Home » Articles posted by Tatiana (Page 2)

Author Archives: Tatiana

Interactions Between Crime and Schooling in the Housing Market

By Billy Marsden    Interactions Between Crime and Schooling in the Housing Market


Arguably the largest and most important purchase that a consumer will make in his or her lifetime is a house. Housing costs make up a considerable portion of one’s annual spending. According to the Wall Street Journal, Americans spend an average of around 30% of their income on housing. Thus, a lot of time and energy goes into determining exactly what one wants in a new home. A homebuyer must determine exactly what qualities they want to pay a premium for, and which to forego. The buyer must figure out their preferences regarding the location of the house, including the city, neighborhood, and proximity to certain amenities, as well as the desired physical attributes, including size, number of bedrooms and bathrooms, and general aesthetic style.

Thus, significant research has gone into determining exactly what characteristics affect housing prices. Homeowners, construction companies, real estate firms, and the government alike could all benefit from better understanding the preferences of homebuyers, whether that be knowing exactly how much they’ll pay for an extra 300 square feet or from the addition of a pool into the backyard. However, due to the complexities of housing characteristics and the pure number of different traits that go into a purchasing decision, it becomes much more difficult to pinpoint exactly how much consumers will pay for different attributes.

Technical Review

Two such characteristics that have received a significant amount of attention in academic and technical papers are crime and school quality. Theoretically, homebuyers would pay a premium in order to live in a physically safe environment to avoid crime, as well as the mental security of knowing that they are safe. Additionally, parents will pay a premium to live within the district boundaries of the best public schools, allowing their children free access to quality education. While these make sense theoretically, many economists and psychologists have attempted to quantify these effects.

One study conducted by Stephanie Swift of the University of Troy examines the effect of crime on housing prices in Jacksonville, Florida. Segmenting by violent and non-violent crime, the researcher runs a simple linear regression using certain housing characteristics, including square

footage and other basic amenities, and local crime rates to predict housing prices. Based on the regression output, both violent and nonviolent crime rates had a significant effect on housing prices. However, nonviolent crime actually had a negative effect on housing prices. The paper addresses the fact that this could potentially be due to the fact robberies and property damage occur in areas with more expensive real estate.

A second study that addresses this dual relationship between nonviolent crimes and housing prices is a study published by Keith Ihlanfeldt and Tom Mayock of the Economics Department at Florida State University. The 2009 analysis focuses on the effects of crime on housing prices in Miami-Dade County. Through a nine-year time-series regression analysis, the researchers attempt to investigate the effect of changes of crime rates, segmented by violent and property crimes, on changes in property values. As the researchers argue, crime is never treated as an endogenous variable when regressed with housing prices, despite the fact that housing prices may affect crime rates, specifically nonviolent ones. Thus, they not only use a simple OLS regression, but also an instrumentation approach to attempt to derive the causality of crime rates on housing prices.

Their regressions provide overwhelming evidence that property crime has little impact on housing prices, while violent crime rates have a significant impact. Homeowners pay a significant premium to avoid living in violent crime-ridden areas. For a 1% increase in violent crime rates, holding all other factors constant, housing prices decreased by .25%.

While crime is one significant factor that homebuyers may consider when thinking about a new purchase, local public school quality is another key factor, and one that has also been the subject of significant academic research. Much like crime, school quality is highly correlated with other unobservable characteristics, including quality of the neighborhood, so it is difficult to determine causality via regression analysis.

One such paper that attempts to quantify the effect of school quality on housing prices is Which School Attributes Matter, a paper by John Clapp, Anupam Nanda, and Stephen Ross of the University of Connecticut. Using housing and schooling data from the state of Connecticut, along with neighborhood demographic and socio-economic qualities, the researchers attempt to determine the true effect of school quality on the housing market. A shortfall of past studies, according to these researchers, is their inability to control for unobservable neighborhood characteristics. Thus, this analysis incorporates a fixed effects model, allowing them to control for all unobservable neighborhood traits that may have been masked into the school effect in previous models. Comparing the general OLS model to the neighborhood fixed effects model, the effect of school

quality, as measured by math test scores, is reduced to 20% of its original value. The magnitude of the school quality coefficient reduces from .074 to .013 when the neighborhood fixed effect is introduced. Without controlling for neighborhood characteristics, the importance of test scores is overstated by a factor of five. However, the effect is positive and statistically significant in both cases, implying the quality of schools is a significant factor when purchasing a home.

While many studies have looked at both of these elements separately, or have even incorporated both in the same regression, no analysis has been done connecting the two. It could be possible that those who put a premium on schooling put a different premium on crime than those who choose not to put a premium on schooling. Thus, my regression will not only incorporate schooling and crime data and their effect on housing prices, but an interaction between the two. Based on the data collection and regression output, we will be able to determine the premium on crime that homebuyers who live in high performing schooling areas, low performing schooling areas, and areas without any schools have, and whether any difference exists among these groups.

Data and Methodology

The data used in the analysis were collected from a variety of online sources. The housing data, including housing price, number of bedrooms, number of bathrooms, housing square footage, and lot size were collected from Zillow. The 98 analyzed houses were a random collection of houses sold in the last month in Durham.

Houses were then segmented into 3 different groups based on their proximity to schools. Schools given a School Category rating of 0 did not have any schools within a half-mile. Schools with a School Category rating of 1 were in close proximity to a low performing school, and those within a half mile of a high performing school were given a School Category rating of 2.

The quality of the school was determined through a rating from greatschools.org. The website aggregates publicly available test score data and assigns all US public schools a rating between 1 and 10. Schools that received a rating between 1 and 5 were assigned to the low performing group, and schools with a rating between 6 and 10 were deemed high performing. Due to the fact that test score information is only available for public schools, private schools and houses located near private schools were omitted.

Crime statistics were collected through the Durham Crime Mapper website. For each house, the total number of crimes that occurred within a half-mile of the house between January and March of 2014 were counted. Violent and nonviolent crimes were bundled together, which includes arson, assault, burglary, homicide, larceny, motor vehicle theft, robbery, and rape.

With the collected data, OLS regressions were run with the log of housing price as the response variable. A combination of number of bedrooms, number of bathrooms, housing square footage, housing square footage squared, lot square footage, crime rates, school category, and an interaction between crime rates and school category were the explanatory variables. The 3 primary models are shown below. The first model is the most basic, while the second model incorporates the interaction term of interest, and the third model includes squared footage squared to help with the fit of the model.

Model 1:

LogP_i = beta_0+beta_1*Bed_i+beta_2*Bath_i+beta_3*SF_i+beta_4*Log(Lot_i)+


Model 2:

LogP_i = beta_0+beta_1*Bed_i+beta_2*Bath_i+beta_3*SF_i+beta_4*Log(Lot_i)+


Model 3:

LogP_i = beta_0+beta_1*Bed_i+beta_2*Bath_i+beta_3*SF_i+beta_4*SF^2+



The results of the model are shown in Table 1. Model 3 appeared to provide the best fit for the data based on the r-squared and residual plots. Interpreting the coefficients from model 3 yielded insights into which characteristics affect housing prices. Unsurprisingly, bedrooms, bathrooms, and housing size all have positive coefficients. The log of lot size has a slightly negative but statistically insignificant coefficient, rendering it not useful in predicting housing price. In all models, crime has a negative and statistical significant coefficient. This implies that higher levels of crime are a predictor for lower housing prices. Additionally, the category 1 and 2 school variables have positive and mostly statistically significant values. The category 2 coefficient is generally larger than the category 1 coefficient, implying that homebuyers will pay a larger premium for high performing schools than low performing schools. In model 3, the School: Category 1 coefficient is .15 and the School: Category 2 coefficient is .16. The interpretation of this coefficient is that on average, holding all other factors constant, the log of housing price will increase by .15 for low performing schools and .16 for high performing schools in relation to houses with no schools in nearby proximity.
Finally, when interpreting the coefficient for the interactions, we receive a counterintuitive result. For model 3, the interaction between crime and school quality has a positive coefficient. This means to determine the true coefficient of crime on housing prices for houses sold near schools, the sum of both the original crime coefficient and the interaction must be taken. The true coefficients of crime on housing prices for each category of school from model 3 are shown below.
Screen Shot 2014-04-08 at 11.58.29 PM
For houses not near any type of school, crime is negatively associated with housing price, an expected result. However, for houses near schools, higher levels of crime predict higher property values. Thus, the model predicts that homeowners purchasing homes near schools are actually less sensitive to crime than those not purchasing near schools.
While our model predicts that homeowners buying near schools appear to put a premium on crime, this may point out a flaw in the model. It is very difficult to imply causality in a regression model. Instead, it is necessary to control for all factors that would affect housing prices in order to imply the causality of crime on housing prices. For example, in the paper Which School Attributes Matter, the researchers controlled for unobservable neighborhood characteristics, which diminished the importance of school effect. The model in this paper does not incorporate fixed effects, meaning the unobservable neighborhood characteristics that would normally affect housing prices are bundled into other coefficients.
When looking at the crime coefficients for houses in category 1 and category 2 schools, we see a positive coefficient. This probably does not imply that homebuyers put a premium on crime, but instead that high amounts of crime are positively correlated with wealthier neighborhoods.
Another flaw in the model is the fact that violent and nonviolent crimes are not segmented. If they had been segmented, then we could have more appropriately determined the effect of violent crimes (homicide, rape, assault) and nonviolent and property-based crime on housing prices.
A potential method to imply causality of crime on housing prices is to use a time-series regression, something not utilized in these models. Using crime data from the previous year, we could attempt to view the causal relationship on that year’s housing sales data. However, this would require a more complex time-dependent linear model and a large sample size. The model could look something like the formula shown below.

LogP_i = beta_0+beta_1*Bed_i+beta_2*Bath_i+beta_3*SF_i+beta_4*SF^2+



An additional flaw is the sample size used in this analysis. 98 data points is relatively small to achieve statistically significant results given the number of explanatory variables used, especially when interactions are incorporated. So, while many of the results were not statistically significant, more coefficients may have been with a larger sample size.
Given our model creation it is plausible to imply correlation between housing and crime. Crime is negative and statistically significant when on analyzing housing that is not within a half-mile radius of a school. However, the picture becomes more complex when analyzing homes near schools. The coefficient turns positive, though not statistically significant. Instead of assuming that homebuyers near schools put a premium on crime when making a purchasing decision, it is more likely to assume that higher crime rates are correlated with higher housing prices due to the fact that they are more prone to nonviolent property crimes, including robberies, motor vehicle theft, and larceny. With a more sophisticated model and larger dataset discussed in the limitations above, it would be possible to more correctly derive the sensitivities of homebuyers to crime given their proximity to schools. However, with the simple regression model derived in this paper, we can only imply correlation.


Table 1
Screen Shot 2014-04-09 at 12.02.11 AM

Works Cited:
1. 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. Print.
2. “Durham Crime Mapper.” Durham Crime Mapper. N.p., n.d. Web. 24 Mar. 2014.
3. “How Much You Should Spend on a Home.” Personal Finance RSS. Wall Street Journal, n.d. Web. 24 Mar. 2014.
4. Ihlanfeldt, Keith, and Tom Mayock. “Crime and Housing Prices.” Department of Economics and DeVoe Moore Center: Florida State University (2009): n. pag. Web.
5. “Join GreatSchools.” GreatSchools. N.p., n.d. Web. 23 Mar. 2014.
Swift, Stephanie. “Do Crime Rates Affect Housing Prices?” Troy University Economics Department (n.d.): n. pag. Abstract. (2005): n. pag. Print.
6. “Zillow: Real Estate, Apartments, Mortgage & Home Values in the US.” Zillow. N.p., n.d. Web. 23
Mar. 2014.

