The Affordable Housing Problem and Investor Behavior in Pittsburgh

By Nick Kharas, Claire Jacquillat, Erin Yanacek, and Maksim Khaitovich

Our cross-functional team from Heinz College and Tepper is interested in looking at the affordable housing scene in Pittsburgh. This was part of a case competition we won, jointly organized by SUDS and the Data Analytics clubs at Heinz and Tepper. Eventually, the topic grew on us as we shared similar experiences while searching for a house to rent or buy. Currently, Pittsburgh has an affordable housing deficit of 17,000 units. Affordability is a reflection of the price of the house, its condition, and the livability of the surrounding area. Thus, solutions to Pittsburgh’s housing challenges need to focus on healthy community development. Keeping that in mind, we propose an “Assess and Address” framework that identifies investor behavior and creates a system of incentives and penalties to address negative influences.

Hazelwood – Our Pilot

Like many other neighborhoods in Pittsburgh, Hazelwood has been hit hard since the city’s steel mills shut down. Still, it has maintained its tight community spirit. New developments like technology company investments, startups and malls are reviving communities in neighborhoods like East Liberty and Lawrenceville, but at the risk of gentrification. Hazelwood is representative of what has been happening across Pittsburgh. Recent developments like Almono and Summerset at Frick Park have the potential to revive the neighborhood. At the same time, we must also ensure that these new developments bring positive change and not displacement for longtime residents.

1

Hazelwood’s community focus is still intact, but the recent developments could attract bad investors. As we see from the comparison of average property prices, Hazelwood does not have cases where properties in poor or unsound condition are sold at inflated values, unlike the rest of Allegheny County. While this is a good sign for Hazelwood, it still has some vacant properties and many houses in average or fair condition, which could potentially attract attention from bad investors in the future, if they aren’t already there.

2

Note: The graph highlights outliers indicating that some properties in poor or average condition are sold for significantly inflated values. The edges of each box in the plot indicate the interquartile range of sale values for each property condition. The flat line within each box is the median. The dotted lines are the outliers. For our assessment, we considered all valid sales from the property assessments data maintained by the WPRDC. Also, we used zip code 15207 to distinguish Hazelwood from the rest of Allegheny County. Although 15207 covers parts of Glen Hazel and Greenfield, it represents the challenges we aim to address.

3

Investor Behavior – Who Are These “Bad” Investors?

A neighborhood stands to benefit if houses in poor or unlivable conditions are purchased and redeveloped. However, some investors are only interested in buying and reselling houses to make a quick profit, without improving their condition or spending on maintenance. Unlike “rehabbers”, “flippers” and “milkers” buy and keep properties in distressed conditions, and hope to sell them off for a profit as quickly as possible. Such behavior does not attract the healthy investors, and also negatively affects the community and quality of life in the neighborhood.

4

We decided to look at data driven ways to identify and predict bad behavior using the property assessments data maintained by the WPRDC. Although we do not have any information on house owners, our cognitive solution identifies and flags potentially bad investments and highlights insightful characteristics.

The data does not have class labels that identify bad investments. We ran a k-means clustering algorithm on ownership duration, property value appreciation, and sale price to set our class labels for good and bad investments needed for analytical modeling. We avoided using general definitions for flippers to determine class boundaries as that could ignore any hidden patterns and add bias. For example, any house resold within 12 months for less than $100,000 is potentially a transaction conducted by a flipper. However, using this information alone to set class boundaries ignores several transactions where the property was held for only a little more than a year. The results from clustering directed us to target parcels that were

  • Owned for less than three years, and,
  • The property value depreciated, or appreciated less than 15%, or was sold for less than $90,000.

Any parcels that met the above condition were flagged as properties owned by potentially bad investors, while the rest were labeled as unsuspicious.

5

6Once we were able to set the labels for good and bad investments, we ran few classification algorithms to predict investor behavior. We excluded 20% of the data for testing, and trained the remaining 80% data on different classifiers like Gradient Boosting, Random Forest, and Conditional Inference Trees (a modeling technique based on unbiased recursive partitioning). All returned an AUC between 0.70-0.72. Gradient Boosting and Random Forest with feature selection returned marginal improvements, but not significant.

