The rising income of first-time homebuyers

Reposted with permission from BuildZoom.

During the housing boom, mortgage lending was available to people with a broad range of incomes. Since the bust, first-time buyers have grown more financially select.

During the last decade’s housing boom, it became much easier to obtain a mortgage, and many people who previously couldn’t buy a home grasped the opportunity. In 2005 the annual number of first time home buyers in the U.S. hit an all time high of 3.2 million, but by 2008 it fell below 2 million and remains below that level to this day. The relative absence of first-time buyers from the current housing recovery should worry us because it can dampen the housing recovery, but also because of its social implications.

The important thing to understand is that fluctuations in the number of first-time buyers go hand-in-hand with changes in their nature: when buying a home gets more challenging it is the least financially able who drop out of the race first, and vice versa.

Does the persistently low number of first-time buyers mean that homeownership is becoming the privilege of a more select few?

Turning to the data

To see how first-time buyers’ financial ability evolved, we turned to the data. The Annual Social and Economics Supplement of the Current Population Survey (CPS ASEC), a joint project of the U.S. Census Bureau and the Bureau of Labor Statistics, provides the most recent national snapshot of household incomes. Although it doesn’t identify first-time buyers per se, it identifies owner-occupancy and asks people who moved over the past year why they moved. Two of the available responses single out the group of first-time buyers that concern us: “wanted to own a home, not rent” and “to establish own household.” We used respondents who gave one of these reasons as a proxy for first-time buyers.

The graph below shows the average household income of first-time buyers since 1999, when the question “why did you move?” was first asked. It references first-time buyers’ income against that of all other households, i.e. the general population, and reports both annual household incomes and a five-year moving average, whose smoothing helps discern the long-run picture.

While income in the general population remained fairly stable during the housing boom, first-time buyers’ income dropped significantly. The ease of borrowing during the boom allowed households that couldn’t previously afford a home to buy one, and the inclusion of such households among first-time buyers reduced the group’s average income.

Matters have changed since the housing boom ended. Although average household income fell during the Great Recession and has yet to fully recover, first-time buyers’ income substantially increased. It spiked in 2008 when lending dried up at the peak of the financial crisis and then subsided in 2009 and 2010 when the First-Time Homebuyer Credit was in place, but overall it has risen, and is now back to its level circa 2000.

A skeptic might argue that first-time buyers’ income is rising faster than the general population’s just because of widening wage inequality, i.e. because first-time buyers are drawn from rungs of the socioeconomic ladder in which wages have risen faster than at the very bottom. This is not the case.

As the next graph shows, the average first time buyer at the peak of the housing boom was drawn from the 56th percentile of the income distribution, whereas now she is drawn from the 59th. In other words, during the boom first-time buyers were able to step into homeownership from lower rungs of the socioeconomic ladder.

Why has first-time buyers’ income risen since the boom?

The precipitous decline in first-time buyers’ income – most evident in the dashed annual line – was stemmed from 2004 to 2006 when interest rates were raised from 1 percent in June 2004 to 5.25 percent in July 2006. During that time the overall volume of transactions in the housing market hit peak and began falling sharply, foreshadowing the subsequent bust. Sales to first-time buyers dropped even more sharply than the overall volume, because first-time buyers rely more heavily on mortgages than other buyers and are therefore particularly sensitive to interest rates.

By 2007, when the housing bubble had burst and the financial crisis erupted, lenders responded by severely restricting credit. They tightened lending standards by requiring larger down payments, lower debt-to-income ratios and better credit scores, which was enough to preclude many from buying a home. The restriction of credit disproportionately affected first-time buyers, who rely most heavily on mortgages, and among them it had the greatest effect on households whose financials were the weakest. When homeownership slipped out of reach for these households their earnings stopped counting towards first-time buyers’ average income, so the average increased.

More recently, tight lending standards have begun easing and responsibility for keeping first-time buyers out of the market has begun shifting to the recovery itself, especially in the most heated local markets. As the recovery gains steam, rising home prices inflate the required down payment and at the same time rents that are rising faster than incomes make it harder to save towards one. The result is the same: as the least financially-able households are excluded from buying, first-time buyers’ average income rises.

How else are first-time buyers changing?

The nature of first-time buyers has fluctuated over the years in other respects, as well. According to the same CPS data, the share of married households among first-time buyers fell during the boom and — although it has fluctuated — the trend since has been upwards. In contrast, the share of married households in the general population declined smoothly, as illustrated in the following graph.
The married share is relevant because it correlates with the presence of multiple incomes in the household, but the CPS data allow us to look at households with multiple incomes directly.

The share of multiple-income households is generally much higher among first-time buyers than in the general population, but while the share steadily declined in the general population over the entire period, for first-time buyers it shows the same V-shaped pattern as income and marriage. A related pattern also emerges with respect to age. Whereas the average age (of heads of household) in the general population steadily increased over the period, for first-time buyers it dipped during the boom and has since risen. The following table reports the figures.

A more comprehensive but technical set of results, including difference-in-differences estimates and the associated standard errors, is available here.

The upshot

Lending practices during the housing boom opened a brief window of opportunity for people who couldn’t previously afford it to become homeowners. That window has closed, and the ability to transition into homeownership is gradually becoming the privilege of a narrower group of first-time buyers that is more financially select.

This is a situation in which the glass is half full and half empty. Given the role of irresponsible lending practices in bringing about the housing boom and bust, it is a good sign that the current cohort of first-time buyers is more financially robust. At the same time, the notion that homeownership is slipping out of reach for a growing share of the population is an uncomfortable one, especially if the trend continues. Do we really want homeownership to become a privilege rather than a choice?

Methodology

We assembled data from the 1999-2015 CPS ASEC supplements (each supplement covers the period from April of the preceding year through March of the current year, i.e. the 1999 CPS ASEC supplement covers April 1998 through March 1999). As described in the main text, we proxied for first-time buyers using owner-occupants who had moved in the past year, and who answered the question “why did you move?” with the response “wanted to own a home, not rent” or “to establish own household.” This question first appeared in 1999. We conducted the analysis at the household level, weighting the sample to represent the national population. Household incomes were adjusted for inflation using the CPI-U, and converted to constant 2015 dollars.

Issi Romem is the chief economist at BuildZoom.

Email Issi Romem.