The Millennial Generation and Durham’s Housing Market

By Gabrielle M. Ware  The Millennial Generation and Durham’s Housing Market


In the most recent decade, Durham has experienced a large influx of members of Generation Y, otherwise known as the Millennials. Millennials, defined throughout this paper as those born between 1980 and 1998 are currently between ages 16 and 34 and described as educated, diverse, creative, connected and open to change. Forecasted to be the most educated generation in the United States’ history, greater than one-third, or 37%, are unemployed or out of the workforce as a result of the nation’s 2008 Financial Crisis1. In the search for financial stability, members of this generation are drawn to Durham and its surrounding area because of the high caliber universities located there, as well as the promising jobs in healthcare, pharmaceuticals, and other professions found in Research Triangle Park. There is definite incentive for Durham officials to continue to attract and retain members of Generation Y to sustain the county’s high growth rate. The rise of an educated, and hopefully successful, generation will not only bring more young professionals into Durham, but will also benefit businesses and Durham’s overall economy. Typically, Millennials want to live in high-density neighborhoods, accessible to centers of business and employment either by walking or public transportation, and desire affordable urban rental spaces as they  try to achieve financial stability. As Durham’s housing market becomes dominated by Millennials, demand for the aforementioned types of housing will certainly increase and it will be interesting to examine how Durham officials will respond to this changing demand.

The goal of this paper is to explore which communities and neighborhoods Millennials are most attracted to and what factors determine their housing choice. To achieve this goal I will compare neighborhoods with the highest and lowest proportions of Millennials in Durham County and will go over the study of Millennials in Philadelphia conducted by the PEW Charitable Trusts. Because patterns have shown that members of this generation are more concerned with the quality of their surroundings than of their housing itself, I will focus primarily on which neighborhood and county qualities will play the largest role in these buyers’ decisions.

Millennials and the Housing Market

Over the past five years, as the majority of members of Generation Y have reached their twenties and thirties, there has been increasing concern about the housing market. In the decades of their lives in which previous generations have driven growth in homeownership, many Millennials are putting off owning their own home due to massive amounts of student loans they must pay back and uncertainty in the job market. Additionally, members of this generation have been known to switch jobs more often than their parents, spending only about two years in each position before moving on to a new one. This creates another deterrent to homeownership, as Millennials prefer not to be tied down in any sense, and many members of Generation Y are even more hesitant after witnessing the suffering of their slightly older friends, many of whom had just bought their first home when the housing bubble burst. Furthermore, Millennials are marrying much later than previous generations with 75% still single between the ages of 18 and 28, compared to 67% of Generation X and only 52% of their baby boomer parents at the same age. This could also be working to slow home purchases, as marriage is a milestone that often coincides with homeownership. However, this must be taken in stride as members of Generation Y are also much more open to couples living together before marriage, and much more social in general, than previous generations were. This combination of both financial and preferential reasons against owning a home has led to largest drop in homeownership of any age group in the United States for 24 to 33 years old between 2005 and 2011. Despite this decline, approximately 93% of Millennial renters still plan to own their own home at some point in the future, indicating that this decline is driven by lack of funds and may only be temporary. Still, many experts point out that this reluctance to purchase now may leave members of Generation Y stuck in a far less favorable real estate market down the road and unable to reverse the trend. Although there are many benefits to purchasing a home now, with Trulia estimating that it is 35% cheaper to own a home than rent in America’s largest cities due to slightly undervalued homes and interest rates that are still historically low, many Millennials have insufficient funds and are too afraid of the commitment to one place to make a down payment.

In addition to whether or not they will rent or buy, another important factor to consider when assessing the housing needs of this generation is whether or not needs are likely to change in the long term and by how much. While it is true that Millennials are currently seeking out small and affordable housing in high-density city centers, with only 14% living in rural areas, these needs may change in the next five to ten years. A look at the population between ages 20 to 34 in Philadelphia gives some insight into how this generation’s needs may be rapidly changing. During the period between 2006 and 2012, Philadelphia experienced a larger increase, totaling 6.1%, in members of Generation Y as a percentage of its total population than any other US city; however, this growth may only be short lived as slightly over half of the Millennials living in Philadelphia plan to relocate in the next five to ten years and would definitely not recommend the city as a good place to raise children. The there main considerations survey participants cited for planning to leave the city were jobs/career, school quality/child upbringing, and the crime/safety/drugs. A full list and more detailed comparison between Millennials and older generations can be found in Table 1 (Appendix at end). From the case of Philadelphia, it is implied that successful policy to attract and retain Millennials will need to be adapted as the generation’s needs change and that Durham may benefit from incorporating both long-term and short-term strategies.

Millennials in Durham

As shown in Figure 1, Millennials have located throughout Durham in accordance with what is predicted by patterns and preferences of the generation. The map shows Durham County, divided into 60 smaller census tracts as defined by the US Census Bureau for more uniform and complete data. The five tracts with the highest proportions of 16 to 34 year olds are marked with green diamonds, while the five tracts with the lowest proportions of 16 to 34 years olds are marked with red circles. Even from a quick glance, it is apparent that the five tracts with the highest proportions of Millennials are clustered near the city-center while the five tracts with the lowest proportions of Millennials are scattered around the Durham County’s edges. Table 2, created by data obtained from the US Census Bureau American Community Survey five year estimates from 2008 to 2012, show some of the most relevant characteristics of each of these ten tracts and the differences between the two groups can be used to demonstrate which factors have led Millennials to distribute themselves as they have throughout Durham.

When examining the five census tracts with the highest proportions of Generation Y members living in them, it is not surprising that two, 15.03 and 15.01, contain Duke’s East and West Campuses, respectively, two, 15.02 and 4.02, correspond with areas touching Duke’s Campuses and the final tract, 13.03, contains North Carolina Central University’s campus. When examining the eight included characteristics of each tract, including population density, number of occupied rental units, median gross rent, median household income, proportion of residents with at least some college education, median commuting time to work, and the proportions of residents that are black and Hispanic, any extreme outliers can be explained by the presence of specific institutions in or adjacent to the tracts. For example, 100 percent of the population in tract 15.03 has at least some college education and the median household income, $67,454, is uncharacteristically higher than the surrounding area as a result of the presence of Duke’s East Campus, filling almost the entire tract.  This fact also explains the lack of data to estimate the median gross rent. When examining the tracts with lowest proportions of Generation Y members, it is also not surprising that four out of five tracts feature either a golf or country club or luxury development, while the tract with the absolute lowest proportion of Millennials, 9801, is fairly non-residential including two major highways and Research Triangle Park. Again, the aforementioned features of each tract can explain away any outliers, such as the extremely low population density and proportion of Millennials in tract 9801, as young adults commute to work in RTP, but do not live there.

The differences between the groups with high and low proportions of Millennials were also not surprising. Because of their documented affinity for high-density city centers, the clustering seen in Figure 1 and higher population densities of the tracts with high proportions of Millennials were to be expected. All of these areas are within walking distance from some sort of city center and are accessible by public transportation. This desire to be close to some sort of community is also reflected in the relatively shorter commuting times to work places in tracts with higher proportions of Millennials. Additionally, it is consistent with trends in other cities that tracts with high proportions of Generation Y members have many units available for rent at affordable prices, as many Millennials are burdened with student loans and trying to be debt free and secure some sort of emergency fund before purchasing a home10. It can also be seen that tracts with higher proportions of Millennials are more diverse, another defining characteristic of this generation.

What Does this Mean for Durham?

Because of the great universities located in and surrounding Durham County, as well as the promising professional opportunities in Research Triangle Park, Durham has already begun to attract members of Generation Y at high rates. Still, Durham has not experienced an influx of the same magnitude as cities like Philadelphia, Boston, Nashville, and Baltimore11. Because of this, policy for Durham should take a two-sided approach. First, in the short-run, Durham authorities should learn from and attempt to mimic some of the conditions in the aforementioned cities and parts of Durham that have already attracted high numbers of Millennials. Understandably, it is impossible for Durham to match the size, density, and connectedness of a city like Boston in the short run, but it can take some steps in the right direction. As seen in Figure 1, the most important factor for members of Generation Y when deciding whether or not to relocate is for job or career opportunities. It is clear that in order to attract Millennials, the city must make it worth it for them professionally. Durham’s government can help achieve this by offering incentives for companies that provide professional opportunities located in Durham or within reach of the city’s public transportation.

Once attracting Millennials to the opportunities Durham has to offer, the city must accommodate the living needs of Generation Y. Durham authorities must make an effort either to expand upon community centers already in existence with more affordable housing or to plan additional centers that will be walkable to some degree, and have access to centers of employment. As an extremely connected and collaborative generation, members will also be attracted to high concentrations of young adults, as well as youthful, liberal, and creative ambiances. The addition of a 26-foot sky-scraper at the corner of Main and Parrish Streets, a redevelopment project led by Austin Lawrence Partners set to break ground in the Fall, is a step in the right direction for Durham. The high-rise, which will consist of a mix of retail, office, and residential space and is already in surrounded by award-winning restaurants, theaters, and sports centers, will be certain to attract young and professional adults and solidify the city’s downtown area, creating a high energy and urban environment.

In addition to meeting the needs of Millennials today, Durham officials should learn from the case of Philadelphia and try to anticipate how the needs of Generation Y members will change in the next five to ten years. As shown in Table 1, young adults between the ages of 20 and 34 are concerned about living in a family oriented city that has good school districts and is safe in addition to having good career opportunities in the future. It is also true, that members of this age group still plan to invest in a home at some point down the road. Durham authorities should plan ahead for this need, so when the time comes to make the more permanent decision of buying a home and settling down with a family, Millennials do not choose to locate elsewhere. To do this, officials must work to ensure a safer and cleaner city, as well as school quality that rivals not only neighboring districts, but is competitive on a national level. The low quality of Durham Public schools in relation to Orange and Chatham County school districts will be one of Durham’s biggest challenges in retaining Millennials in the next decade. Possible solutions include raising teacher salaries as a way to attract better teachers, which may enhance school quality, or increasing per pupil spending; however, these solutions will also make Durham a less affordable place to live. Additionally, there is reasonable doubt to believe that increasing per pupil spending will translate into increased school quality in the homebuyer’s eyes12. Furthermore, Durham officials must work to create clean and safe neighborhoods consisting of affordable homes so that educated professionals will make the more permanent decision to stay in Durham. Over time, this may translate into better school quality for Durham, as the children of educated parents will be far less expensive to teach.

Overall, Durham’s goals to create a better living environment for Millennials are complementary to one another and cannot be achieved alone. Professional job opportunities and urban centers must exist to attract members of Generation Y initially, and then safe and affordable housing communities must follow to retain them. Better school quality will hopefully ensue as Durham begins to consist of a more educated demographic. Policy throughout the next decade will be critical in whether or not Durham will continue to sustain high growth levels.


Table 1: Main Reasons for Expecting to Leave Philadelphia in the Next 5-10 Years for Millennials in Contrast to Older Adults

Screen Shot 2014-04-08 at 11.23.20 PM


Figure 1: 5 Census Tracts Least and Most Populated by Millennials

Screen Shot 2014-04-08 at 11.24.13 PM


Table 2: The 5 Census Tracts in Durham with the Most and Least Millennials and their Characteristics

Screen Shot 2014-04-08 at 11.25.12 PM



1. Bissonnette, Zac. “Homeownership: The Elusive American Dream for Millennials.”CNBC.com. N.p., 28 Nov. 2013. Web. 23 Mar. 2014.


2. Bracken, David. “Report: Triangle Home Prices Kept Rising in January.”&newsobserver.com (n.d.): n. pag. Web. 20 Mar. 2014.


3. 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 Economics63.2 (2008): 451-466.

http://digitalcommons.uconn.edu/cgi/viewcontent.cgi?article=1094&context=econ_wpap ers

4. DeBruyn, Jason. “Pew: Millennials with a College Degree Earn 38% More than Those without.” Triangle Business Journal. N.p., 12 Feb. 2014. Web. 20 Mar. 2014.


5. Durham, North Carolina Government. Durham City-County Planning Department.

Forecasting Land Use and Trends. By Daniel Band and Holli Safi. N.p., Apr. 2013. Web. Mar.2014.


6. “Millennials in Philadelphia: A Promising but Fragile Boom.” The PEW Charitable Trust(2014): n. pag. Web. 20 Mar. 2014.

http://www .pewtrusts.org/our_work_report_detail_full.aspx?id=85899534472>.

