National Multifamily Housing Council
It’s common knowledge among apartment professionals that rents are rising as a result of increased demand for rental housing and record low levels of new construction the past two years. The downside of this otherwise beneficial supply/demand situation is a growing number of media stories about escalating rents adding to the burdens of already struggling households.
If past experience is a guide, it probably won‘t be long before the tone of these reports shifts to imply that profit-seeking owners are taking advantage of unfortunate renters, who are vulnerable because they can’t afford to buy, and there aren’t enough rental residences.
But are high rents really the problem? There are many reasons to think not. This issue of Research Notes suggests that the culprit for cost burdens faced by renters is not rents, which in many markets are still below their pre-recession levels, but rather income, which has been stagnant since before the latest downturn.
Rent Trends: The Federal Data Perspective
The first step in analyzing this issue is to review rent trends over an intermediate- to long-term time frame. Unfortunately, this is not an easy task. Average rents can be calculated from annual and biannual surveys like the American Housing Survey and the American Community Survey, but these are not intended for time series analysis and have drawbacks in analyzing trends.
That leaves the consumer price index (CPI), which includes an estimate of “rent of primary residence,” as the only federal data source. As with all CPI data, there are no estimates of actual price (rent) levels, just the percentage change in prices (rents). Also as with other CPI data, the rent series is a “same-store” estimate.
There are limitations with this data series, however. First, it does not distinguish between single-family, small multifamily and apartment rentals. Second, there is no quality breakdown (class A, B, C); all are combined in one rent estimate. Third, the method used to collect the data causes estimates of rent inflation to lag actual rent changes. Notwithstanding these drawbacks, it is still worth examining because it has the longest history (monthly seasonally adjusted data go back to 1981, and monthly not seasonally adjusted data go back to 1947 and less frequently back to 1913) and because it is a widely recognized, publicly available series.
The following chart shows CPI rent inflation from 1990 to the first quarter of 2011. The lagged nature of the series is clear: the peak gain occurred in 2001 Q4, about a year later than most apartment rent series peaked. The lag also mutes the highs and lows. That peak increase, according to this measure, was only 4.7 percent, well below the nearly 10 percent rent rise indicated by other estimates. Also, the recent low was zero percent when all apartment data series showed actual rent drops as a result of the Great Recession.
The lack of “surging rent growth” is even clearer when rents are adjusted for overall inflation. While nominal (unadjusted) rents have increased an average of 3.0 percent annually from 1990 to 2011, inflation-adjusted real rents are up just 0.3 percent annually. The story is roughly the same over longer periods. For example, from 1970-2011, nominal rents increased 4.2 percent annually, but real rents actually fell slightly (-0.1 percent annually). In short, these data suggest that over time, rents haven’t diverged much from the overall rate of inflation.
Apartment Rents: The Private Data Provider Perspective
Apartment rent data from private providers tells a more dynamic story. Keeping to same-store data, MPF Research provided a special analysis aimed at eliminating any effect of their increasing sample size. Here are their results:
MPF Same-Store Rent Recovery
Peak to Trough
(2008 Q2 to 2009 Q4)
Trough to Today
(2009 Q4 to 2011 Q1)
Peak to Today
(2008 Q2 to 2011 Q1)
Source: MPF Research.
While there is considerable regional variation, nationally investment grade apartment rents today remain 2.6% below the previous peak level (mid-2008). And unlike the for-sale housing market, rents never underwent a bubble in the first place. Since the previous 2001 recession, the biggest annual pickup in same-store rent nationally was only 4.4 percent (back in 2006 Q3).
To look at the data another way, NMHC constructed an index going back to 1994 using the MPF Research same-store quarterly rent changes. We note upfront that there are some drawbacks to this approach. In particular, although we call it “same-store” analysis, the set of “same stores” changes over time so that the 2011 universe is quite different from the 1994, or 2000, universe due to new construction, demolition/destruction and increased share of apartments responding. However, this approach is broadly similar to the “chain” indexes used in calculating GDP, hence is deemed a worthwhile addition to the analyst’s toolbox.
The chart below shows this rent index in both nominal and real terms. Here again, the average rent nationally in 2011 Q1 was 2.6 percent below the previous peak. But we also calculate the average annual rent increases over longer periods of time:
The first period represents a strong rebound from the real estate debacle of the late 1980s-early 1990s, hence we would expect stronger-than-average rent increases. Even so, the average annual increase in apartment rents over the whole period is muted at best.The first period represents a strong rebound from the real estate debacle of the late 1980s-early 1990s, hence we would expect stronger-than-average rent increases. Even so, the average annual increase in apartment rents over the whole period is muted at best.
The inflation-adjusted data are even more striking. While apartment rents rose a little faster than overall inflation in the 1990s, they have actually trailed inflation since peaking in the first quarter of 2001. Although real rents are up a slight 1.1 percent total since their trough, they remain 13.7 percent below the 2001 Q1 peak. Again it is clear that rents can’t be the real problem facing renters.
For good measure, we examine one other rent measure using yet another source (Reis) and looking at overall market rents rather than same-store rents. This means that new apartments, with higher rents, are included in the mix, pushing the average higher over time. While those higher rents generally represent higher quality, a case can be made for including them because they represent part of the array of rents available to current renters.
Here’s what these data show: from peak to trough, apartment rents fell by 2.4 percent. Since then, they’ve rebounded, but remain 0.4 percent below the peak. No sign of high rents there. And again, adjusting for inflation makes this even clearer: real rents in the first quarter of 2011 were at the low point for the current cycle, 4.2 percent below the peak.
The upshot is this: thus far, there is no evidence that rising rents are responsible for any housing cost burdens that renters might be currently shouldering. Might that change in the near future, given the widely anticipated rent increases to come? A significant pickup in rents is indeed likely to materialize. This is simply the inevitable outcome of changes in both demand and supply – both the result, in different ways, of the bursting of the housing bubble.
Renting has become more popular as people realize that homeownership (with high leverage) is not a “can’t miss” investment. Even for those who want to buy, tighter underwriting is making it difficult. This new demand is being delivered into a market that recorded the lowest level of new multifamily starts in 50 years because of a virtual shutdown of private construction lending during the financial meltdown. But that too will change. Capital markets are improving, and barring a further constriction, new apartment construction is likely to ramp back up to needed levels. This will moderate future rent increases.
The real problem many renters (and homeowners) face—indeed, the elephant in the room for some time now—is on the income side. A future issue of Research Notes will examine that part of the problem.
Questions or comments on Research Notes should be directed to Mark Obrinsky, NMHC’s Senior Vice President of Research and Chief Economist, at email@example.com or 202/974-2329.