of influence is so important. If one uses the county as the defined impact area, a higher multiplier will be obtained than if only the local neighborhood is defined as the impact area. Inconsistent determination of relevant areas of influence can lead to great disparities between economic impact claims across studies.

Colleges and universities commonly claim that they create jobs, boost tax revenue, and stimulate their local economy (Siegfried 2007). Other studies have claimed that universities resist business cycle fluctuations, attract outside revenue, and attract and develop human capital. The purpose of many of these studies is to articulate the value of these institutions in order to help them compete for state funding and to maintain tax-exempt status. A university’s goal of persuasion leads one to doubt the concrete accuracy of their reports. Siegfried et al. point to inconsistencies across studies as a reason to question their validity. For example, multipliers for job impacts ranged from 1.03 to 8.44 in a review of 138 university impact studies.

Steinacker argues that universities and colleges located in large metropolitan counties should use smaller geographical units in order to more accurately reflect their impact. She also argues that only employees who have moved to the area specifically because of the university should be included in economic impact studies. Additionally, for employees who live outside of the zip code areas, only the purchases they make in the target area should be counted. For example, if a professor at Duke purchases a bagel and a coffee at the Erwin Road Dunkin’ Donuts every day, this should be included in the calculations. The third adjustment calls for studies to include more detailed information on student expenditures. Some studies include student expenditures, but they are often only a standardized estimate from the university’s financial aid office. Steinacker prefers a survey, and her student expenditures are calculated in a similar manner to the expenditures of employees (2007). Calculating student expenditures on housing can be difficult. On-campus housing is included within university budget data, but students who rent or buy off-campus are not. Local circumstances determine the impact of student housing spending. For example, if there are a lot of vacant housing units surrounding a university and students merely fill these units, then all rent can be considered new. However, where vacancy rates are low, student demand can drive rent higher and force local residents to move to new areas. Landlords will benefit, but it is unclear if the community as a whole does.

Some people assume that universities always have a positive impact on their surrounding housing markets. Alvaro Cortes investigates this relationship by studying local neighborhood housing markets from 1980 to 1990, and determines that the characteristics of neighborhoods next to universities are significantly different than the citywide neighborhoods. Like Steinacker, he uses zip codes to define the geographical units. Cortes also examines the differences in impact that private and public universities have on their surrounding neighborhoods and identifies why these effects are not always positive (2004).

Universities have both a direct and an indirect impact on real estate in their surrounding neighborhoods. By 1996, US universities held over $100 billion (book value) in land and buildings (Cortes 2004). Universities directly impact their surrounding neighborhoods through housing development partnerships, direct expansion, and through campus generated externalities. For example, these effects can increase the desirability of a neighborhood by increasing the amount of cultural events offered. On the other hand, a university can create negative effects if, for example, new student housing leads to a large increase in noise complaints in an adjacent neighborhood. Universities indirectly impact their surrounding neighborhoods through the effects of their partnerships and by attracting a distinct population. Many municipalities recognize the influence of universities and universities are occasionally regarded as important players in a city’s development plans (Cortes 2004).

Cortes empirically investigates the impact of universities on their local neighborhood by selecting five public universities and pairing them with a private university that exists in the same city. His empirical work uses three ordinary least squares multiple regression models to evaluate the research questions. Only census tracts that are directly adjacent to census tracts that have been defined as part of the university are included. They are then compared to the other citywide neighborhoods. Data for these examinations is drawn from the “Under Class Data Base” (UDB) (Cortes 2004). One problem with Cortes’s research is the size of his sample, as effects are likely to vary considerably from campus to campus.

The descriptive statistics show that there is a clear difference between the neighborhoods abutting a university and the other citywide neighborhoods. The poverty rate of university neighborhoods is typically 50% higher than that of the other neighborhoods. Additionally, in half the total number of cases, the percentage of persons without a high school diploma is higher in university neighborhoods than in city neighborhoods. This is somewhat surprising, as university faculty and students should all have achieved a high level of education, but the overall neighborhood effect is to the opposite effect. University neighborhoods also have higher levels of renters than other areas of the city do. Most of these university neighborhoods demonstrate lower monthly rental payments. Lastly, new residential unit construction is higher in university neighborhood (Cortes 2004).

The results of the regression and path analyses demonstrate that there is a statistically influential effect of neighborhood proximity to an urban university, but these effects are not systematic (Cortes 2004). Differences in changes between the two types of neighborhoods could be attributed to university decisions. For example, some universities have a large amount of student housing on-campus while others do not. Since students often desire low-rent, low-quality housing, a lack of on-campus housing could result in an overall pattern of neighborhood downgrading. It is hard to determine what goals a university should have with regards to its surrounding neighborhoods. For example, inflated rent may price university students out of neighborhoods. On the other hand, deflated rent may attract poorer individuals to a neighborhood, which could have negative consequences. It is important to note, however, that a university’s decisions have an important effect upon their surrounding neighborhood. Knowing that these effects can be both positive and negative, it is important for universities to align their goals with those of their community.