7. Millennials: Confident. Connected. Open to Change. Rep. Pew Research Centers Social Demographic Trends Project RSS, 24 Feb. 2010. Web. 21 Mar. 2014.

<http://www .pewsocialtrends.org/2010/02/24/millennials-confident-connected-open-to-change/>.

8. Roth, Bryan. “The Future of Duke’s Workforce.” DukeToday (2014): n. pag. Web. 20 Mar. 2014.


Durham Tour by David Lillington


I decided to begin my tour on East Pettigrew Road; little did I know that this would be one of the most economically depressed areas I would visit. I noticed that I was entering an area less developed than Main Street and its surrounding crossroads, especially after crossing under the train tracks on Roxboro. The entire left side of the street followed train tracks, dead grass bordering the curb up the hill to the train. The right side of the street at Roxboro did have a sidewalk, which I followed past the Venable Center, seemingly a tobacco building recently converted into office space. I felt alone but rather safe in this area. It was not until I crossed the intersection of the on/off ramps of Highway 147 that I felt the area changing. The road became more beaten up, probably due to the fact that large transport trucks use Pettigrew as their route (denoted by signs) and houses built along the highway seemed neglected. I walked past a house (one of many) that had been boarded up and whilst vacant had been vandalized with a BB gun (Appendix A1). It was perched on stacks of brick (Appendix A2); perhaps it was a mobile home, where grasses and plants had begun to grow in the area where a porch had once existed. Yards consisted of patchy brown and yellow grass that faced the train tracks on the opposite side of the street. At some point a train came tumbling down the tracks blaring its deafening horn – this would not be a comfortable place to own a home. Based on conditions of homes and types of businesses in the area, I could hypothesize that this was a low- income area of people of color. Businesses included a hair salon, possibly closed forever, named “Chloe’s too – a little touch of soul”, a tattoo shop named “The Inkwell – Tattoos that Hurt So Good”, a tire shop, “Llanteria Padilla – New and Used Tires”, and a taquería. The only people I had seen were two African American women walking a baby in a stroller, two Caucasian men cutting a tree branch way from power lines, and three African American men smoking outside the Brenntag building.


Just before Maple St on E Main a shorter African American woman approached me and said “Sir, can I ask you a question? Do you have any spare change? A dollar?” Admittedly I could not understand her well. I turned on Maple and noticed that the houses were nicer compared to those on Pettigrew, with paved roads in the neighborhood instead of dirt. Roofs were thin, in many cases sagging, and drainage in the street was poor – sidewalks were still mudded from the rain the day before. The speed limit sign on the street was covered with a trash bag and many houses were boarded up or empty. One retaining wall was falling down and one house had a notice from the city

to clean their lot of debris. Cars were mainly old trucks and sedans. I did see one early 90s sedan that had a hydraulic system installed (Appendix A3), or was raised, and had been painted Tar Heel blue and had UNC flags flying off its windows. An African American man rocking on his front porch stared at me as I walked up the street. Another yelled something incomprehensible at me as I walked past him. I passed a two bedroom, one bath, 865 square foot house for sale on Liberty and Maple at $35,9001. Spruce Street felt much the same as Maple, though one might say the homes were a fraction nicer. A large church, the We Church, owns a stretch of the left side of Spruce Street, giving a sense of more space that Maple did not possess on this particular block. The houses that existed on the whole stretch of Spruce, however, were quite close together. I did not see any construction/remodeling of houses in this neighborhood. It seemed to be a lower-class African American neighborhood.


This had to be one of the most interesting neighborhoods I had visited on my tour. A friend’s girlfriend lives in this area and she had told me that gentrification was occurring. House prices ranged from $14,100 (610 Canal St., right on the other side of Elizabeth) to $299,900 (604 Primitive St., originally $315,000 and two blocks from 610 Canal St.)2 and there was construction/remodeling on a good number of houses (Appendix A4). The neighborhood consisted of more raised foundation, older homes placed close together much like those on Maple and Spruce, however these seemed to be kept in better condition. I saw a mix of older African Americans and younger Caucasian residents; perhaps this is an area transitioning from a lower income, uneducated neighborhood to one that is more educated.


On Geer Street houses were a tad more spread out than the other neighborhoods I had been in, but they were still of the craftsman or Tudor style. On most of Geer St I saw many signs for businesses in Spanish. The area consisted of a mix of craftsman and mid-century modern homes in okay condition. At the corner of Roxboro and Geer I noticed a mini-mart that advertised that it accepted food stamps, suggesting that this was not a high-income neighborhood. Cars were older, mid-range brands and there was a mix of all races from what I could see. Nearby there were three identical homes boarded up, all next to each other. Arriving near Washington Street I noticed a Latino market on the right, soon after Motorco on the left and a crossfit gym further down. It is clear that this stretch of Geer closer to downtown has been commercialized and gentrified.


I felt as though this area had been gentrified, because although it was bordered by a lot of apartment buildings on Trinity Avenue, most of the older homes were in relatively good shape. There was the occasional house that was falling apart and in need of redoing, but for the most part it seemed that most homes had been renovated. On the corner of Trinity Avenue and Washington Street an African American woman stopped me and asked me for a bite to eat (again I could not understand her the first time). I entered into the west part of the neighborhood at this point and houses were very small on small lots. Despite the area’s name “Old North Durham” the houses on the west side seemed newer, perhaps from the 50s, 60s, or 70s. It was a quiet, treed neighborhood without train and highway noise. I saw a younger Caucasian man walking his dog here. Cars seemed to be mid-range, older sedans for the most part. It was not until I made my way over to Mangum Street that I encountered the more stately southern Victorian and craftsman homes. These were intermittently mixed with smaller homes. It seemed to me that an educated, lower middle class population and/or a middle-class population inhabited this neighborhood.


Weaver Street at its intersection with Cornwallis Road began with apartment complexes spread out far from each other by grass lawns and parking lots. They seemed to have been built anywhere from 1950 to 1980 and seemed to have an African American population, based on my observation of three African American people. Cars in the parking lots surrounding apartment buildings were older mid-range sedans. After passing this group of apartments and a stretch of woods I came upon a development of split-level homes that seemed to be built around the same time as the apartment complexes down the street. These houses were of medium size on larger lots and tended to be rather colorful. I noticed again that many residents were African American and that cars tended to be middle of the road and of varying size. Some roofs were in need of work as they were sagging, and I did notice an older house that had been boarded up. The street came to a dead end and backed up to a forest, allowing the last residents on the street breathing room. Perhaps people of the lower-middle class inhabited this area.


The majority of buildings on campus were very new, perhaps built within the last ten to fifteen years. The campus was spotless: buildings were very well maintained and their large glass windows were kept clean. The grass lawns were also green relative to those that I had seen in other neighborhoods. There was a large, beautiful football field and track fenced from the outside road (as is the entire university). Cars seemed to be newer mid-range sedans, coupés, or older luxury sedans. I noticed a very large African American population at the university. I visited on a Friday when classes were in session with plenty of students to be seen and I counted three Caucasian students during the time I spent there. Houses surrounding the university were simple but in good condition. I noticed a mix of craftsman and mid-century architectures. I would guess that the population here consists of perhaps a lower-income, educated population, such as students.


South Street was rather close to NCCU’s campus, however there was quite a large change upon entering this neighborhood. I saw four individuals in this neighborhood, three of which were Latino. The houses were older and run-down, some even had bars protecting their windows and doors from break-ins, and others were boarded up. As my friend drove me up the street, she pointed out a broken umbrella hanging from power lines outside of a duplex (Appendix A5). Being from Los Angeles, I had seen objects hanging from power lines many times and I was told that it meant that either a drug dealer was nearby or that it was gang territory. There were small single family and multi-family homes that were placed close together. Most cars were older low to mid- range sedans. South Street abruptly ended just before University Drive due to construction. Professor Becker added that is an area about to transition to a more stable lower-middle income community. This will be done through a project that encourages home ownership.


East Forest Hills Road was by far the most beautiful street that I had visited. Secluded, large homes sat on well-maintained, big lots; even the street wound in such a way to provide privacy to residents. Cars in the area were new mid-range or luxury sedans and SUV’s and some houses had one or more sitting in the driveway. Note this was on a working day, Friday, around 1:30 pm. Perhaps this is a community of families or couples that is wealthy enough for only one spouse to have to work. Houses were perched above the street, giving a figurative message of being upper middle class. I saw one person in the neighborhood, a blonde, Caucasian, young woman decked out in exercise gear. She was walking along the street. Despite being very close to South Street, this community had a completely different feeling. There was also a park that separated the neighborhood from the main street.


Kent Street continued the theme of larger wooded lots, however the houses were much simpler and smaller. I noticed ranch and split level homes most likely from the 50s. There was no sidewalk in the neighborhood and I did not see any residents out. These houses were set back from the road thus giving privacy to residents. Based on the newer mid-range cars that I saw and good condition of houses this neighborhood seemed to be inhabited by middle class residents. Speed bumps slowed down drivers on the road and kept the roads safer, suggesting that residents might have younger children living with them.


I found Kent Street to be a dividing line of Bivins Street. The west side did not have the bigger, more private lots that the east side of the street had. The houses also became larger as I headed farther east. Homes on the west side were quite small and some were even run down. A large modern home stuck out amidst the smaller bungalows. Some yards were overgrown and others were kept nicely. I saw two older Caucasian women walking a dog on the west side of Bivins turning on Kent Street. Most cars were a mix of old and new mid range cars. I did see one BMW X3 SUV. A strikingly new development of houses on the west side could be seen up a small cul-de-sac and did not really blend in with the neighborhood. Professor Becker added that this is a racially mixed neighborhood that has been gentrified over the past few years.


“1108 Liberty Street”. Realtor.com. Accessed January 15, 2014.

“Area in 502 Gray Avenue”. Realtor.com Accessed January 15, 2014.


A1. BB gun bullet holes in siding of vacant house on Pettigrew Street. Author’s own.

Screen Shot 2014-04-03 at 10.31.10 PM

A2. Home sitting on top of stacked bricks on Pettigrew Street. Author’s own.

Screen Shot 2014-04-03 at 10.32.27 PM

A3. Car on Maple Street with hydraulics and UNC flags. Author’s own.

Screen Shot 2014-04-03 at 10.33.16 PM

A4. Victorian house being remodeled in the neighborhood between Roxboro, Geer, Elizabeth, and Holloway. Author’s own.

Screen Shot 2014-04-03 at 10.34.23 PM

A5. Broken umbrella hanging on power lines on South Street. Author’s own.

Screen Shot 2014-04-03 at 10.35.24 PM



Durham Tour by Spencer Rasmussen

Location 1: Cole Mill Rd. and Stoneybrook

This area contained some of the nicest houses that I visited, but it also had great disparity in the houses depending upon which side of Cole Mill Rd. the houses were on.  The houses that were on the side of Cole Mill Rd. with the golf house were very nice and sat on large properties.  In addition, these houses had very well kept lawns, had ornamental stone decorations, and the driveways had some of the nicer brands of cars.  I am not sure if the houses sitting on the golf course made up members of a country club, but I could very easily see this as being the case.  On the opposite side of Cole Mill Rd. the houses were part of a small development called Stoney Brook Cottages.  These houses were substantially smaller than the ones that sat on the golf course, but they were still nice houses with much smaller lots and considerably smaller lawns. The majority of houses on both sides of the street were two-story houses.





Location 2: Northgate Mall

From my own previous experience I know that Northgate Mall is not nearly as nice as Southpoint Mall, which I will talk about later.  When visiting the mall it was clear that there are many vacant storefronts throughout the mall, which shows that it is not getting nearly the foot traffic that it was originally expected.  The mall does contain a fairly nice movie theater that always seems to be crowded.  In the parking lots the cars that I noticed tended to be older models and not nearly as nice as ones seen at Southpoint.  In addition, I found the layout of the mall to be a little odd, with very limited views of the actual mall because of parking structures and more recently built buildings.  On the backside of the mall there is also a strip mall that shares the parking lot with Northgate.  This building also seems to be a little worn-down.

