Housing

Housing

Sources for Housing Data

• US Census

• Technically the census is “Census of Population and Housing”
• Added to the 1940 population count as a result of depression era attempts to assess the degree of inadequate housing and how new construction might improve the economy
• American Housing Survey

• Compiles data on housing size and quality, neighborhood characteristics, home financing, and recently moved households
• Conducted biennially between May and September in odd-numbered years

• Building Permits Survey
• Home Ownership and Vacancy Rates

• S&P Case-Shiller Index

• Analyzes changes in prices of recorded homes over particular time frame
• Controls from many variables since the same home is being analyzed across time

• Federal Housing Finance Agency (FHFA)
• Bureau of Labor Statistics (BLS) Shelter Index
• National Association of Realtors

The Housing Bubble

• Shiller Index revealed high volatility in bubble prices”

• 240% increase from 1997-2006; 120% decrease from 2006-2009
• FHFA was much less volatile

• Shiller includes sub-prime financed units; FHFA does not
• Shiller is a more comprehensive measurement

• Both limit data to metropolitan areas and larger

• Predicting the bubble was challenging

• Housing prices change due to fundamental and speculative factors:

• Fundamentals (less volatile): rental value, inflation, vacancies, demographics, etc.
• Speculative (highly volatile): buy low and sell high for a quick profit

• Researcher Dean Baker predicted the bubble by comparing fundamental to actual prices, identifying a divergence around 2002

Homeownership Rates

• Rates increased to an all time high of 69% in 2006 and racial gaps had shrunk significantly
• Calculated as: (owner-occupied households)/(owner and renter occupied households)
• Rates can increase due to:

• Renters becoming owners
• Renters consolidate (move back home, take in roommates, etc.)

• Important: When the numerator and denominator are simultaneously changing, quick conclusions should not be made

Mortgage Interest Deduction

• Largest federal subsidy for owner-occupied housing
• Touted as a “middle class tax break”

• More likely to buy homes
• More likely to itemize taxes
• Higher Tax bracket, greater benefit (up to $1,000,000 cap)

• Data suggests otherwise

Quality of Housing

• Changes in housing prices may reflect quality changes
• Case-Shiller and FHFA control for many variables by looking at the same home over time (lot size, sq.ft, layout, structure) and make adjustments upon each new sale
• Downward skew in prices often emerges due to short-sales and foreclosures

Geographical Units

• “City”, “County”, “Rural Area” are often subjective and arbitrary
• Important for detailed geographic issues and consistency across time
• OMB attempts to locate geographies that are linked economically and socially

• Metropolitan Statistical Area: over 360 in the US
• Micropolitan Statistical Area: 560 in the US

• Census defines

• “Urban” as any incorporated place with more than 50,000 residents and “Built Up” characteristics
• Census Blocks (11.5 million in U.S)
• Census Tracts (65,000 in U.S.)

Racial Discrimination

• 1975 Home Mortgage Disclosure Act (HMDA) requires banks disclose lending practices in census tracts
• Early data suggested “Redlining” but did not control for other pertinent characteristics (e.g. wealth, income, FICO)
• In 1990, the Boston FED controlled for other factors and found race remained a factor
• FED Report has been contested

• Omitted variables; however, they controlled for over 60
• Similar default rates in minority and non-minority neighborhoods

• If standard is higher for minorities than default rates should be lower for those neighborhoods

• More recent studies have demonstrated that applicants identical in all respects except race receive less information and are quoted higher rates if Hispanic or Black

• Discrimination may not carry through to application denial

Segregation

• Typically measured by census track demographic data, obscuring neighborhood segregation
• Dissimilarity Index

• The proportion of a group that would need to move in order to achieve perfect integration
• 1970 to 2010 index suggests less segregation
• May be due to movements of Asians and Hispanics rather than Blacks

Homeless

• Estimates suggest anywhere from 600,000 to 3 million homeless in the U.S.
• Lower estimates: point-in-time head counts

• Records people in shelters, transitional housing, and on the street
• HUD reports 643,067 homeless people on one night in January 2009
• Fails to consider length of homelessness

• Overestimates chronic homelessness as some individuals are only temporarily homeless
• Underestimate the number of people that have homeless at some time in their life

• Larger estimates: one year estimates

• HUD reports 1.56 million people spent at least one night in a shelter from 2009-2010

• Underestimate; does not include homeless not in shelters

• Highest estimates: extrapolation

• (Point-in-times)/(Population in poverty)
• National Law Center on Homelessness and Poverty and the Urban Institute generate about a 3.5 million figure based on a 1996 study
• Fails to consider that the proportion of those in poverty that are homeless may change over time

Best Places to Live

• Different studies use different variables (Climate, crime, housing, culture, education, income, wealth, public transportation, etc.)
• Different studies use different weights despite using same variables
• Hedonic Pricing: analyzing price differences to impute a value for a qualitative variable

• How much more would the same house sell for in San Diego vs. El Centro

• Limitations: Impositions to mobility (job access and transportation)

Affordability

• Qualifying Income
• Proportion able to afford a median priced home