Institutions of higher learning can greatly impact their surrounding communities in a variety of direct and indirect ways. The impact of universities and colleges upon their surrounding neighborhoods is not always positive, and it is important to acknowledge both the good and the bad effects. By acknowledging that a university’s economic impact can have both positive and negative effects, it becomes easier to discuss what role a university should take in developing its community.

Works Cited

Cortes, Alvaro. “Estimating the Impacts Urban Universities on Neighborhood Housing Markets: An Empirical Analysis.” Urban Affairs Review 39.3 (2004): 342-75. Web.

Siegfried, John J., Allen R. Sanderson, and Peter McHenry. “The Economic Impact of Colleges and Universities.” Economics of Education Review 26.5 (2007): 546-58. Web.

Steinecker, Annette. “The Economic Effect of Urban Colleges on their Surrounding Communities.” Urban Studies 42.7 (2005): 1161-75. Web.

Urban Poverty and Geographically Concentrated Low-Income Communities

by Gini Li Li_Gini-Lit_Review_Revised

Urban Poverty and Geographically Concentrated Low-Income Communities


When trying to understand any type relationship between phenomena, the hardest point to establish is causation. Two seemingly correlated variables do not necessarily cause or have a significant impact on each other. This is the case in studies on the causes and effects of concentrated urban poverty. Factors such as race and social networks have been identified as factors that cause low-income urban communities to form and prevail, but the results are conditional on the data used, theory postulated, and method employed. The following research explores the causes and implications of concentrated urban low-income communities.


Urban poverty is usually defined in two ways: as an absolute standard based on a minimum amount of income needed to sustain a healthy and minimally comfortable life, and as a relative standard that is set based on average the standard of living in a nation (McDonald & McMillen, 2008, p. 397). The poverty line, as defined by the U.S. federal government, is an annual income that is three times the cost of a nutritious diet, as computed by the U.S. Department of Agriculture (ibid, 398). These definitions for poverty shape the nature of public policy focused on poverty reduction. However, they define poverty as the standard of living at a given point in time. Low agency (or the ability to make choices for oneself), low standards of living, and limited mobility can also be seen as vulnerability, the definition of which can transcend a monetary and temporal definition.


Moser (1998) defines asset vulnerability as the limited ways in which the urban poor can manage their “asset portfolio”, which includes labor, human capital, housing, household relations, and social capital (Moser, 1998, p. 1). This definition differs from those described by McDonald and McMillen in that it identifies those who are at risk of being in poverty and those who are systemically stuck in poverty instead of just those who are currently “poor”. This is possible because by looking at the broader range of assets that are available to the urban poor, researchers can identify their capabilities and ability to recover from crises. Using the asset vulnerability framework, Moser chooses urban research communities that were from different regions of the world[1] but had in common a decade of economic difficulties, declining per capita income, and an increasing rate of urbanization. Additionally, 17.8% – 77.2% of residents in these communities live under country-specific poverty lines (ibid, p. 6). This context provides Moser with information on how vulnerable communities use consumption modifying or income generating strategies to cope with declining real income. The findings show that when households became “poorer”, more women and children join the work force and men migrate to generate remittances. Although this increases a household’s labor asset, it also diminishes its household relations asset, as many children grow up without a paternal figure and family relationships weaken. Increasing a family’s immediate labor asset also erodes its human capital asset, as children are sacrificing their educations for low skill labor. This increases overall vulnerability within a household because it perpetuates poverty from one generation to the next (ibid, p. 9).


Moser also found that the human capital asset is largely dependent on economic and social infrastructure provision by local governments. As households earn less income, they substitute more private goods and services for public ones. Household heads’ education level was strongly correlated to household income level. This means that if the public provision of education in times of economic downturn diminishes, the asset vulnerability or urban poor populations also increases.


In terms of housing, not having tenure security and legal titles to property made households extremely vulnerable in times of crisis. Not only did they not have incentives to upgrade their homes, but also they were unable to productively generate rent-income from their properties (ibid, 10). This produces a relatively transient population, which can affect human capital development and social capital by weakening social ties and investment in a community. The regulatory environment has a large impact on how households can maximize their housing asset.