Location 3: Old North Durham neighborhood

Old North Durham was a pretty interesting mixture of decrepit one-story houses and some newer, nicer two-story houses.  Some of the really bad houses were clearly vacant as they had their doors and windows boarded up.  This area also seemed to have nicer cars relative to the houses: I saw two BMWs, two brand new Lexus’ and multiple customized trucks and low-riders.  Also when I examined the local convenience store I noticed that it had bars on all of the windows and had a sign hanging outside that said that it accepted food stamps.  From this single observation it was easy to tell that this area had undergone some economic hardships. One interesting aspect of this area was that there were a few churches all within a couple blocks of each other.  The area also housed a Trosa building, which is a residential option for substance recovering abusers.  There was also a community garden in the area, which could mean that the area had a larger sense of community than others.




Location 4: Maple from E Main to Liberty; then Spruce from Liberty to Juniper

This area reminded me slightly of the Old North Durham Neighborhood; it had a few houses that were boarded up with no trespassing signs on the fronts, had customized trucks and cars, and contained mostly one-story houses.  Although some of the houses were abandoned and the lots that the houses were on were quite small, people still seemed to keep their lawns very well kept.  There was also very little spacing between neighboring houses and also houses and the street.  Most of the divisions between houses were just chain-linked fences that were a little taller than waist high.  I also saw two different people walking around drinking from bottles concealed within brown paper bags at about 1:00 pm.  There were also two schools that were relatively close by one another (presumably a middle and high school) that looked to be old but still well maintained.

 Location 5: Junction Rd., from Holloway to Geer

This was one of the worst locations to live in my opinion because all of the houses on the street were directly across the street from train tracks and on the other side of the train tracks there were multiple industrial buildings.  With all of the traffic and noise that these two locations produced the land value of these properties has to be on the lower end.  Most of the houses in this neighborhood were single story houses with a couple of nicer two storied ones sprinkled in.  Lots of the houses had multiple personal items in the front yard: lawn chairs, children’s toys, jet skis and even a bench press.  The presence of the jet skis in two different peoples lots perplexed me, because it shows that people living in this area do have a significant amount of disposable income.  Almost all of the houses that had driveways were not paved, but rather gravel drives.  Further down the street there was an apartment complex that seemed a bit out of place in the otherwise single story house area.  The further down the street you go the more undeveloped the area got and there was even a rather sizeable forest at one end that had yet to be touched.


Location 6: Bivins St (entirety)

Bivins Street had a large amount of disparity amongst the houses.  One end had smaller houses that were primarily single storied, but these houses were all located within a short walk of a fairly large park, which was located at one end of Bivins St.  Towards the middle of the street I noticed quite a few modern houses, which is one of the only areas in Durham where I have noticed modern architecture in houses.  This modern architecture makes me think that these houses were built rather recently, and that the owner hired a specialized architect to design the houses.  In addition, at this middle point of Bivins there were also a lot of overcrowded front yards that contained many different personal items, not too different from the items seen at Location 5.  Then at the end of Bivins Street furthest from the park there was a really nice neighborhood that was probably the nicest or second nicest group of houses (other was E. Forest Hills Blvd.).   At this end of the street there was one of the only signs of a new house getting built that I saw on my tour of Durham, and from the looks of it the house that was getting built was going to be quite sizeable.




Location 7: Fayetteville and NC 54 (three blocks in all directions)

To me this was the most interesting place that I visited on the list because it combined so many of the different things that I saw at all of the other locations: a mall, apartment buildings, businesses, shopping centers, houses, and housing developments.  Not to mention that it was the intersection of a freeway and one of the larger roads in Durham.

The shopping centers: In this area we have the nicest mall in Durham, Southpoint mall, whose parking lot is full of cars covering the whole pricing spectrum.  When driving through their parking lot I saw many top of the line car brands (BMW, Lexus, Audi, etc.), but I also saw lots of cars that have clearly been heavily used and were a little beat up.  Across the street from Southpoint there is a very nice shopping center that has some of the typical mainstream eating places (Buffalo Wild Wings and PF Changs).  This shopping center has much nicer shops than the shopping center that shared a parking lot with Northgate Mall.  On the other side of 54 the shopping center is a little less nice, but the buildings are all relatively new and very well kept.  There was also quite a bit of construction going on further down the road, adding even more commercial storefronts.  With Southpoint being so close, and the heavy use of NC 54, I think that this would be the ideal area to open up a new business.

Residential: Depending upon which direction you go you find very different housing options.  There were new apartment buildings that were in the medium range, and there was also a housing development that was a step down from the housing development seen at location 1.  This housing development had cookie-cutter houses that had very small lawns.  The housing development seemed like a cheap version of “Suburbia America.”  Moving in the other direction from the intersection of Fayetteville and NC 54 there are some small run down one-story houses that lay on large properties.  These houses seemed like they could be in a Midwest because of the low elongated architecture that they utilized.  The draw of living here is probably somewhat negated due to the closeness of NC 54 that is only a few hundred yards away from these properties.

Businesses: In this area I saw the only real office building area.  These buildings were designed so that they might easily be mistaken as houses or apartments, but are in fact actual office buildings.  From what I could see from the outside that it looked like most of the offices were occupied by either lawyers or other similar professions.  It seemed like these were high-quality places in a good location, because of all of the traffic that would be attracted by the shopping centers in the immediate area.

Location 8: E. Forest Hills Blvd.

If I had to choose a place to live permanently in Durham, this is the place that I would want to live in.  Not only did all of the houses seem to be placed on top of a little hill, but also they all had nice views of Forest Hills Park.  E. Forest Hills Blvd. is also very nice because the park has multiple tennis courts and lots of walking and running trails.  The nicest thing about this neighborhood was that there were only houses on one side of the street.  In addition to not having houses across the street, the properties that these houses sat on were much larger than the other areas that we visited in Durham.  There was also much care taken to the appearances of both the houses and the lawns of these properties.  In multiple front yards people had added decorative items or statues.  One nice addition that most people had added to these houses were big porches with tables and chairs.


Location 9: NC Central University and surrounding area

NCCU is located directly within a residential neighborhood.  The houses in this area are by no means the nicest houses I saw on my tour of Durham or the worst.  Most of the houses are single story houses that are located on small plots of land, but like most of the rest of Durham there are a few two-story houses sprinkled in.  The school buildings seem to range in terms of date of construction much like Dukes campus.  It is quite obvious that two of the major dorm buildings (or buildings that look like dorms) do not have central heating or cooling, because every window in the building has a window ac unit (this appearance is somewhat similar to project buildings in NYC).  Also as would be expected there were tons of people walking around the school and surrounding area.  In addition, there were more people hanging out porches in this area of Durham than other areas.  All of the buildings on campus are made out of brick, with the professional schools seeming to be the newer buildings because of the color of the bricks and the architecture.

Location 10: Parkwood neighborhood

Parkwood neighborhood had the most communal feel out of any of the areas that I visited around Durham.  It had a baseball field, church shopping center, and school all named after the neighborhood.  One of the most interesting things is that the Parkwood Neighborhood had a fishing pond that had a sign that said fishing for residents of Parkwood Neighborhood only.  There appeared to be some amount of code that all of the houses had to apply to, because all of the houses were extremely well kept even though they were small mainly one story houses.  The code is likely enforced by some sort of neighborhood association, which probably increases the value of the properties.

 Location 11: South St. from Apex to University Dr.

This area might have been the worst area that I visited.  Almost all of the houses had been boarded up by the Self Help organization and had signs that read no trespassing.  In addition, some of the boards used to close the windows had been spray painted to make it look like the houses had not been boarded up.  This might be a trick used by real-estate companies to increase interest in the area  Other than the boarded up houses there were a few vacant lots that only had the cement foundations of where houses used to be.  There were also large amounts of construction that was occurring on South St. that completely shut down the road.


Does Living Near a University Boost Home Prices? Duke and Durham As a Case Study

By Mischa-von-Derek Aikman  Does Living Near a University Boost Home Prices?

The purpose of this paper is to explore the possible existence of a correlation between the proximity of one’s home to a higher education institution (such as Duke University), and the monetary value of that respective home. I have decided to use the residential structures surrounding Duke University’s East Campus as the sample population for this study. More specifically, I analyze and contrast the historical price trends for those homes located within one block of the perimeter of Duke’s East Campus, with homes located two blocks away from the same perimeter. The details of the specific geographic locations of these homes are discussed more thoroughly in the paper’s analysis. I propose that there is increased property value associated with living closer to the physical location of the University relative to living farther away. This difference in value appears to be apparent even in homes that are within one block of each other.


As mentioned above, the homes selected for the study were those located within a 2-block radius of Duke University’s East Campus. The annual historic prices of each of these residential homes between 2004 and 2014 were gathered using Zillow.com’s respective “Zestimate.” The “Zestimate” value is the median Zillow estimate of prices of all the houses in a given geographic location. As of May 2010, the index had tracked over 200 metropolitan areas, and had successfully calculated the index for 120 of these locations. Therefore, the extensive nature of the index made it suitable for the purposes of this study. The homes were divided into two groups; one for those located within a one-block radius of East Campus, and the other for those located within the second block of the East Campus perimeter. The reasoning behind this division was to determine if there is a significant price gap on average between homes located more closely to the University relative to those that are farther away. The specific monetary values for each home can be found in the appendix. The Zestimate at the Durham County level was also collected for this time period to be used as a benchmark. It is important to note that commercial buildings and apartment complexes within the given radius were excluded in an attempt to control for the types of residential structures being observed. The table and annotated map below show which streets the two groups spanned.

Table Showing the Streets each Respective Group Spanned 

Screen Shot 2014-04-03 at 12.44.46 AM

Figure 1: Annotated Map Used for Study

Screen Shot 2014-04-03 at 12.46.50 AM

The homes that fall between the green and blue borders constitute those placed in Group 1 (located closer to Duke), while those that fall between the blue and red borders constitute Group 2 (located farther from Duke). Pricing information was gathered for a total of 485 homes over the 10 years.

Observed Trends and Analysis

The mean home prices for each year was then calculated for both respective groups, and were then plotted against each other along with the Durham County level Zestimate.

Figure 2: Mean Home Prices for Group 1 and Group 2

Screen Shot 2014-04-03 at 12.48.32 AM

It is obvious that there is a significant and consistent price gap between those homes located within one block of Duke’s campus, and those situated two blocks away. It is also very interesting to notice that despite the price gap, both housing groups seem to have been appreciating at more or less the same rate over the past decade. Plotting the price gap itself, as we do in Figure 3 on the next page, shows an unequivocal spike in the price gap between 2007 and 2008. This can be attributed to the culmination, and subsequent burst of the national housing bubble in 2009. Trulia’s chief economist asserted that “geographical home prices was widest in 2007, the peak of the housing bubble.” This may have translated into the unusual widening of the gap at that particular point in time within this subsector of the wider housing market.

Figure 3: Price Gap Between Groups 1 and 2 

Screen Shot 2014-04-03 at 12.49.18 AM

Regarding the significant price gap between the two defined housing groups, we can look to the apparent economic implications of being located within close proximity of a prestigious University. Although extensive literature does not exist on these effects, it is common knowledge that Universities impact communities socially, culturally and economically.


First, is the factor of employment. As of 2006, Duke University was the second largest private employer in the state. A significant portion of the population living within immediate proximity of the University (i.e. within the one block radius) will tend to be well-paid members of faculty and staff within some facet of the University such as health care workers and professors. Hence, these residents are likely to be more financially stable, and equipped to pay more rent than their average counterpart. While it is also probable that some percentage of these employees also live within the two block radius of campus as well, the desired real estate will be that which is more convenient, and therefore, closer to one’s place of employment. This augmented demand within a targeted group of individuals may contribute to the higher prices found within this geographic region.

