As reported in the Austin Monitor, Mayor Pro Tem Delia Garza offered the following assessment of the local housing market during a recent discussion about Austin’s land development code rewrite:
“…the city’s population is growing faster than new housing supply, driving up both demand and costs. With demand steadily increasing at around 150 people per day, she said, the way to see prices go down is to build new housing fast enough to meet that demand.”
Measuring housing supply and demand is extremely complicated, and it’s becoming increasingly contentious in fast-growing cities grappling with escalating costs, gentrification, and displacement. Politics can explain a portion of the controversy, but limitations of publicly available data—margin of error, time lags, geographic specificity, to name just a few—are far more important to understand as we strive for a “common set of facts” to guide community discussion and policy making.
The underlying challenge for analysts is the dynamic nature of housing markets. Changes in population and housing are constantly fluctuating nominally and spatially. People are born, they die and they move. New homes are permitted, built, and occupied. And thousands of people from around the country view real estate listings online, thereby influencing the perceived value and various local market conditions in rapidly growing cities like Austin. Publicly available data simply cannot keep up.
So until some clever technologist figures out how to accurately track those dynamic changes—likely through means I don’t want to know about—we will continue to be frustrated by our inability to accurately measure true supply and demand in local housing markets. But for the purpose of informing public debate and policy making it’s important to figure out what we can know with a reasonable degree of confidence, what we can’t know given currently accessible data, and what is probably impossible to know.
First, let’s deal with the Jason Voorhees of local sound bites about population growth that, according to Garza, could be impacting housing demand—is it “steadily increasing at around 150 people per day?” Here’s what we know from publicly available data: The U.S. Census Bureau estimates that Travis County’s population has grown by an average of 16,750 per year, or 46 per day, since 2010 due to net migration, or people moving in minus people moving out. For the metropolitan area, it’s nearly 39,000 per year, or 106 per day. Unfortunately neither estimate addresses net migration to the city of Austin, specifically. Nor do we know how to group those net new residents to Austin into households in order to properly estimate the impact on demand—one net new resident, of course, is not necessarily equivalent to one net new housing unit demanded. People live in families, and children do not purchase or lease homes.
So where does that leave us in terms of Garza’s claim about the 150 people per day? We do know from another census estimate that total population growth in the city of Austin has averaged nearly 20,000 per year since 2010, or about 54 per day, which is helpful context—and, notably, many fewer than 150 or even 100 people per day—but not very useful for estimating demand for housing units, primarily because the Census Bureau uses housing unit data to derive that figure from county-level population estimates.
But there’s another, more fundamental problem here, and it has very little to do with data issues. While it’s important to understand what data can and cannot tells us, it’s more important to understand what users of that information—residents, advocates, policy makers—are really asking or want to know. Garza’s point, I think, was less about aggregate demand, in which case 150 per day might be the relevant statistic even if it is overstating the matter, and more about demand in a certain segment of the market. To put it in my own words, is supply increasing fast enough to stabilize prices for the most price-sensitive people in the Austin housing market? That’s a different, and potentially better, question that can tell us something meaningful about affordability and equity in Austin.
Of course, it’s also much more difficult to answer convincingly, and not just because of the standard data issues outlined above. For example, existing publicly available data is silent on one of the most important puzzles to sort through when it comes to affordability: How many people want to live in the city of Austin but for whatever reason choose to live somewhere else, whether that’s elsewhere in Travis County, in an adjacent county or someplace else entirely? To be clear, displacement is occurring in Austin due to the rising cost of housing; that is an undeniable fact. But, it’s tempting for some participants in this debate to characterize development and population growth in suburban areas outside Austin city limits – and even those who oped to locate in nearby cities within the region – as would-be Austin residents who were “priced out” and, consequently, had to settle for another location. That’s an overstatement, at least for people lacking primary data on why exactly those residents opted to live outside Austin.
Consumer preferences present another dilemma. There are not many clear substitutes for housing, so you can’t treat it like a typical good. Despite what some participants in the debate might contend, it’s not merely a factor of “supply and demand.” Preferences are important. For example, let’s assume a three-person family looks at two homes of equivalent cost – a two-bedroom on a small lot in Austin, and a four-bedroom on a larger lot in Pflugerville. Say they choose the home in Pflugerville. Is that evidence of an affordability problem in Austin? Has that family been “priced out” of the Austin market? Perhaps our hypothetical family also considered a three-bedroom rental home in Austin they could afford, but they preferred to own a home, and that was the deciding factor in favor of Pflugerville. How do we account for ownership preferences when trying to estimate the number of housing units needed based on population growth, especially when the stated policy goal for many people is “housing?”
The opportunities for more research in this area are immense. Academic researchers in Austin are already making progress using surveys and interviews to sort through issues that can’t be addressed satisfactorily with existing, publicly available data. We hope to contribute more to our body of knowledge on this topic in the months to come.