Similarly, Moser evaluates the tradeoffs inherent in optimizing the household relations asset, as well as the social capital asset. Her study highlights a problem with the traditional definition of poverty; although a family might not be traditionally below the poverty line (for example, if children are generating income), it can still be vulnerable and cut off from economic and social mobility. However, for a government agency, the amount of resources necessary to evaluate households based on asset vulnerability is significantly higher than the amount of resources it takes to measure income. This raises the question of whether public policy should target poverty as defined by income or vulnerability as defined by capabilities.


Although Moser sees poverty as the consequence of necessary trade-offs between interrelated capabilities, other scholars see more linear causes and solutions to urban poverty. For example, one study attempts to uncover the effect of racial segregation on urban poverty in Canada. Although low-income areas in cities are positively correlated with a high concentration of minority residents, it is unclear whether this is a causal relationship. Walks and Bourne (2006) use census tract files for the 1991 and 2001 census of Canada to identify “visible minorities”[2]. With this data, Walks and Bourne then classify neighborhoods in Canadian metropolitan areas as isolated host communities (with less than 20% visible minorities), non-isolated host communities (between 20% and 50% visible minorities), pluralism/assimilation enclaves (from 50% to 70% visible minorities), mixed-minority neighborhoods (greater than 70% visible minorities, but no dominant minority), polarized enclaves (greater than 70% visible minorities and over 2/3 of population comes from the same minority), and ghettos (same as polarized enclave with additional criterion that at least 30% of all members of that minority in the urban area must live in the same neighborhood). The purpose of these distinctions is to isolate ghettos, in which minorities are forced into by the host community through discrimination, from minority enclaves, in which minorities choose to strategically live with other people from the same ethnic background. Traditionally, when immigrants arrive in Canada, they will locate in areas with high concentrations of their minority group for support services. However, as they start assimilating in the host culture, they will move to suburbs and decentralize (ibid, p. 276). This leads the authors to conclude that minorities who stay in concentrated urban minority communities are not there by choice, but are rather trapped there because of external forces.


To test this theory, they use ordinary least squares regression models that use census tract data for the most segregated metropolises. The independent variables are the proportion of each minority group, the proportion that immigrated to Canada between 1991 and 2001 (recent immigrants), neighborhood type (as defined above), and the type of apartment housing stock (ibid, p. 281). Ultimately, the study did not find that Canadian cities have ghettos, as defined above, despite high levels of spatial segregation in some of Canada’s largest cities. This is because there isn’t one dominant minority in low-income areas inhabited by visible minorities, but rather a few visible minorities that co-inhabit an ethnic community. Further study could elucidate whether these ethnic communities are formed because of social exclusion or by choice. However, the authors did confirm that low income was strongly correlated with certain minority groups, such as Aboriginals, blacks, and Latin Americans (ibid, p. 294). Additionally, it seems that certain types of housing stock, low-income residents, and minorities are highly correlated, with a huge poor minority population residing in high-rise apartments. This probably has an impact on the social capital that Moser mentioned, but more research is needed to verify this.


Ludwig et al. (2001) also explore the role of living environments in how poverty affects juvenile crime. The purpose of their study is to uncover the extent to which spatial concentration of low-income families in high-poverty and high-crime urban neighborhoods affects the criminal activities of youth. One explanation for how high-poverty neighborhoods can affect the rate of crime among youth is that they can depress the opportunity costs of crime by restricting access to quality schools, jobs, and role models (Ludwig, Duncan, & Hirschfield, 2001, p. 656). This study tests this theory with data from the U.S. Department of Housing and Urban Development’s Moving to Opportunity (MTO) experiment, which has assigned 638 families from high-poverty Baltimore neighborhoods into three treatment groups (ibid, p. 656). Unlike the previous two studies, which observe natural phenomena in urban environments, this study has a controlled experiment that has less outside influence. The MTO participants are broken down into three groups: the experimental group, the section-8 comparison group, and the control group. The experimental group receives housing subsidies, counseling, and search assistance in relocating to low poverty areas (less than 10% poverty rate) and private housing markets. The section-8 comparison group receives the same as the experimental group, but with no limitation on where they can relocate. The control group receives no aid. The families chosen for this program entered the program voluntarily, which might bias the results. Additionally, almost all of the MTO households are headed by African-American women, who prioritized escaping from gangs and drugs as primary motivators for enrolling in the program. There are 336 teens in the study that are between the ages of 11 and 16.