 

Our primary challenge was to not let our results get affected by any bias in the data. A majority of the records are not classified as bad investments, and around 90% of the parcels are not red flagged. This makes sense, as we cannot expect the property market to be completely overrun by flippers. Additionally, the assessments data has missing values for some fields, including those which determine our class labels. For example, some parcels do not have previous sale records, making it difficult to determine how long the house was held before it was sold. We cannot assume that this is missing data, as the property may not have changed ownership more than once in its lifetime. Further, the assessments data set gives us the parcel characteristics as of now, and not as when the property was last sold. For example, the condition of a property may have either improved or deteriorated after it was last sold four years ago, but we have no way to find out. For this reason, the simpler models did not perform well. Logistic regression proved to be computationally expensive with a large number of categorical variables, and decision trees performed poorly on test data.

7

We have shared our detailed analysis on GitHub.

 

Our Proposal – Assess and Address

A city like Pittsburgh would like to identify and avoid bad investor activity. However, in an effort to maintain housing affordability, the city cannot drive away potentially good investments that can develop and enrich its vibrant community. To maintain this balance, we propose an Assess and Address framework that gives actionable recommendations.

8

Assess

Confirm Results with In-Person Observation

Our analytical model can highlight and flag properties that are potentially at a risk of being owned by flippers. The city inspectors and community leaders can export a list of such at-risk targets, look up information on their owners or landlords, and monitor their behavior.

Address

Sanction Bad Practices

Install practices and systems that would discourage landlords from mistreating their tenants and violating requirements for minimum standards. Our suggestions include:

  • Minimum Property Standards established by the Allegheny County Health Department for rental properties.
  • Rental Registration – Force landlords to act responsibly towards tenants. A good example is the Probationary Rental Occupancy Permit (PROP) set by the city of Raleigh, NC, which aims to ensure better housing quality for tenants and discourages landlords to violate City Codes.

Foster Positive Practices

These are programs that would encourage good investors and community homeowners to invest in enriching and developing the city’s community spirit. Some suggestions and examples of their implementation include:

 

We hope to take this initiative forward with the help of SUDS at Carnegie Mellon University and present our findings to the city council at Pittsburgh.

 

Nick Kharas graduated from Carnegie Mellon University with a Masters degree concentrating in Data Analytics and Business Intelligence. Prior to his time at CMU, he was a business intelligence and data warehousing SME at a Japanese multinational financial holding company. When not a data buff, he enjoys travel, sport and meeting new people. Click here to check out his work on GitHub. You can also connect with Nick at https://www.linkedin.com/in/nickkharas.

Claire Jacquillat is an MBA candidate at Carnegie Mellon’s Tepper School of Business. She focuses her studies on Operations management and Operations research. She is the president of Tepper Data Analytics Club where she strive to foster an integrated use of business analytics in various industries. Before starting her MBA at the Tepper School, she worked as a strategist in Sales Enablement for a Fortune 500 company. You can connect with Claire at https://www.linkedin.com/in/clairejacquillat/en

Erin Yanacek is an MBA candidate at Carnegie Mellon’s Tepper School of Business. In summer 2017, Erin will join McKinsey as a Summer Associate. Prior to business school, Erin was a classical musician. She founded a non profit organization, the Chamber Orchestra of Pittsburgh, and toured internationally performing and teaching classical trumpet. You can connect with Erin at https://goo.gl/BEWz1L

Maksim Khaitovich is an MBA candidate at Carnegie Mellon’s Tepper School of Business. In summer 2017, Maksim will join A. T. Kearney as Summer Data Science Associate. Prior to business school Maksim worked as an engineer and IT consultant in fintech and wireless communications. You can connect with Maksim at https://www.linkedin.com/in/maksim-khaitovich-828a2b47/

The Overlapping Police Departments of Allegheny County

By Lizzie Silver and Lauren Renaud

banner

This article was written as part of SUDS’ partnership with the Alliance for Police Accountability.

“Did you know that there are over one hundred police departments in Allegheny County alone?” Prof David Harris offered this factoid at the panel discussion following the April 12 screening of Peace Officer. It sat at the back of our minds until finals were over and we started looking for a new mapping project. Spoiler alert: we were ‘only’ able to count 85 departments. [Update: we’re up to 127 departments. Details here.]

Recently, two adverse interactions between the police and the public involved officers from at least two departments: the Port Authority (PAT) Police and the City of Pittsburgh Police. The first incident involved fighting between youths and police officers at the T station downtown. In the second incident, PAT Police Officer Brian O’Malley shot and killed Bruce Kelly Jr, raising questions about PAT police policies. As Tony Norman writes, “Why the disparity in response between the officers? Why did only two of [at least eight officers] fire if Mr. Kelley was so dangerous?”

