People

PUMS

All other charts and tables in this section of the report are derived from the Census Public Use Microdata Sample (PUMS) which is much more granular and adaptable than typical Census American Community Survey (ACS) tables. 

The PUMS dataset is organized geographically into Public Use Microdata Areas (PUMAs) which typically conform to county geographies but often do not neatly align with municipal boundaries. This analysis utilizes the PUMAs that most closely align with the municipality's boundaries, including all municipal area but also some adjacent communities in certain cases. 

Public Use Microdata Areas (PUMAs)

The following PUMAs were combined to compose the dataset reflected in this section:

• Texas 02303 (map)

• Texas 02305 (map)

• Texas 02306 (map)

Analyzed ACS tables

• B01001 Sex by age

• B01003 Total population

• B11016 Household type by household size

• B19001 Household income in the past 12 months

Place

The Place section includes two subsections that draw from a range of datasets to derive observations and findings.

Current housing mix

Analyzed ACS tables

• B25004 Vacancy status

• B25014 Tenure by occupants per room

• B25024 Units in structure

• B25032 Tenure by units in structure

• B25041 Bedrooms

• B25042 Tenure by bedrooms

• B25047 Lacking complete plumbing facilities

• B25051 Lacking complete kitchen facilities

• B25063 Gross rent

• B25068 Bedrooms by gross rent

• B25075 Value (owner)

• B25094 Selected monthly owner costs

• B25118 Tenure by household income in the past 12 months

Cost of housing

This subsection incorporates data from the Census as well as economics resources including the Federal Reserve Bank of St. Louis (FRED) and Zillow.

• Median home price affordability: Pulling data on pricing and incomes from FRED (Federal Reserve Bank of St. Louis Economic Data), this analysis compares the median price of a home with what a household earning the local median household income could afford without incurring cost burden. Price data is pulled directly from FRED and includes monthly estimates as current as one or two months prior to publishing. Gauging affordability requires some computation, translating median incomes into the home purchase price that income could afford based on the typical loan term, interest rate, down payment, private mortgage insurance (PMI) cost, and property tax rate at the time. Values for these factors are found from a variety of sources that vary based on the region of the country being studied.

• Home sales prices: Deriving real-time data from Zillow's Home Value Index (ZHVI), this metric tracks the average sale prices over time across three terciles (i.e. the average of the lowest 33% of prices, the average of the middle 33%, and the average of the highest 33%).

• Average asking rent: Pulling from Zillow's Observed Rent Index (ZORI), this metric tracks average asking rent prices over time. This number should be understood as the asking rent for an apartment that is newly on the market and being advertised to new tenants (rather than what people currently in leases are paying, some of whom may have signed their lease long ago for rents that no longer reflect the latest market trends).

• Current owner and renter costs: These charts visualize PUMS data to illustrate the distributions of monthly costs currently borne by the local population for owner and rental housing respectively. These are distinct from sales prices and asking rents in that they include the majority of residents who have not moved recently and therefore are paying mortgages and lease rents that were set in prior years when market conditions may have been different. These are provided as comparative context for the ZHVI and ZORI measures to indicate how affordable newly offered homes and apartments might be to current residents.

Demand

How many units are needed?

The 10-year housing production target translates projected household growth plus the adjustment factors listed below into the number of housing units needed to maintain the existing housing supply and keep up with anticipated new demand.

Regardless of growth prospects or latent demand pressures, every local market should maintain sustainable vacancy rates and offer hospitable housing stock to best serve community residents. Some housing production is often necessary to keep each of these indicators in a healthy range. 

• Household growth: Forecasted from 2023 to 2033.

• Overcrowding adjustment: Comare the locate rate versuses the national average. Overcrowding is measured by >1 occupant/room. Often related to vacancy rate, the degree to which supply limitations drive households to occupy under-sized units.

• Replacement housing: 0.05% of the housing stock is replaced annually, which includes uninhabitable or obsolete units requiring replacement.

• Vacancy adjustment: Compare the local rate of 5.7% to the healthy market minimum of 5% for ownership and rental combined. Vacancy is the “slack” in the housing market (too low and prices can spike, too high and neighborhoods can suffer blight)

• Substandard adjustment: Compare the local rate to the national average. Substandard housing is measured by incomplete plumbing or kitchen. It is the portion of units that are functionally inadequate.

Analyzed ACS tables

• B25106 Tenure by housing costs as a percentage of household income in the past 12 months

• B25074 Household income by gross rent as a percentage of household income in the past 12 months

• B25095 Household income by selected monthly owner costs as a percentage of household income in the past 12 months

What is the right mix?

The preference model combines several factors, each corresponding to a parameter of housing need.

• Absolute scale of need: The number of units required is from the 10-year demand model as derived above.

• Future income distribution: Multiple sources, including trends-extended, regional movers, and cost-burdened households, all expressed within AMI brackets.

• Ability to pay for housing: This model caps each household's ability to pay for housing at 30% of income. This is intended to not reinforce the construction of housing that will exacerbate cost burden.

• Willingness to pay for housing: This model uses regional PUMS data to identify what households are willing to pay for newly constructed units. This is intended to take into consideration the fact that many households spend below 30% of their income on housing, especially at higher income levels, and therefore the projection of demand needs to account for these preferences.

• Number of bedrooms: This model uses regional PUMS to identify the number of bedrooms households move into at each income level and each tenure type (owner or renter). Lower income households are generally smaller, have fewer earners, and tend to prefer smaller units, but of course there is a range of preference, and this model accommodates that.

• Rental or owner: This figure is based upon the regional norm. This model and report does not put a lot of emphasis on tenure for several reasons. First, it is not something that communities have control over, as tenure cannot be part of regulating new housing unit construction. Second, units can change tenure from owner to rental and back again, so what is more important is the cost and size of the unit. However, the ratio of rental to owner can be useful to monitor, so the final scenario in the report uses the regional PUMS mix of rental and owner at each price point and unit size to recommend a distribution of units.