The independent variable in this experiment is the poverty-rate in a low-income family’s neighborhood. The dependent variable is juvenile crime from these families, which is measured through juvenile arrest records. The results show that up to 14 quarters after relocation, teens in the experimental group have significantly smaller violent-crime arrest rates (75 versus 250) but slightly higher property-crime arrest rates (300 versus 50). Possible sources of error for this study include the differences in police action in various locales and different risk-reward ratios for violent and property crimes in various locales. For example, gang involvement was probably more common in areas with high-crime and high-poverty rates, increasing the risk of not joining a gang. Additionally, in low poverty areas, there is probably more expensive property to steal and fewer people watching, driving up the reward in risk-reward analyses. In general, this study shows that environment does seem to have a positive impact on the violent crime rate for children in low-income families.

While Moser provides a general framework for understanding poverty, Walks and Bourne, and Ludwig et al. evaluate the implications of poverty’s spatial distribution. They both explore the impact of decentralized poverty, but do not go into more detail about the pathway through which poverty decentralization affects the vulnerabilities of the poor. Is it through social assimilation or better public services that the poor can reduce their vulnerabilities? And if so, can one assume that spatial decentralization actually promotes access to better social networks or better public services? These questions would be great topics for further research.

Works Cited

Ludwig, J., Duncan, G. J., & Hirschfield, P. (2001, May). Urban Poverty and Juvenile Crime: Evidence from a Randomized Housing-Mobility Experiment. The Quarterly Journal of Economics , 655-679.

McDonald, J. F., & McMillen, D. P. (2008). Urban Economics and Real Estate. Oxford: Blackwell Publishing.

Moser, C. O. (1998). The Asset Vulnerability Framework: Reassessing Urban Poverty Reduction Strategies. World Development , 26 (1), 1-19.

Walks, A., & Bourne, L. S. (2006). Ghettos in Canada’s cities? Racial Segregation, ethnic enclaves and poverty concentration in Canadian urban areas. The Canadian Geographer , 273-297.





[1] Chawama in Lusaka, Zambia; Cisne Dos, in Guayaquil Ecuador; Commonwealth, in Metro Manila, the Philippines; and Angyalfold, in Budapest, Hungary

[2] Which includes Chinese, South Asians, Blacks, Latin Americans, Lebanese/Arabs/West Asians, and Aboriginals

Literature Survey: The Effect of School Quality on Housing Prices

by Lauren Taylor Lauren_Taylor_Literature_Survey (1)

There are many factors that go into an individual’s or a family’s decision to purchase a home.  Such factors include structural characteristics of the house such as square footage and number of baths, property tax levels, proximity to amenities, neighborhood quality, and school quality, all of which are reflected in a house’s retail price.  Of particular interest to homeowners, economists, and policy makers is the effect of school quality on housing prices in any given area.  Numerous individuals have performed studies attempting to quantify how much individuals value school quality by analyzing housing prices in school districts of different quality schools.  Some studies have placed emphasis on “output-based” means of measurement such as standardized test scores and school ranking while others have used “input-based” measurements such as teacher-pupil ratio and per-pupil spending.  Most recently, researchers have focused on the use of standardized test scores as a measure of school quality and have compared these to the prices of housing in corresponding school districts.


Much of the research on the impact of school quality on housing prices can relate back to the model developed by Charles Tiebout in his 1956 paper, “A pure theory of local expenditures”.  In conducting his research, Tiebout’s model predicts that consumers pick a community to reside in based on which community best satisfies their preference patterns for local public goods (LPGs), which include schools, parks, and other amenities.  Furthermore, these individuals will move to the community whose local government best satisfies their sets of preferences, resulting in individuals self-sorting into homogenous communities with residents who demand equal levels of quality of LPGs (Tiebout, 1956).  This model can be applied to the analysis of school quality on housing prices.  According to such a model, individuals with similar preferences will populate a community.  Thus, individuals who prefer better quality schools may be willing to move to a community in which they pay higher housing prices because their desired LPG quality exists there.


Many studies have been conducted in the past two decades to further explore the effect of school quality on housing prices in various cities across the United States.  The Reinvestment Fund (TRF) conducted such a study which measured school quality in Philadelphia and quantified its impact on the value of Philadelphia real estate.  In this study, residential sales between 2006 and 2007 are geocoded and combined with data on the elementary zone  in which each sale lay as well as the percent of elementary school students scoring proficient or above on the combined Pennsylvania System of School Assessment (PSSA) for Reading and Math at the schools in that zone.  TRF also used a multilevel modeling analysis that accounts for correlation at various levels and among several factors, in its study in order to more accurately assess the relationship between school quality and housing prices.  This model takes into account the fact that other neighborhood characteristics are correlated with school quality and may affect the prices of housing in a given school district as well.  This model attempts to get rid of some bias, but some bias still remains due to omitted variables that haven’t been controlled for.   This study concluded that point increases in Structural Decline scores, which measure the impact of new construction and neighborhood disinvestment on housing prices, reduce sale prices of homes by $1.50 per square foot.  Similarly, point increases in Crime scores, which measure the impact of crime on housing prices, reduce the sale prices of homes by $1.00 per square foot.  Of even greater interest to this paper, TRF’s study found that for each percentage point increase in school district PSSA score of students who scored proficient or above, the prices of housing in that area increase by $0.52 per square foot. The correlation between housing prices and school quality can be seen in the Figure 1 below.  The two maps in Figure 1 show that central Philadelphia districts tend to be of lower housing price and lower school quality, and higher quality school districts and districts with higher housing prices tend to be clustered in the top right and top left regions of the Philadelphia area.  Thus, this study shows that overall school quality, as measured by test scores, is positively related to the price of housing in that school district.