Police departments work together to solve problems when their jurisdictions overlap. Officers from different departments must coordinate with each other in high-pressure, high-stakes situations. Different departments sometimes have different training or different policies, so the way they interact with the public can vary. After the Peace Officer screening, Pittsburgh Chief of Police Cameron McLay mentioned that training on implicit bias will be available to City of Pittsburgh officers in the near future. However, that training will not be available to Port Authority officers, and officers from many other departments that have jurisdiction within the City of Pittsburgh.

Furthermore, the laws that govern Pittsburgh only apply within city limits. Recently Pittsburgh decriminalized possession of less than 30 grams of marijuana. So if you are walking down S Braddock Ave between Biddle and Whitney, smoking a blunt, the local police can only fine you $100. But walk past Overton St, and suddenly you’re committing a criminal offense that could cost you $5000, land you in prison for 30 days, and give you a criminal record, because Swissvale is outside city limits.

Even if the law is the same in different areas, the “unwritten rules” can differ. As Ryan Deto writes for the Pittsburgh City Paper:

“Mount Lebanon police cited Esquivel-Hernandez on March 26 for driving without a valid license and without insurance. He paid his fine on April 21, and according to his U.S. District Court case, he was identified as undocumented on April 25. Mount Lebanon police have not returned multiple calls requesting comment about its communication policy with ICE.
[…]
Since 2014, Pittsburgh Police have had an unwritten policy not to initiate communication with ICE about undocumented immigrants. The department will comply if ICE initiates contact. However, there is no indication that other smaller police forces in Allegheny County have adopted similar policies.”

Different police departments are also held accountable by different bodies. For example, the Citizens’ Police Review Board (CPRB) can investigate complaints against the the City of Pittsburgh police department, but the Port Authority Police do not answer to the CPRB, only to the District Attorney. DA Stephen Zappala exonerated Officers O’Malley and Rivotti for shooting Mr Bruce Kelly Jr, a questionable but unsurprising result, given how rarely police officers face charges for shooting citizens.

[Update: see also our post about the problems of small police departments.]

We thought it’d be helpful for the public to know where the different police departments’ jurisdictions are located. This may help if:

  1. You are wondering which police departments you are likely to interact with in a given area;
  2. You need to contact the police, and are wondering which department to call; or
  3. You have had an interaction with police officers, but you are not sure which department they were from.

We welcome feedback and additional information. We hope this map helps inform citizens about the different departments that serve them, and helps to start a conversation about standardizing training and policies for officers from different departments.

The Map

[Note: This updated version has more police departments included and a better color scheme. Our original, less complete version can still be accessed here.]

http://carnegiemellon.maps.arcgis.com/apps/View/index.html?appid=813f1c964eef4bf59fbbfc18b1ed418e&extent=-80.3644,40.2634,-79.6448,40.6323

  • Larger version of this map here
  • To see the number of overlapping jurisdictions in any given area, see this map with semi-transparent overlays here.
  • To see both, with the “spyglass” feature for more detail, see this map here.

[Old versions available here, here and here.]

The Departments

Municipal Police

There are 130 municipalities within Allegheny County, of which 70 have their own police department. We matched the county’s list of police departments to a list of municipality borders from the county’s GIS department.

One municipal police department could not be matched by name: the Northern Region Police Dept in Gibsonia, PA. It may be a joint department covering multiple municipalities.

We’ve assumed that the other municipal police departments only cover the areas they are named after. However, some municipalities have joint agreements so that a single department will patrol two municipalities. For example, North Versailles police currently patrol Wilmerding (although from 2017 onward that duty will fall to Allegheny County police). We are not aware of most of these agreements, so we can’t put them on the map. If you have more information about these agreements, please email us at our contacts below.

[Update: Thanks to feedback from JI Swiderski, we have corrected this layer of the map. Allegheny County has 130 municipalities, of which 109 have their own police department, 4 share the Northern Region joint police department, 15 contract their police services out to a nearby municipality, and 2 contract out to state police. Allegheny County’s list of police departments includes information on these contracts. We matched that list to the list of municipality borders from the county’s GIS department, so the updated map shows which PD has jurisdiction in every municipality. More details we’ve updated here.]