Investment Incentive

Second, is the very attractive opportunity to invest in real estate near Universities. Zillow’s chief economist, Stan Humphries, asserts that “a lot of students will live off campus, there’s built-in rental demand.”4 The very high flow of students and faculty from year to year lowers the risk investors run by renting homes to tenants near Duke, or any university for that matter. Vacancy rates will be much lower relative to other areas given the continuous demand for housing. Therefore, the heightened demand from the faculty and staff’s perspective, feeds the investor’s growth of demand, who are more confident in the long-term returns on their investment, as well as the short-term security of it. This might be another reason those houses in Group 1 were consistently valued higher than those in Group 2.

Location, Location, Location…

Just as acquiring a beachfront property will typically cost more than the average home, it can be argued that the same is true with purchasing a home close to a University. Housing is an asset that, despite crashes like that of 2008, ultimately appreciates over time. Being located near Duke University, one of three major points in the Research Triangle, intrinsically implies that one is located in an “established” neighborhood. It is very unlikely for the ‘status’ of this neighborhood to decline over time, as the physical University is essentially an immovable asset. This point is further supported if we take a closer look at Figure 2, which plots the Annual Mean House Price for both groups. Although the housing bubble burst in 2008, Durham real estate prices in both groups did not experience a dip in prices until late 2012, leading into 2013. This three-year lag in the reaction of housing prices suggests that residencies located near a powerhouse University such as Duke may be privy to some level of insulation from national market occurrences. This supports the paper’s rationale even further as to why properties in Group 1 would be more desirable, and therefore, more expensive compared to those in Group 2.

Isolating the Outliers

Another interesting observation was that while the house values were cheaper on average in Group 2 than they were in Group 1, there were a few outliers. More specifically, there were occasional strips of Group 1 homes that were far cheaper than quite a number of Group 2 homes. In order to isolate the streets along which these outliers existed, the historical mean price of homes were calculated for each street within each individual group (see appendix for Group 2 data).

Figure 4: Iredell as Outlier for Group 1 Housing 

Screen Shot 2014-04-03 at 12.50.36 AM

For the streets covered within Group 1, the majority of the mean historical prices were range bound between approximately $175,000 and $250,000. The means for Broad St., Minerva St. and Watts St. were all higher relative to the others with a maximum mean of $531,529. The obvious outliers in this case were those houses located along Iredell St, whose historical means floated very consistently around $75,000 throughout the entire decade. Why are these houses valued so much lower than the others found in its group? If one were to look closely at the annotated map in Figure 1, he will notice that Iredell Street was on the furthest most point of the boundary used to confine houses in Group 1. It is possible that the ‘one-block vs. two-block’ measure might have been too neat of a divide, and that the very outskirts of Group 1’s boundary had already transitioned into homes which fit the characteristics of Group 2 more appropriately. However, this outlier still did not cause for the hypothesis to be rejected.


While the results of this study were informative, and supported the original hypothesis, there were a few facets of the experiment that may have limited the level of conclusiveness. First, is the use of the Zillow estimate to gather the historical prices for the homes. It was a suitable index to use within the scope of this experiment since it uses public data on house attributes and actual sales prices to develop its model. However, the academic community often criticizes it for its lack of publically available historical time series.

Second, it is clear that the sample size used in the experiment is relatively small. Having surveyed 4,850 historical prices for 485 homes in Durham provides a nice picture for the community immediately surrounding Duke University. However, there would be great value in expanding the boundaries throughout a larger geographical spread within Durham.

The final limitation is concerned with the method used to define the boundaries that divided the residential homes into two groups. As was seen in the “Isolating the Outliers” portion, the evidence suggests that the border may have been too rigid of a split. Perhaps one could observe more accurate price correlations using the metric distances from the center point of the university and each respective home. This allows the distance factor of the model to be continuous, not discrete, and can speak to even more meaningful relationships.

Summary and Conclusions

Using Duke University and Durham as a case study, we were able to observe significant relationships between historical housing prices for homes located closer to the campus (Group1) relative to those located farther away (Group 2). We noticed that while the houses in both these groups appreciated at rates that were relatively very similar, there was a consistent price gap between houses located within one block of Duke’s campus, and those located two blocks away. More specifically, the homes within the first block were consistently more expensive than those in the second block by significant amounts. Various reasons that could potentially contribute to the existence of this gap were discussed. These included the impact that Duke University has on the employment of those who live near campus, the attributes of homes situated near a university that attract investors, and what seems to be some kind of cushion against larger market phenomena such as the housing crash in 2008. All of these supported the hypothesis that homes located closer the Duke’s East campus, were consistently more expensive than those located farther away over the past 10 years.

Moving forward, it would be very interesting to conduct the same study on a larger scale for numerous universities across the United States. The differences in results between private Universities and State Schools, or between Universities whose campuses are compactly designed (such as Duke University) vs. those that are dispersed throughout a city (such as North Carolina State University) would prove to be very useful within this topic.

Work Cited:

  1. Dougherty, Conor. “Gap Between Most, Least Expensive Housing Market Still Wide.”Real Time Economics RSS. The Wall Street Journal, n.d. Web. 27 Mar. 2014.
  2. Duke and Durham:AnAnalysisofDukeUniversity’sEstimatedTotalAnnualEconomicImpactontheCityandCountyof Durham. Rep. Durham: Office of Public Affairs, 2006-2007. Print.
  3. The Identification and Estimation of A University’s Economic Impacts.G.GeoffreyBoothandJeffreyE.Jarrett.The Journal of Higher Education. Vol. 47, No. 5, pp.565-576
  4. TrackingtheHousingBubbleAcrossMetropolitanAreas–ASpatio-TemporalComparisonofHousePriceIndices.Laurie Schintler and Emilia Istrate. Cityscape. Vol. 13, No. 1, Discovering Homelessness (2011), pp. 165-182)
  5. Woolley, Suzanne. “Real Estate: Investing in College Towns: A Degree in Real Estate”. Bloomberg.com. Bloomberg, 5 Nov. 2012. Web. 20 Mar. 2014.

Appendix: [you may find all the data in Appendix here Does Living Near a University Boost Home Prices? ] 

Screen Shot 2014-04-03 at 12.56.51 AM

Screen Shot 2014-04-03 at 1.07.45 AM


Gentrification’s Effect on Crime Rates

By Mischa-von-Derek Aikman   Gentrification’s Effect on Crime Rates


Many scholars have explored the behavior of crime rates within neighborhoods that are considered to have been completely gentrified, or are still currently undergoing the process of gentrification. They do this largely by studying the changes in crime trends in numerous neighborhoods that display typical characteristics of gentrification. This literature survey pays careful attention to the definition used to select examples of gentrified neighborhoods for examination. It will also exert the claim that crime rates of particular categories seem to rise on average within these neighborhoods upon the commencement of gentrification. It will look at the models used to normalize crime rates across neighborhoods of different populations and densities, as well as those that account for the issue of crime rates regressing towards the mean. Using these normalized statistics, the survey will outline conjectured reasons as to why crime rates seem to rise in gentrified neighborhoods.

Defining and Selecting Gentrifying Neighborhoods

The issue of neighborhood selection proves to be inherently complex as the literature quickly realizes that one of the root difficulties is that what is considered a “gentrified/gentrifying” neighborhood is subject to many different definitions and interpretations. When selecting “gentrifying” neighborhoods for study, it is important to differentiate between neighborhoods that are simply experiencing a cycle of appreciation, and those that are truly gentrifying (Taylor, 1989). In other words the average increase in dollar value of houses and land alone do not define a neighborhood that is gentrifying per se as this can be the result of inflation in the larger housing market (McDonald 1986). Additionally, McDonald (1986) distinguishes between gentrification and “incumbent upgrading” in which current residents improve housing stock, and there is no apparent population change. Rather, Taylor (1989) defines gentrification as “the migration of younger, middle-, and perhaps upper-income households into centrally located urban neighborhoods and the accompanying upgrading of the worn-out housing stock that previously had “filtered down” to lower-income occupants.” It is also commonly accepted that gentrification is accompanied by the inevitable displacement of lower-income residents who previously resided in these neighborhoods (Taylor 1989).

Even with this relatively common definition, methods of choosing neighborhoods for study vary between authors. McDonald (1986) selects a sample of fourteen neighborhoods in which “gentrification has been reported.” These neighborhoods were all located in Boston, New York, San Francisco, Seattle, and Washington, D.C. The literature studied these particular neighborhoods based on various principles. Most importantly, they were chosen due to the availability of time-series crime statistics between 1970 and 1984, as well as an attempt to capture neighborhoods that underwent both commercial and residential gentrification (McDonald, 1986). However, this methodology used by McDonald did not go on to compare gentrifying neighborhoods with non- gentrified neighborhoods, and was generally arbitrary in its selection process (Taylor 1989).

Negative Impacts of Displacement

Given the definition of gentrification used in this survey, displacement proves to be a necessary byproduct. Atkinson (2002) uses cross-sectional data in gentrified neighborhoods where population outflow exceeds citywide averages to determine the extent to which displacement becomes an issue. Atkinson (2002) argues that displacement is typically short lived, but may be prolonged depending on the rate of inflow of new residents. Although using less quantitative methods, Atkinson outlines issues that arise from displacement, which contribute to an environment conducive to increased crime. These include evictions due to the inability to afford the rising price of rent associated with gentrification. This inevitably leads to increased homelessness directly through the loss of Single Room Occupant (SRO) dwellings (Atkinson 2002). However, Atkinson admits that there was no conclusive evidence confirming that the loss of SRO’s were caused by gentrification directly, and not by occurrences in the wider housing market. Atkinson’s research on crime directly produced contradicting results (which we explore in more detail). While crime seemed to fall in some gentrified neighborhoods, others showed that crime actually increased within certain categories (Atkinson 2002). There is also the issue of social conflict sparked by the presence of new residents with “different cultural backgrounds” (Atkinson 2002).

Expectations Surrounding Gentrification’s Effect on Crime

Rational expectations about gentrification’s effect on crime can be made in either direction. We can expect a general decrease in crime due to the fact that statistically, middle to upper income residential spaces typically have lower crime rates (Taylor 1989). Additionally, the more affluent people migrating into the neighborhood are more likely to have more political influence, and can therefore successfully request a heightened police presence (Taylor 1989). We can also reasonably expect increased crime in gentrified areas due to the fact that displaced young adults may move to neighborhoods within close proximity of their original homes, and may view their wealthier replacers as more attractive targets (McDonald 1986). Another practical reason for crime rates to rise is that the presence of richer residents living among those who would typically be below the poverty line could feed an atmosphere of social conflict (McDonald 1986). This occurrence has the potential to manifest itself in physical violence between cohabitants.


As mentioned before, McDonald (1986) utilized the time-series data from 14 arbitrarily chosen gentrified neighborhoods to determine plausible effects of gentrification on crime. While his findings are also discussed in the ‘Results’ portion, this section will look more closely the approach used in Taylor’s (1989) study. Taylor (1989) utilizes census data available for all 277 Baltimore City neighborhoods, and 1979 – 1980 Part I crime data for the same neighborhoods. To obtain the ‘beginning of the decade’ and ‘end of the decade’ crime counts for each offense, 2-year averages were used (1970 and 1971 for beginning, and 1979 and 1980 for the end). Crime counts were then divided by respective total neighborhood population in order to calculate crime rates per 100,000 (except for burglary which was divided by number of households).

In order to capture neighborhood dynamics ‘in the context of what was happening in other neighborhoods,’ both the predictor and outcome scales were made relative. Therefore, both beginning and end of decade crime rates were transformed to weighted percentile scores. With this information, Taylor (1989) was able to rank all the neighborhoods (relative to one another) while accounting for each respective population size. This rank is essentially an ordinal representation of the various crime rates for all the neighborhoods. Taylor (1989) found this measurement attractive, as its skewness (measure of the asymmetry of the probability distribution) is lower than that of logged or raw crime rates.