Figure 1: Median sales price by school catchment (Map 1) and Percent of elementary students scoring proficient or above on PSSA (Map 2)


In another study, Kwame Owusu-Edusei and Molley Espey used data on housing transactions between the years of 1994 and 2000 to estimate the effect of K-12 school rankings on housing prices in Greenville, South Carolina.  These two researchers used a relative measure of school quality – school rankings – rather than an absolute measure.  Like many of the other studies that will be discussed later in this paper, this study applied a hedonic pricing model to estimate the quality of schools on housing prices.  Owusu-Edusei and Espey made two important conclusions from their data: 1) high-ranked schools have values embedded in single-family housing prices and 2) greater commuting distances to schools has a negative impact on the value of property.  In this specific hedonic housing pricing technique, the price of a house in Greenville, SC was modeled as a function of the following characteristics of a house: structural characteristics including condition, number of baths, square footage, air conditioning, lot size, and garage; block characteristics; proximity to parks, golf course and schools; and school rank categories.  The study then used ordinary least square estimations of a semi-log model and regressions to interpret the data collected.  As will be seen in many other studies documented, Owusu-Edusei and Espey found that proximity to and quality of a school does affect the prices of housing in its respective school district.  Of the houses studied in Greenville, SC, houses with elementary schools within 2640 feet (a half of a mile) of their properties have prices 18% higher than those of houses located further than 10560 feet (2 miles) from an elementary school.  Similarly, houses with middle schools within 10560 feet of their properties have prices 16% higher than those of houses located further than 10560 feet from a middle school and houses with high schools within 10560 feet of their properties have prices 12% higher than those of houses located further than 10560 feet from a high school.  Furthermore, if an elementary school rated Good, houses in that school district sell for 12% higher than those in districts with schools with a worse rating.  If a middle school rated Average, houses sell for 31% higher than houses in a district with a school of a worse rating.  Lastly, if all K-12 schools in an area rated Average and Above, the value of homes is 19% higher in that area than those in areas with Below Average schools.  Thus, it can be concluded that both greater proximity to and better quality of schools does positively affect the prices of housing located in their attendance zone.


Sandra Black conducted a study of great importance, which also used housing prices – this time in suburbs of Boston, Massachusetts – to infer the value homeowners place on school quality (1999).  Black used a sample of single-family residences within 39 school districts across 3 counties outside of Boston from 1993 to 1995 and used test scores on a statewide 4th grade assessment called the Massachusetts Educational Assessment Program as her measurement of school quality.  Black set out to calculate how much more people are willing to pay for houses located in areas with better schools.  However, Black made an extremely important observation that such a calculation can be complicated by the fact that better schools tend to be located in better neighborhoods, a characteristic which also influences the price of housing.  Black concluded that estimates of the effect of school quality on housing prices that do not adequately control for neighborhood characteristics may overestimate the value of better schools.  Thus in order to control for variation in neighborhood characteristics, property taxes, and school spending, Black used a hedonic housing price regression – which again describes house sale price as a function of the characteristics of the house and its location – that includes boundary fixed effects which restrict the sample of houses studied to those close to and on opposite sides of school attendance district boundaries.  Instead of the traditional hedonic price function, Black used the formula ln(priceiab) = α + X’iabβ + K’bφ + γtesta + εiab in which boundary dummies (the K term) account for unobserved characteristics shared by houses on either side of the attendance district boundary.  In this way, Black was able to eliminate bias caused by omitted variables such as neighborhood characteristics and property taxes.