County Police

The Allegheny County Police Department’s jurisdiction includes ‘county-owned property’, as well as ‘Pittsburgh International Airport, the County Airport, nine County Parks and other regional parks’. We weren’t sure which ‘other regional parks’ were included, so we only mapped out the county parks. We have marked all county buildings with a spot that extends 250 feet from the center of the building. We got the building locations and the park shapefiles from from the county’s GIS website.

The Allegheny County Police might also patrol municipalities that don’t have their own police departments, but we weren’t sure which ones (if any), so we have not included any on the map.

The Allegheny County Sheriff’s Office also has jurisdiction in Allegheny County, but we weren’t sure what areas it extends to, or whether they do any patrolling, so we have not included them on the map.

Federal and State Police

The federal US Marshalls (Western District PA) has jurisdiction throughout Allegheny County, as do the Pennsylvania State Police

[Update: The PA Fish and Boat Commission polices the rivers.]

Transit Police

The Port Authority (PAT) Police have jurisdiction over PAT routes and nearby areas. We have marked the PAT routes with a line that extends 1/10 of a mile (528 feet) around the route. We think this is a conservative estimate of how far the PAT jurisdiction extends. However, if we widened the buffer much further — say, to 1000 feet — the PAT jurisdiction looked like it blanketed the entire city of Pittsburgh. We got the shapefiles from the City’s GIS website.

Amtrak police have jurisdiction over Amtrak routes and nearby areas. We have marked the Amtrak routes with a line that extends 1/10 of a mile (528 feet) around the route. We got the shapefiles from ArcGIS.

[Update: Thanks to Mx Daria Phoebe for pointing out that Norfolk Southern rail police are also accredited.]

School Police

Most of the universities in Pittsburgh have their own police. Their jurisdictions extend across the college campuses, to student dorms, and nearby areas. For example, Carnegie Mellon University Police’s “primary patrol zone… includes all campus property, Off Campus Housing & Sites, and residential areas immediately in the vicinity of the CMU Main Campus.” We do not have shapefiles for the college campuses, so instead of tracing the shape of each campus, we’ve just put big circles on the map (centered on each campus, with a radius of half a mile). There is one circle, half mile radius each for:

We left the Community College of Allegheny County (CCAC) off the map. Each of their campuses has “a Director of Safety and Security who is also a sworn police officer”, but we felt that one officer per campus was too few to count as a “department” for the purposes of this map.

Hospital Police

UPMC employs police. St Clair Hospital Police “will issue citations to those who violate parking and traffic regulations”. Neither hospital has a webpage describing their police departments; they may simply employ police officers alongside security officers. However their police have arrest powers and are employed by the hospital so we put them on the map. We assume that their jurisdictions cover the hospital complexes. We have put circles on the map for each UPMC building complex and for St Clair Hospital. We weren’t sure how large to make the circles, so we put circles with radius of both 250 feet and 500 feet around the center of the buildings; you can turn one overlay off if you prefer.

Humane Society Police

Humane Society Police Officers enforce animal cruelty laws. They are appointed per county so their jurisdiction covers the whole of Allegheny County. There are a number of Humane Society Police Officers serving Allegheny County.

Assumptions and Open Questions

We’ve tried to be conservative in our estimates of where the departments’ jurisdictions extend to, but we may have made some mistakes. We also have some open questions, such as:

  • What kind of authority, if any, do Block Watch Groups have?
  • Do some correctional facilities have their own police? Do corrections officers have arrest powers in PA?
  • Does the Allegheny County Police Department only have jurisdiction over buildings that house county departments and services, or all county-owned buildings?
  • Should the Allegheny County Sheriff’s Office count as another department? If so, what is its jurisdiction?
  • Do UPMC Police cover all UPMC hospitals, or only certain ones? What sort of authority do they have?
  • Which police department(s) have jurisdiction over the municipalities that do not have their own police force? We believe it may be the County Police or that some departments that are listed as a specific municipality may also cover their neighboring towns, but we need more clarification on this.

This map may not be complete. If you know about another police department within Allegheny County, please send us a link to the police department’s webpage, or let us know who we can contact to confirm the department’s existence. Also, if we have made a mistake, please let us know so we can correct it. You can contact us at silver@cmu.edu and lrenaud@andrew.cmu.edu.

Lauren Renaud is a masters student in Data Analytics and Public Policy at CMU.
Lizzie Silver is a PhD student in Logic, Computation and Methodology and a masters student in Machine Learning at CMU.
In their spare time they participate in Students for Urban Data Systems at CMU, and do research for the Alliance for Police Accountability.