As opposed to simply taking the difference between scores to determine change in crime rates from year to year, residualized change scores were used (Pt = A + BPt-1 + e). The residual itself (e) should theoretically represent the unexpected change from year to year as it was (as per the assumptions of regression) uncorrelated with the predicted scores. Additionally, each parameter value for 1980 was regressed on its respective 1970 score. The residual generated from this regression was used as the gauge of change (Taylor 1989). Ultimately, these regressions controlled for fluctuating population levels throughout the time period, as well as for each neighborhood’s respective initial level of crime.

Furthermore, in an attempt to provide a more “clear-cut” method of identifying gentrified neighborhoods, Taylor (1989) utilizes a single (not multiple) indicator of gentrification. This measure uses a census-based item whereby households were polled, and asked to provide the current market value of the homes. Using this information, a ‘dynamic index’ representing the appreciation in neighborhood house values was constructed to determine a house-value percentile score for each neighborhood (Taylor 1989). This was done by using a regression model similar to that used in the development of percentile scores for the crime rates. Calculating the percentile changes in house values produced residuals that accounted for the unexpected increases or decreases given the neighborhood’s initial house-value score. Controlling for initial levels accounts for the “regression to the mean” issue, and since the only relevant factor is the ordered ranking of the neighborhoods at any given point in the time-series, inflation is also accounted for (Taylor 1989).

Finally, it was determined that neighborhoods with very high residualized relative house- value scores were those neighborhoods that truly “gentrified” (and not merely experienced appreciation). This obviously presented the issue of determining a “cut off point” (how far down that list would be considered as gentrified neighborhoods?). For this reason, the study was conducted with the top 15, as well as the top 20 scoring neighborhoods (Taylor 1989).


Using the regression analysis outlined in Taylor’s (1989) paper, it was concluded that gentrification was associated with unexpected increases in both larceny and robbery. This result held true both when using the top 20 gentrifying neighborhoods, as well as the top 15.

According to McDonald’s (1986) study on the 14 neighborhoods (See attached table for details on crime rates for each neighborhood), every gentrified neighborhood studied had total Index crime rates above the average of their respective cities. It most be noted, however, that these observations were based on ‘per capita’ crime rates. This poses an issue since the population of almost all these neighborhoods declined during the period of observation. Therefore, these rates could have been just as influenced by population fluctuations as they could be by the actual shift in number of crime incidents (McDonald 1986). More specifically, McDonald (1986) notes that presence of higher crime rates in the gentrified neighborhoods were actually lower for personal crimes, but higher for property crimes (with a few insignificant exceptions). Table 1 (attached in the appendix) indicates that the effect of gentrification on crime is not of a linear nature. The crime rates rise to a significant climax in 1980, and then subside again shortly after (McDonald 1986).

This tells us that the time frame of the observations plays a crucial role in the results one gets. In an attempt to correct for this, McDonald (1986) calculates each neighborhood’s crime rate as a ratio of their respective citywide rate (these values can be seen in Table 2). The results show significant declines in personal crime from what they were in 1970 in 6 of the 14 neighborhoods (McDonald 1986). The analysis of property crime rates showed just the opposite result. Property crime rates for all but one neighborhood showed a decline (McDonald 1986). Finally, it was the general observation that despite the apparent decline in personal crime rates, most of the gentrified neighborhoods maintained crime rates higher than their citywide averages (McDonald 1986).

Questions Moving Forward

Obviously, the results of these two studies produce slightly contradicting results (Atkinson 2002). Whereas McDonald observes a decline in personal crimes and an increase in property crimes, Taylor observes an increase in both. An interesting exercise would be to conduct McDonalds’ methodology of determining trends in crime rates on those neighborhoods selected using Taylor’s (1989) regression analysis. Additionally, studying time-series data for more than a decade could shed light on the results’ sensitivity to the span of time we noticed in McDonald’s piece. Lastly, it would be interesting to explore the existence of any implemented policies used as a response to heightened crime in gentrifying neighborhoods, as well as any influences these policies might have on the crime rates themselves.


Table 1: Table Showing Crime Rates in Selected Cities and Neighborhoods, 1970-84 (McDonald)

Screen Shot 2014-04-01 at 10.33.04 PM

Table 2: Table Showing Crime Rates of Selected Neighborhoods, Indexed to the Crime Rates of Their Cities, 1970- 84 (McDonald)

Screen Shot 2014-04-01 at 10.33.17 PM


  1. Atkinson, Rowland, Dr. “Does Gentrification Help or Harm Urban Neighbourhoods? An Assessment of the Evidence-Base in the Context of the New Urban Agenda.”ESRC Centre for Neighbourhood Research (2002): n. CNR. Web. 06 Feb. 2014.
  2. Covington, Jeanette, and Ralph B. Taylor. “GENTRIFICATION AND CRIME Robbery and Larceny Changes in Appreciating Baltimore Neighborhoods During the 1970s.” Urban Affairs Quarterly 25 (1989): n. pag. Sage Publications, Inc. Web. 06 Feb. 2014.
  3. McDonald, Scott C. “Does Gentrification Affect Crime Rates?” Chicago Journals(1986): n. The University of Chicago Press. Web. 06 Feb. 2014.

The Holdout Problem, Urban Sprawl, and Eminent Domain

By Spencer Rasmussen  The Holdout Problem, Urban Sprawl and Eminent Domain

1.  Introduction

Purpose: To acknowledge the holdout problem, which is a type of land market failure, that contributes to urban sprawl by creating a bias towards the fringes of cities for large land developments

The Holdout Problem: “is a form of monopoly power that potentially arises in the course of land assembly.  Once assembly begins, individual owners, knowing their land is essential to the completion of the project, can hold out for prices in excess of their opportunity costs” or “individual owners, realizing that they can impose substantial costs on the developer, seek prices well in excess of their true reservation prices.”

A holdout problem must require assembly, which is the need for at least two distinct properties for a development.

Result: Large-scale projects that require assembly, like housing developments, parks and open spaces, stadiums or shopping malls, will have high bargaining costs.  This will create incentives for developers to look for land where ownership is less dispersed, which will minimize assembly.  This will lead to these large building projects taking place on the fringes of cities leading to unnecessary urban sprawl

2. The Economic Literature on Urban Spraw

Definitions of Urban Sprawl:

  • Galaster: “Sprawl is a pattern of land use in [an urban area] that exhibits low levels of some combination of eight distinct dimensions: density, continuity, concentration, clustering, centrality, nuclearity, mixed uses, and proximity.”
  •  Nechyba and Walsh: “the tendency toward lower city densities as city footprints expand”
  • Brueckner: “the excessive spatial growth of cities…implying inefficient outward growth”

Sources of market failure that can lead to excessive growth

  • When the price of agricultural land does not fully represent its social value that it produces as open space, which causes its conversion to urban areas
  • When commuters do not properly evaluate the costs of congestion when making commuting decisions, which results in excessively long commute times
  • When real estate developers fail to acknowledge the full social cost of the required infrastructure, which artificially lowers the cost of development
  • Miceli and Sirmans develop a fourth potential market failure, the holdout problem

3. Land Assembly and the Holdout Problem

There is a possible relationship between the holdout problem and urban sprawl suggested by the fragmentation of ownership in urban areas spatial variation in the fragmentation of ownership in urban areas…this is because lot sizes are generally smaller towards the center of cities and become larger as you move away from the heart of the city center.  As a result assembly in the middle of the cities generally requires more participation for a given area then at the fringes of a city.

3.1. A Simple Model of the Holdout Problem

A developer needing to acquire two adjacent, individually owned lots.  We assume that bargaining between the two parties, the developer and land owners takes place in one of two periods: now (t=1) or later (t=2).  The developer can only proceed with development if he (a) acquires both parcels of land in t=1 (b) acquires one parcel of land in t=1 and the other in t=2 or (c) acquires both parcels in t=2.  If the developer is unable to acquire all (two) pieces of land then he is forced to scrap the project.  In this model the lot owners have the option to wither “bargain,” selling their property to the developer, or “hold out,” not selling their property to the developer in t=1, 2.

i.     If the developer is able to acquire both pieces of land in t=1 V > 2v

  1. V = the profit that the developer expects to earn if he is able to acquire both areas of land in t=1
  2. 2v = the combined values of the two properties of land to the individual owners.  So v is the value of one property to one owner.
  3. We know that V must be greater than 2v because the land developer would never assemble the two parcels of land if he could not obtain a profit from the proposed development

ii.     If the developer is able to acquire one piece of land in t=1 and one in t=2 or if the developer is able to acquire both pieces of land in t=2 V – ε > 2v

  1. ε is the cost of the delay that the developer will incur.  We still assume that the development is profitable at this date because otherwise the project would not progress

iii.     Potential Outcomes

If both sellers bargained in t=1

Both sellers will get P~ = V / 2

If one seller bargained in t=1, and received P1, and the other held out in t=1…

  • And then the seller who originally held out in t=1 sells in t=2 for P2

Net return for the project is V – ε – P1 – P2.  P1 = v because as shown below P2 = V – ε – v.  And assuming the developer is able to acquire the second parcel of land the V – ε – P1 – P2 = V – ε – P1 – (V- ε – v).  Here the development is able to take place because the developer obtained both parcels of land, but he does incur a loss because of the delay caused by own of the sellers holding out until t=2

  • And then the seller who originally held out in t=1 holds out in t=2

Net return from the single parcel is v – P1 because the development was scrapped because the developer was not able to acquire both plots of land.

So the net gain from acquiring the second parcel of land in t=2 is V – ε – P1 – P2 – (v – P1) which when set equal to zero yields P2  = V – ε – v.  Obtaining the second parcel of land allows the developer to proceed with his land development and attempt to obtain the profit he expected.

If both sellers held out in t=1

And then sell in t=2 the price per parcel is P* = (V – ε) / 2. This is true because the two sellers will split the payment equally.  In this case the developer is once again able to start his development because he was able to assemble both parcels of land, but he once again has to incur the loss due to the delay of assembly


1. Sellers would prefer to sell jointly in period one as opposed to period two, because the later involves a delay, ε.  This delay causes the overall profit of the development to be decreased.

2. It is better for a seller to be the lone holdout in period two, as opposed to the case where both sellers holdout and sell in period two, as seen by P2 > P*.  This is the same as the classical prisoner’s dilemma problem where both owners are better off selling out promptly, but each individually has the incentive to delay selling.

P2  = V – ε – v and P* = (V – ε) / 2

3. It is unknown if being the lone holdout or selling jointly in period one is better, because while the holdout has superior bargaining power, the available surplus in period two is smaller due to the cost of delay.  The ambiguity comes from the relationship between ε and (V-2v)/2.

4. The worst possible outcome for a seller is to be the lone seller in period one because the price the seller obtains is only v. P1 = v

5. The overriding takeaway form this model is that costly delays can arise in projects involving land assembly.  We can assume that the more parcels of land necessary for a certain development will make assembly even more difficult, and thus lead to more costly delays.  So developers will prefer locations where ownership is less dispersed, all else equal.

  1. Equilibrium strategies (depend upon the relationship between P~, the price the sellers would get if they both bargained in t=1 and P2, the price that the lone holdout in period one would receive if he then sold in period 2)

i.     Suppose P~ > P2  which is true whenever ε > (V-2v)/2

  1. The two Nash equilibriums are (bargain, bargain) and (holdout, holdout) holdout assumes that the seller will then bargain in period two

ii.     Suppose P~ < P2  which is true whenever ε < (V-2v)/2

  1. There is only one Nash equilibrium (holdout, holdout)

3.2. The Spatial Configuration of Lot Sizes and Urban Sprawl

Lot sizes decrease towards the city centers for two reasons: 1) increasing land prices toward the city center cause housing producers to substitute land for capital and 2) increasing housing prices toward the city center also cause the demand for housing to decrease.  Due to these two reasons there is greater population density nearer to the city center.  Which we can reasonably extrapolate from and say that ownership of a piece of land of a given size is more dispersed the closer it is to the city center, meaning that more people own a given area of land the closer the closer this area of land is to the center of the city.