Black conducted her calculations twice: once using a simple hedonic housing price regression and once using a hedonic housing price regression which incorporated boundary dummies.  Results from the simple hedonic regression indicate that a 5% increase in the average elementary school test score is associated with a 4.9% increase in house prices in that school district, as shown in Table 1 below.  However, results from the hedonic regression which controlled for omitted variable bias such as neighborhood characteristics show that when houses observed are restricted to those within only 0.15 miles from the boundary of a school attendance zone, a 5% increase in the average elementary school test score is associated with a 2.1% increase in house prices in that school district (Table 1).  This second calculation is roughly half of the estimated effect calculated using the simple hedonic housing price regression.  Thus, in her paper, Black has demonstrated that it is very important to control for neighborhood characteristics by restricting the housing sample used to houses on school attendance boundaries.  Otherwise, one may greatly overestimate the value of school quality as shown by test scores on the prices of housing in the respective district.  Yet, even after controlling for omitted variables, it can be seen that better school quality, as shown by an increase in test scores, has a positive effect on housing prices.

Table 1: Magnitude of Results



Using both the work of Tiebout and Black as background research, John Wulsin more recently conducted a study on the effects of school quality on housing prices in Durham, North Carolina.  In his paper “An Analysis of the Effects of Public School Quality on House Prices in Durham, North Carolina” (2009), Wulsin stated that when families buy a home, they also buy the right for their kids to attend the local public school in that district and that the price of that right is incorporated into the price of the house they purchase.  Like Black, Wulsin used school performance composite test scores, which is the North Carolina Department of Education’s standardized metric for measuring school quality.  Building on Black’s work, Wulsin, too, recognized the importance of using boundary fixed effects to control for neighborhood characteristics in order to ensure that the effects of school quality on housing prices are not overstated.  Wulsin gathered the following data on houses in Durham County: the fair market value of the house, observable characteristics that affect house prices, which school attendance zone the house is in, and the distance the house is to the border of the school attendance zone.  He gathered data from sources such as the Durham Tax Assessors Office and Durham Public Schools.  Wulsin then worked with GIS shapefiles, computer software programs such as ArcMap, StatTransfer, and Stata, and an OLS regression to organize and decipher the data.  After collecting and interpreting his data, Wulsin concluded that “parents do pay more to live in areas with better schools”.  Wulsin’s study found that in Durham Country school districts, a 10% increase in elementary school scores leads to an 11% increase in housing prices, a 10% increase in middle school scores leads to an 11% increase in housing prices, and a 10% increase in high school scores leads to a 5% increase in housing prices.  Thus, it can be seen that an increase in school quality, as measured by test scores, once again leads to an increase in housing prices in that school attendance zone.


As with many other attractive neighborhood qualities, it has been observed that people value the quality of the schools they send their children to.  In fact, a 2000 survey by the Philadelphia City Planning Commission found that the third most important neighborhood characteristic for buyers and sellers with children was the presence of a good school in the area (TRF, 2007).  Many researchers have attempted to determine the exact value consumers place on school quality and have used housing prices as a means of measurement.  As shown by the studies analyzed above, better school quality is correlated to higher housing prices.  Furthermore, this trend has been observed across the United States.  However, there are still several issues that need to be explored further.  Firstly, measuring school quality is often difficult and very subjective.  Many have argued for the use of input-based metrics such as per-pupil spending while others believe that output-based measurements such as test scores are a better indication of school quality.  Future studies should attempt to collect and use a greater range of data on each school observed in order to gain a clearer picture of what makes a school “good quality”.  Furthermore, researchers need to continue to develop better ways to isolate the effects of school quality on housing prices and reduce, with an ultimate goal to eliminate omitted variable bias in which variables such as neighborhood characteristics and property taxes cause an overestimate of the impact of school quality on housing prices.  In conclusion, as each of these papers has shown, the quality of a school has a positive impact on the prices of houses located within that school attendance zone.  This finding is not only important to homeowners and parents, but also to economists and policy makers.  Because it has been found that better quality schools increase the real estate value of houses in their areas, improving schools can be a method for improving neighborhoods and stimulating economic growth.





Works Cited:


Black, Sandra. 1999. “Do Better Schools Matter? Parental Variation of Elementary Education”. Quarterly Journal of Economics, Vol. 114. 4 February 2013.


Owusu-Edusei, Kwame and Molley Espey.  2003. “School Quality and Property Values in Greenville, South Carolina.” Department of Agricultural and Applied Economics, Clemson University. 30 January 2013.


The Reinvestment Fund.  2007. “Schools in the Neighborhood: Are Housing Prices Affected by School Quality?”  Reinvestment Brief: Issue 6. 30 January 2013.


Tiebout, Charles. 1956. “A Pure Theory of Local Expenditures”. Journal of Political Economy. 5 February 2013.


Wulsin, John.  2009. “An Analysis of the Effects of Public School Quality on House Prices in Durham, North Carolina.” Economics Department, University of North Carolina at Chapel Hill.  2 February 2013.






Literature Survey

by Christopher Bradford Bradford_Christopher_2013_LitSurvey


It is my hope to explore in subsequent work the Research Triangle Park (RTP) as a factor for economic and urban growth in and around Durham. Accordingly, the sources chosen here are ones that pertain to some facet of this topic. The Jenkins et al paper explores causes of growth in the proportion of employment in high-technology industry, and proposes that state and local government policies can be effective in achieving high-technology employment growth. Kodrzycki et al investigate the factors associated with resurgent post-industrial cities, finding chief among them competent leadership and long-term collaboration between the private and public sectors. Seeking to understand the institutional contexts that underlie high-tech clusters, Sternberg et al compare federal, state, and non-state cluster policies in the dual cases of the RTP and the Greater Munich area in Bavaria, Germany.


Jenkins, Craig J, Kevin T Leicht and Arthur Jaynes. “Creating High-Technology Growth: High-Tech Employment in U.S. Metropolitan Areas, 1988-1998”. Social Science Quarterly. Vol. 89, No. 2, Pp. 456-481. 2008.


This 2008 study is concerned with the issue of whether state and local policies can induce growth in the proportion of high-technology employment. The authors argue that sustainable economic growth necessitates a transition from older to newer, high-technology industries, especially as traditional high-paying manufacturing jobs continue to be outsourced. Accordingly, the proportion of total employment in high-technology industries is vital for future economic growth, more so than is the pure numerical increase of high-technology jobs, which the authors believe has received far more scholarly interest. Another point of difference between this and other studies is that this work emphasizes the role of agglomeration effects in the proportional growth of high-technology employment, whereas, according to the authors, others have focused primarily on location effects. Jenkins et al find that location effects may explain numerical growth in high-technology employment, but that location effects fail to account for the proportional increase of high-technology jobs.

In their analysis, the authors use data from 291 Metropolitan Statistical Areas (MSAs) from 1988-1998 to construct a generalized linear model that predicts the probability of percentage change of high-technology employment in this time period. Data include population density, 1988 mean high-technology wage, various amenities associated with location effects, and numerous determinants of agglomeration effects (469). The generalized linear model finds many factors to be positively correlated with increasing shares of high-technology employment, among them research parks, public venture capital programs, technology incubators, small business innovation research, and technology grants and loans.

The conclusion that research parks are correlated with an increasing proportion of high-technology employment is significant to my interest in the RTP. Moreover, the authors propose that research parks magnify the effects of private venture capital firms in increasing the share of high-technology employment. Perhaps this amplification in high-technology employment that arises from such a partnership will emerge as a matter of importance in the context of the RTP. The study’s conclusion that “state and local government can play a strategic role in high-technology development” (456) is certainly relevant to the past, present, and future relationship between North Carolina, Durham, and the RTP.

The authors indicate that their study does not address the distribution of benefits from increasing shares of high-technology employment in the labor pool. This has relevance for Durham, where it is unclear to me at present whether unskilled workers that are largely excluded from positions in high-technology employment benefit from growth in this industry. The study also points to other works that address “counterproductive ‘bidding wars’” for investment in high-technology industry (477). Though only mentioned in passing, this appears to be a question of significance for an evaluation of how RTP has and will continue to affect economic and urban growth in Durham.


Kodrzycki, Yolanda K and Ana Patricia Muñoz. “Reinvigorating Springfield’s Economy: Lessons from Resurgent Cities”. Federal Reserve Bank of Boston. Community Affairs Discussion Paper No. 09-7. 2009.


This study is concerned with identifying factors that are associated with economic growth and revitalization in post-industrial cities. The authors investigate 25 municipalities that were similar to Springfield, Massachusetts in 1960. Ten of these cities experienced significant economic growth after restructuring their economy on a post-industrial model. The authors brand these as “resurgent” cities and consider them in detail. In doing so, the authors hope to elucidate potential policy options for catalyzing growth in Springfield. The study is largely qualitative in scope, examining factors responsible for success in various resurgent cities. Among the data used to evaluate urban success are income levels, poverty rates, population figures, educational attainment, and employment in various industries (e.g. manufacturing versus healthcare).

The authors stress leadership and long-term collaboration between the public and private sectors as factors chiefly responsible for the success of the studied resurgent cities. This applies to the case of the RTP; for example, Link and Scott (2003) emphasize the importance of the vision of certain individuals and coalitions to RTP’s success. Kodrzycki et al find that the initial preponderance of manufacturing in a metropolitan area is the factor most correlated with post-industrial growth. Namely, studied cities that were more reliant on manufacturing in 1960 are less likely to experience dynamic post-industrial economic growth. The study notes that each of the 10 resurgent cities studied had more employment in healthcare and in social assistance than in manufacturing by the mid-2000s. This transition from manufacturing to healthcare is a prominent feature of Durham, and will likely come up as a subject of investigation in my future work.