In order to combat urban sprawl…

i.     Developers can maintain their secrecy about projects by utilizing dummy buyers to help acquire assemblies.  This would be useful because sellers would not know that a single buyer is attempting all of the land in a certain area.  This is more difficult for government-backed projects because they often require openness.

ii.     Governments can create incentives or subsidies for building in city centers or disincentives for building in the suburbs.  The justification for this can come from redevelopment of central areas.

iii.     The use of eminent domain, but this often raises issues about whether or not a private organization should be able to benefit from the use of eminent domain.

4. Conclusion

The holdout problem “represents a situation where landowners whose property is essential to the completion of some large development project to seek to block completion of the project in an effort to extract monopoly rents”

This biases development away from areas where ownership is the most dispersed, city centers, and towards areas where ownership is more concentrated, the fringes or suburbs of cities


All quotes are from the following citation

Miceli, Thomas J., and C. F. Sirmans. “The Holdout Problem, Urban Sprawl, and Eminent Domain.” Journal of Housing Economics November 16.3-4 (2007): 309-19. Web. 1 Mar. 2014.


Innovation in cities: Science-based diversity, specialization, and localized competition

By Olivia Nicolaus


This paper by Feldman and Audretsch attempts to address the question of “whether diversity or specialization of economic activity better promotes technological change and subsequent economic growth” (409).  It finds considerable support for diversity as a catalyst for innovation and little support for specialization (409)


A number of scholars including Krugman, Romer, and Lucas support the importance of concentration of people and firms as the most important factor for economic activity (410).  Concentration creates knowledge spillovers, which are the transmission of knowledge “through face-to-face interaction and through frequent contact” (411).  Scholars disagree over the significance of knowledge spillovers within and across disciplines, but it is widely accepted that physical proximity is key for the transmission of “sticky knowledge,” or that which is highly contextual (411).  Knowledge spillovers create increasing returns to scale within a geographically bounded space, primarily the relatively compact area of cities (410).

An important question to ask in relation to agglomeration economies is “does the specific type of economic activity undertaken within any particular geographic region matter?” (410). This opens the debate to two options: a geographic region that specializes in a particular industry, or a geographic region with diverse firms and economic agents.  In order to answer this question, Feldman and Audretsch attempt in this paper to classify the extent of diversity or specialization in geographic regions and then measure “how this composition influence innovation output” (410).

Connecting Innovation and Cities

In this paper Feldman and Audretsch use data from the United States Small Business Administration Data Base as a direct measure of innovative output.  This database is composed of product innovations, each with a four-digit standard industrial code (SIC).  Limitations to this data include the emphasis of product innovations over process innovations, variation in the quality of innovations, and the necessity to treat all innovations as homogenous (414).

By attributing each SIC to a Consolidated Metropolitan Statistical Area or Metropolitan Statistical Area, the researchers are able to rank cities in terms of gross quantity of innovation.  The results are exhibited in Table 1, which shows that the most innovative city in the United States in 1982 was New York.  It is also important to note the overwhelming source of innovation is urban areas, with less than 4% of all observed innovations occurring outside of metropolitan areas (415).  For reference, 70% of the population at this time lived outside of metropolitan areas (415).  The table also uses population statistics to provide a more accurate calculation of innovation, finding San Francisco with the highest innovation rate per capita (415).

Connecting industry clusters, academic departments, and geographical areas

Feldman and Audretsch then attempt to link “products on their closeness in technological space” (415).  To do so they utilize the relevance ranking scale in the Yale Survey of R&D managers to establish groupings between industries that share a common scientific base.  The results are shown in Table 2.

The researchers find that industries that rely on a “common science base” exhibit a tendency to cluster together geographically with regard to the location of employment and innovation (418).  This is the initial information that the researchers use to create a model for determining the effect of a variety of factors on the quantity of innovations in different locations.

Modeling Framework

Feldman and Audretsch establish the dependent variable of their analysis as the number of innovations attributed to a specific SIC industry in a particular city.  They isolate three explanatory variables: a measure of industry specialization, a measure for the presence of science-based related industries, and an index for localized competition (419).  The equation, mean, and standard deviation for these three variables are exhibited in Table 3.


Feldman and Audretsch’s results are shown in Table 4, titled Poisson estimation results for the Poisson regression estimation method.  This method was selected because to model count variable because “the dependent variable is a limited dependent variable with a highly skewed distribution” (420).  This means that the events represented by the data are somewhat rare.  This type of distribution can be used for cancer, cases, number of accidents, or number of bird sightings, but in this case is used for counts of product innovation (Schwartz 1466).

The first column (Model 1) provides results for the three independent variable measures (specialization, science-based related industries, localized competition).  For industry specialization, the negative and statistically significant coefficient suggests that cities that specialize in economic activity in a certain industry have a lower rate of innovative activity.  For science-based related industries, the positive and statistically significant coefficient means that innovative activity is correlated with a strong presence of complementary industries sharing a common science base.  Finally, the negative coefficient on the third variable, localized competition, suggests that innovative activity of an industry is actually associated with less localized competition.  To translate these correlations: the results provide support for diversity in spurring innovation as opposed to specialization spurring innovation (421).

Potential Concerns

There are a few potential drawbacks to using this model.  The first is of city size; that large cities might be expected to have more innovation purely as a result of advantages in total manpower and resources.  There may be a greater degree of economic activity and localized competition.  In the second column of Table 3 (Model 2), total employment is normalized and the results for the third variable change.  This new positive coefficient means that localized competition is, in fact, conducive to innovative activity.  The other two coefficients remain unchanged. Another concern with Model one is the variation in innovation across industries.  In the third and fourth columns (Models 3 and 4), the number of innovations recorded for the specific industry is controlled.  The basic results remain the same.

 Policy Implications and Importance

The answer to the debate of specialization versus diversity prompts two different policy implications.  If innovation is fostered more effectively in specialized economies, policymakers should “focus on developing a narrow set of economic activities within a geographic region” (410).  However, since the opposite is true and diversity prompts innovation, policymakers should attempt to “identify commonalities and foster diversity” within the geographic region (410).

The specialization versus diversity question draws parallels to two types of modern development: university research parks versus the traditional urban form.  According to this study, the diversity of work types that occurs in a traditional urban setting is more innovative and therefore more economically productive than a more focused research park.  If policymakers are purely hoping to pump out a vast quantity of innovation in the form of new products, they should focus primarily on developing a diversity of businesses and corporations in cities, and also figuring out ways to encourage face-to-face contact that is valuable to knowledge spillovers.

However, this research does not measure the quality of innovations, and thus should not be taken at face value.  Further research could incorporate the quality of innovations in the spatial analysis.  In addition, further research could measure how the degree of specialization within research parks affects the amount of innovation created.


Table 1: Counts of innovation normalized by population


Table 2: The Common science bases of industrial clusters


Table 3

Table 4: Poisson estimation results



Feldman, Maryann P., and David B. Audretsch. “Innovation in Cities:.” European Economic Review 43.2 (1999): 409-29. Print.

Schwartz, Carl J. “Poisson Regression.” Poisson Regression. Simon Frazer University, 7 June 2013. Web. 28 Mar. 2014. <http://people.stat.sfu.ca/~cschwarz/Stat-650/Notes/PDFbigbook-JMP/JMP-part025.pdf>.



Which Characteristics of Schools Affect House Prices?

By Gabrielle Ware Which Characteristics of Schools Affect House Prices?

When looking for a new home, buyers consider many factors, including neighborhood appearances and demographics, proximity to city centers, safety, property tax rates, and the quality of local school districts and other public services. Specifically, buyers give considerable attention to public school quality and researchers have become increasingly interested in which school attributes are most valuable to home buyers, and hence, influence house prices most strongly. Traditionally, test scores have been used as a good indicator for school quality and have been associated with higher real estate prices; however, the relationship between school characteristics and local home values has been extremely difficult to quantify. The researchers of each of the following papers found unique methods to overcome various data shortcomings.

Downes and Zabel (2), Kane, Riegg, and Staiger (3), and Clapp, Nanda, and Ross (1) all examined the importance of district performance and demographic composition on property values in Chicago from 1997 until 2001, Mecklenburg County, North Carolina from 1994 until 2001, and Connecticut from 1994 until 2004, respectively. For each paper, the researches compiled data from multiple sources in order to eliminate bias from excluded variables and to quantify the effects of specific school attributes on house prices. Although there are similarities and differences in their methods, all three groups of researchers made conclusions largely consistent with one another.

Data Sources
While examining previous work, Downes and Zabel (2) identified two major shortcomings: the lack of controls for intra-jurisdictional variation in schools and for neighborhood quality when trying to determine which school attributes were relevant to home value. In order to combat these, Downes and Zabel (2) combined data from the Chicago Metropolitan Statistical Area (MSA) America Housing Survey (AHS) from 1897 until 1991, the Summary Tape Files (STF) from the 1980 and 1990 Decennial Censuses, and Illinois School Report Cards from 1987-88 until 1991-92. The AHS data include house characteristics and owner reported home information for a random sampling of homes, which eliminate the bias that arises from only using sales data and neighborhood characteristics that are available in the STF data. Additionally, Downes and Zabel sorted the Census and AHS data by census tracts, nearly homogenous areas of 2,500 to 8,000 occupants, enabling more accurate control over neighborhood variation.

To study the effects of a court-imposed desegregation order to redraw school boundaries in Mecklenburg County, North Carolina, Kane, Riegg, and Staiger (3) faced the same challenge faced by Downes and Zabel (2): distinguishing between the influence of school quality and other contributing factors on house prices. To combat this, they used home sales data from 1994 until 2001 from the county’s Property Assessment and Land Record Management Division. Furthermore, data from the tax assessor’s office was used to divide Mecklenburg County into 1,048 homogenous neighborhoods, which was combined with demographic information from the 1990 and 2000 Decennial Censuses and census tract information. Information about school boundaries throughout the rearrangement was obtained from the Charlotte Mecklenburg School District (CMS) from 1993 until 2001 and information on individual school performances and demographics was provided by the North Carolina Department of Public Instruction from 1997 though 2001.

Finally, Clapp, Nanda, and Ross (1) studied the influence of school district performance and demographic composition on home values in Connecticut, as well as whether or not the importance of these indicators varies over time. They used data from a sample of home sales from 1994 through 2004 purchased from Banker and Trademan, which was combined with census tract information from the 1990 Decennial Census to control for neighborhood variation as in the previous two studies mentioned. This data also included information to control for house characteristics. To account for the possibility that housing price appreciation or depreciation may vary regionally or by market, Clapp, Nanda, and Ross (1) included separate year and month fixed effects for each of the ten Labor Market Areas (LMAs) in Connecticut. Information on school attributes from 1994 until 2004 was provided by Connecticut public schools.

Methods and Results
For their work in Chicago, Downes and Zabel (2) first examined the correlation between school characteristics and neighborhood values. Many variables were significantly correlated suggesting that excluding neighborhood controls would bias the coefficient estimates for school characteristics when determining their influence on house prices. Next, Downes and Zabel (2) used multiple regressions, shown in Table 1, with the natural log of the owner estimated home value as the dependent variable to estimate the importance of six measures of school quality for both the district and school level data.

The six district/school quality measurements used were the proportions of African- American students, Hispanic students, limited English proficiency students, and subsidized lunch eligible students, as well as the natural logs of district per pupil expenditures and eighth grade reading tests. A comparison between the coefficients of the pooled regressions (with neighborhood variables included) estimated using the district level and the school level data revealed many biases that can arise from the use of district level data. First, homeowner’s sensitivity to the racial composition of the local school was hidden, and second, the effects of both the school cost variables were of a lower magnitude when district level data was used. There was no significant difference on the importance of test scores. After establishing that both neighborhood and intra- jurisdictional variations could not be excluded, the pooled regression with neighborhood variables included for the school level data became their favored regression, shown in the second results column of Table 1. From this specification, Downes and Zabel (2) concluded that both the proportions of African-American and Hispanic students in the local school had significantly negative effects on home values, while the proportion of limited English proficiency students and the natural log of district per pupil expenditures had significantly positive effects on home values. This is consistent with the argument that homeowners and researchers measure school quality differently. The first-difference regression results are similar, which shows that proper controls for neighborhood and house characteristics remove the need to control for temporally stable, unobserved house and neighborhood characteristics. And finally, the value-added regression tests the hypothesis that the relationship between house prices and standardized test scores is temporally stable. Although, they were unable to reject this null hypothesis, Downes and Zabel (2) note that the direction of change of the coefficient estimates is consistent with the expectation that as states have been more conscious of making school performances public, the correlation between housing prices and standardized test scores will strengthen.

Unlike Downes and Zabel (2), Kane, Riegg, and Staiger (3) were able to use the redistricting of schools to their advantage while trying to isolate the effects of school quality on home values. Their empirical strategy was two-fold, focusing first, on housing values near school boundaries (houses in the same neighborhood assigned to different schools) and second, on house values for homes affected by the court ordered redistricting. To study housing values near school boundaries, Kane, Riegg, and Staiger (3) ran a series of regressions to track the changes in coefficient estimates for elementary school test scores and distance to the elementary school while increasing controls for housing and neighborhood characteristics around each boundary. The natural log of the sales price was the dependent variable in these regressions. Shown in Table 2, all specifications supported the conclusions that mean test scores have a significant positive correlation to property values; however, this impact decreases in magnitude as more controls are added. The distance of a house from its school assignment was found to have a negative relationship with house price, but this relationship became insignificant as more controls for house and neighborhood fixed effects were included. These same specifications were also used for other measures of school quality, including the proportions proficient on the state test and African-American, the median household income, and the “value-added” test score between 1994 and 1999. The proportion of students scoring at the proficient level on the state test and medium income both had significant positive relationships with house prices, as expected, while the proportion of African-American students in the school had a significant negative relationship with house prices. The coefficients for “value-added” were not significantly different from zero in any of the specifications, implying that prospective buyers observe characteristics of potential peers, instead of “value-added” to measure school quality. This is consistent with the finding of Downes and Zabel (2) that homeowners measure school quality differently than researchers do.

Kane, Riegg, and Staiger (3) also examined the relationship between school characteristics and house prices using solely differences in redistricting. As shown in Table 3, all specifications included controls for housing and neighborhood characteristics, as well as fixed effects for every reassignment. The three measures of school quality included were the percent of African-American students, the median household income, and the percent of proficient students on the state test. The percent of African-American students had a significant negative impact on house prices while the average median income and percent proficient had positive effects on house prices at the high school level. At the middle school and elementary school levels, the measured effects were either insignificant or only marginally significant. Both empirical strategies Kane, Riegg, and Staiger (3) confirmed the presence of residential sorting as additional indirect impacts resulted because the population living in any given school boundary is itself a function of the school assignments.

Similarly, Clapp, Nanda, and Ross (1) used three specifications to observe the effects of school attributes using the same dependent variable as Kane, Riegg, and Staiger (3), the natural log of the transaction price. Shown in Table 4, the first results column uses the traditional hedonic regression without controlling for town or census tract effects, the second column presents the regression after controlling for town fixed effects, and the third column after controlling for census tract fixed effects. School attributes included in their study were Math test scores and the fractions of free lunch eligible, non-English speaking, African-American, and Hispanic students. Like the researchers who conducted studies in Chicago and Mecklenburg County, Clapp, Nanda, and Ross (1) found that the effects of school district attributes were sensitive to which specification was used.

The most basic OLS model, controlling only for neighborhood observables, overestimated the effect of test scores on housing prices, and gave coefficient estimates for the effects of the fractions of non-English speaking, African-American, and Hispanic students that were inconsistent with previous findings. As seen before, this regression overestimated the effect of school quality of housing prices. In the second and third columns, the coefficient estimates are not statistically different from one another; however, Clapp, Nanda, and Ross (1) favor the regression that includes census tract fixed effects. This model implies that test scores have a significant positive effect of home values while greater fractions of African-American and Hispanic students have a significant negative effect on home values. These effects are of similar size as the ones measured by Downes and Zabel (2) and Kane, Riegg, and Staiger (3). Clapp, Nanda, and Ross (1) took their study one step further to explore whether or not the effects of key district attributes have changed over time. They found that the effects of the fraction of Hispanic students and test scores are changing over time, becoming less negative and more positive, respectively. This is the type of change that Downes and Zabel (2) suggested, but were unable to confirm statistically.

The three studies surveyed in this review yield many similar conclusions regarding which school attributes influence house prices. First, homebuyers are concerned about changes in the demographic makeup of the local school, in addition to, and sometimes more than, test scores when deciding how much to spend on a home. Second, two sources agreed that prospective homebuyers do not measure school quality in the same value-added way that researchers might. Instead, they rely on many observable factors, such as demographics of their peer groups, to measure school quality. And third, it would be interesting to see if the way homebuyers measure school quality will change as school performances on standardized tests are made more public. One source was able to confirm that this is the case, while yet another suggested it without providing statistical verification. All of this leads to the question of whether homebuyers actually cared more about certain observable demographic factors or if they just used them as indicators because they were more easily accessed. The decline in the use of certain demographic factors as key indicators may also suggest changing opinions regarding race that would be interesting to explore.


  1. 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 Economics63.2 (2008): 451-466. http://digitalcommons.uconn.edu/cgi/viewcontent.cgi?article=1094&context=econ_wpape rs
  2. Downes,ThomasA.,andJeffreyE.Zabel.”Theimpactofschoolcharacteristicsonhouse prices: Chicago 1987–1991.” Journal of Urban Economics 52.1 (2002): 1-25. http://theunbrokenwindow.com/Research%20Methods/Hedonic_school.pdf
  3. Kane, Thomas J., Stephanie K. Riegg, and Douglas O. Staiger. “School quality, neighborhoods, and housing prices.” American Law and Economics Review 8.2 (2006): 183-212. http://www .dartmouth.edu/~dstaiger/Papers/KaneRieggStaiger%20NBERwp11347.pdf


Table 1: Downes and Thomas (2) – Regressions Using School level Data and the Natural Log of Owner Estimated House Value as the dependent Variable
Note: controls for variations house and neighborhood and neighborhoods characteristics were included in this regression but are not pictured below

Screen Shot 2014-03-30 at 3.34.17 PM


Table 2: Kane, Riegg, and Staiger (3) – Sensitivity of Regression Estimates to Neighborhood and Housing Characteristic Controls

Screen Shot 2014-03-30 at 3.35.26 PM


Table 3: Kane, Riegg, and Staiger (3) – Sensitivity of Housing Prices to Changes of within Neighborhood School Characteristics

Screen Shot 2014-03-30 at 3.36.05 PM


Table 4: Clapp, Nanda, and Ross (1) – Regressions using District Level Measure of School Quality and Controlling for Different Fixed Effects

Screen Shot 2014-03-30 at 3.36.36 PM

A Game-Theoretic Analysis of Skyscrapers

By David Lillington  A Game-Theoretic Analysis of Skyscrapers

Skyscrapers have received little attention from urban economists in the past according to Helsley and Strange (2008). What has been discussed relates to their place in the standard urban model, otherwise known as the monocentric city model. In this model, skyscrapers are attributed to the phenomenon of increasing land prices as one approaches the city center. Due to these higher prices, buildings are built up in order to save land costs. In their study, Helsley and Strange (2008) argue that higher land prices are not the only reason for the stratospheric height of these manmade marvels. To builders, the height (and relative height) of their building carries importance for issues of publicity and pride.

In order to capture this importance of building the highest building in any market, the study uses game theory to simulate a skyscraper building contest and then continues on to explain an equation that models overbuilding. Building profits are still a determinant in the builder’s payoff function (as they are in the monocentric city model); however in this particular case whether or not the builder has succeeded in constructing the tallest building becomes part of the function as well. They describe their model as an “all-pay auction” (Helsley and Strange, 2008). That is, builders spend their resources and he or she who bids the highest takes the prize. This prize does come at a cost; its value is partially diminished through the bad economics of skyscrapers. The article proposes two situations: simultaneous and sequential construction. In the simultaneous game theoretic model, no contestant gains any value from competing in the game except for the builder who constructs the highest skyscraper. He or she will enjoy the value of this prize; however dissipation occurs for reasons that will be discussed. In the sequential model, the cost comes before construction “where the leader builds a tall-enough building to deter competitors” (Helsley and Strange, 2008). They use the story of the Empire State Building as evidence for this model.

Introduction of Variables

The model first introduces a situation in which two risk-neutral builders exist, i = 1, 2. Both builders own land on which to build, however we assume that builder 1 possesses a better location.
This causes , where is defined as profit maximizing building height. Because value is given to height in this situation, the model adds some exogenous variable >0,to include building height. This is multiplied by the indicator variable, , which is given a value of 1 if the builder succeeds in building the tallest building in his or her market or 0 if he or she does not succeed. This gives the equation:

where, is builder i’s profit and is building height. Building height can be expressed as equation:

The variable represents what Helsley and Strange (2008) call a “pre-emption”; that is, “if a rival builder j chose height , builder i would concede the contest because it would never be in the builder’s interest to choose a height that would win”. In other words, when .  This expression also explains the prize dissipation experienced in a skyscraper contest. The contest payout will be less than the profit-maximized payout as shown above. In the following two games, is assumed to be less than . This is because there would be no competition in the market otherwise.

The Sequential Game Theoretic Model

 In the sequential model, is chosen sequentially. The model proposes that builder 1 goes first with the strategy to choose a height such that so that builder 2 will surrender to building at his profit maximization height . His or her pre-emption will not surpass builder 1’s actual building height. If builder 1 chooses a height such that , the equation still holds because builder 2 will continue to build at his or her profit-maximizing height since this will win the competition and be economically rational. If then builder 2 will just top builder 1’s height in order to win the competition. Therefore, Helsley and Strange (2008) propose it is better for builder 1 to win the contest by surpassing to achieve . In the sequential game theoretic model, a significant cost is the choice to build high enough that no other competitor will choose to build higher. This can lead to overbuilding, which will be discussed later.

The Simultaneous Game Theoretic Model and Overbuilding
The simultaneous model considers a game in which two builders choose building height simultaneously. Helsley and Strange (2008) propose that in this case, a positive probability weight on is placed on building height . This represents the probability that builder i will build at a height less than or equal to h. The payoff for this situation can be expressed as , the profit maximizing situation. The model proposes that builder i’s payoff is equal to . This expression is claimed to represent the probability that a builder wins choosing a height multiplied by the value of the prize in addition to the value of the building. This gives the general equation
for which j=1,2 and h_i is an element of h* and . This can be rearranged to be rearranged to be

Helsley and Strange (2008) then differentiate with respect to h_i , resulting in .
The expected building height can be calculated by integrating by parts the derivation of this equation. The equation is set up as .
Helsley and Strange (2008) claim that integration by parts yields
, knowing . Through this equation, it is easy to see that overbuilding occurs. Expected building height, E[h], is greater than profit maximizing building height h* (the expression in brackets remains positive as π(h) is decreasing on the interval [h*, h_P ] according to Helsley and Strange (2008)).


Helsley and Strange (2008) present new work in the study of these urban marvels. They come to the conclusion that skyscrapers are not economical when they are built in a contest to reach the highest altitude. Their research includes plenty of historical data, citing stories of the construction of many of the world’s tallest skyscrapers. Some of them, such as the Burj Dubai, fulfill the proposed overbuilding prophecy. It has been hard for them to find enough tenants for their space. The game theoretic model alerts builders of the poor economics of building for height. The authors stress that this can impact the real estate market and its cycles, contributing to “increases in vacancies and declines in rents, leading to subsequent slowdowns” (2008). In fact, they cite that, when built, the Empire State Building, the Manhattan Company Building, and the Chrysler Building brought a whopping 4,000,000 extra square feet of commercial space to New York City (20% of the city’s stock) (2008). Perhaps further research might be done to measure the impact of the construction of a skyscraper on a city’s real estate market. How does it change prices? Also how does pricing within a skyscraper itself change?


Robert Helsley and William Strange, 2008, “A game-theoretic analysis of skyscrapers,” Journal of Urban Economics 64: 49-64.