The authors also cite resurgent cities’ focus on cultivating human capital for a transition to a knowledge-based economy. The authors contend that planning for economic growth has experienced a paradigm shift in recent years. Economic policy in the 1980s and 90s emphasized attracting mega-firms to a metropolitan area. In contrast, planners today focus on the human capital necessary to sustain innovation – the quality matters more than the quantity of workers. Also important for post-industrial economic resurgence, the authors argue, is the successful social and economic integration of minority populations. Both of these necessities apply to Durham. The three major universities that anchor the RTP serve to develop and attract human capital. On the other hand, I am currently uncertain about how successfully Durham has integrated its black, Hispanic, and less-educated populations into an economic regime heavily marked by the RTP. This is likely an issue I will explore in further detail.


Sternberg, Rolf, Matthias Kiese and Dennis Stockinger. “Cluster policies in the US and Germany: varieties of capitalism perspective on two high-tech states”. Environment and Planning C: Government and Policy. Vol. 28, Pp. 1063-1082. 2010.


Believing that literature has often taken the regional cluster model of high-tech firms for granted without exploring the context from which these clusters emerge, Sternberg et al search for institutional determinants of success for cluster policies in their study. The authors compare the governmental policies that have nurtured two technological clusters: the Research Triangle Park in North Carolina, and the Greater Munich area in Bavaria, Germany. These two regions were chosen because they epitomize two very different “varieties of capitalism”: a liberal market economy in the case of the RTP, and a coordinated marked economy in the case of the Greater Munich area (1063). As described by Ketels et al (2006), clusters in liberal market economies are typically formed by the private sector and focus on exports. On the other hand, clusters in coordinated market economies are subject to greater national oversight (1067).

This study interprets qualitative data in the form of interviews with 87 cluster policy experts from both North Carolina and Bavaria. Drawing upon these interviews and a survey of literature in the field, the authors describe key features of cluster policy in the United States and North Carolina, and in Germany and Bavaria. Among the differences: the private sector is the driving force for clustering in NC, which is competition-driven, whereas it is the public sector that drives clustering in Bavaria, which is consensus-driven; the state of North Carolina is the main agent of cluster policy for the RTP, whereas the German federal government cooperates closely with the Bavarian state; the US federal government primarily plays a role in cluster policies by awarding financial grants and fostering a favorable business environment, whereas the German federal government has a history of using contests to dole out large grants while largely allowing the state to pursue its own, autonomous cluster policies.

Sternberg et al propose comparative weaknesses of the RTP and Bavarian cluster models. In the case of the RTP, they find that local cluster policies are very dependent upon individual agencies, such as the North Carolina Department of Commerce and the seven regional nonprofit organizations that market the RTP and plan its development. With regards to Germany, the authors propose that clusters often arise through opaque public-sector processes, with resultant difficulties in mobilizing private sector cooperation (1076). This study’s examination of the federal and local policies, as well as the nonprofit actions, that have contributed to the growth of the RTP should be of use to me in later work. However, the authors clarify that an evaluation of whether the technological cluster is in fact an effective model of economic growth is beyond the scope of their study. This question of whether the cluster model, which has been successful for the past half-century for the RTP, will remain useful to Durham and North Carolina for future high-tech industry and economic growth is an issue that I will likely want to consider in later research.




Jenkins, Craig J., Kevin T. Leicht and Arthur Jaynes. “Creating High-Technology Growth: High-Tech Employment in U.S. Metropolitan Areas, 1988-1998”. Social Science Quarterly. Vol. 89, No. 2, Pp. 456-481. 2008.


Ketels, C. H.M., G. Lindqvist, and Ö. Sölvell. 2006. “Cluster Initiatives in Developing and Transition Economies.” Report, Ivory Tower AB, Stockholm, http://www.cluster-research.org/dldevtra.htm.


Kodrzycki, Yolanda K. and Ana Patricia Muñoz. “Reinvigorating Springfield’s Economy: Lessons from Resurgent Cities”. Federal Reserve Bank of Boston. Community Affairs Discussion Paper No. 09-7. 2009.


Link, Albert N and John T Scott. “The Growth of Research Triangle Park”. Small Business Economics. Vol. 20, Pp. 167-175. 2003.


Sternberg, Rolf, Matthias Kiese and Dennis Stockinger. “Cluster policies in the US and Germany: varieties of capitalism perspective on two high-tech states”. Environment and Planning C: Government and Policy. Vol. 28, Pp. 1063-1082. 2010.