Issue Brief: Volume 2, Number 3

The Rental Affordability Crisis

Author: Professor Todd Sinai

Housing costs are rising because the incomes of households that could live in a city are growing faster than the incomes of households that already live there.


  • Undeniably, renting has become very expensive in many parts of the U.S.—part of a general, long-term erosion in housing affordability overall.
  • In recent decades, this lack of affordability has been creeping up the income distribution, affecting households in the middle-income brackets as well as those in the lowest-income categories.
  • One commonly-proposed policy solution to declining rent affordability is the construction and preservation of low-income housing.  But this will only ameliorate the situation temporarily.
  • To avoid the risk of housing costs outpacing one’s income, one has to be a homeowner in the first place.
  • A subsidy—economists tend to favor a capped refundable tax credit for first-time home buyers—can encourage home ownership, but will not miraculously make the most appealing cities affordable for low- and middle-income households.
  • To become a home owner, people may need to accept that they cannot afford to live in currently high-cost cities, and are better off buying homes in up-and-coming cities that are presently cheaper.

It’s a tough time to be a renter.  According to data from the U.S. Census, half of all renters, and 83 percent of renters with incomes under $20,000, paid more than 30 percent of their incomes in rent in 2011.  These facts recently spurred U.S. Secretary of Housing and Urban Development Shaun Donovan to say, “We are in the midst of the worst rental affordability crisis that this country has ever known.” One commonly proposed solution? Increased development of apartments targeted at low-income households.

But the unaffordability of rental apartments is hardly new, rising rents are far from pervasive, and the notion of “rental affordability” glosses over the fundamental issue our country faces: Not everyone can afford to live wherever they want.

There is no easy solution.  Simply building more apartments is not a panacea, and a policy of subsidizing housing costs so that rents are comparable across cities would be very expensive.  Instead, the focus should return to home owning.  Although encouraging home ownership will not miraculously make expensive cities cheap, home owning insulates households against unexpected jumps in housing costs and rent growth exceeding their income growth over the long term.  

Is there a rental affordability crisis?

One fact is undeniable: Renting has become very expensive in many parts of the U.S.  Between household formation prompted by the improving economy, and the difficulty or reluctance households face in purchasing homes, demand for rental apartments has grown and apartment construction has not kept pace.  Between 2010 and 2013, the national apartment vacancy rate fell by half, from 8.0 percent to 4.3 percent, according to data from REIS.  In some cities, formerly owner-occupied housing units have been converted to rentals, but the scale and geographic distribution has been insufficient.  In Phoenix, for instance, the apartment vacancy rate has declined from slightly more than 12 percent in 2010 to just above 5 percent in 2013 – now well below its 9 percent average vacancy rate over the 1980-2012 period.

However, the bursts of rent growth due to this demand/supply imbalance are more the norm than the exception.  Indeed, the recent rent growth is not unusually high relative to history and is even more concentrated in a handful of cities than past rent booms.  As an example, Figure 1 graphs the average annual growth in asking apartment rents after accounting for inflation for a number of metropolitan areas.  Each vertical bar corresponds to a single city.  The data come from REIS, which surveys the landlords of the highest quality tiers of apartments in a number of major cities.

Figure 1: Average Annual Growth of Real Ask Rent, After Accounting for Inflation

During the 1990 to 1999 period, plotted in panel A, most cities experienced rent growth in excess of inflation.  About a half-dozen were well above that – between 3.5 and 4.5 percent per year.  That is a lot of rent growth – over a 10-year period, 4 percent growth corresponds to a nearly 50 percent increase in real rents.  Between 2000 and 2007, plotted in panel B, rent growth in some cities continued to exceed inflation, although not at the same rate as during the prior decade.

By contrast, during the most recent half-decade of 2008 to 2013, rent growth in most cities did not keep up with inflation.  This result is plotted in panel C.   In only about a dozen cities did rent growth outpace inflation – and even then, just barely.

So, why the sudden alarm about rent affordability?  First, rents have been growing rapidly over the last couple of years.  Between 2008 and 2011, in the midst of the Great Recession, rent growth trailed inflation in every city that REIS tracks (panel D).  However, between 2011 and 2013, rent growth exceeded inflation everywhere – and in some cities, by as much as 3 percent per year (panel E).  This recent rent growth undid the prior rent declines in many cities, and combined with the rent growth between 1990 and 2007 has led to high housing costs. 

Second, income growth has failed to keep pace with rental growth over the last decade. At the national level, between 2000 and 2011, growth in REIS rent exceeded the growth in median renter income by one percentage point each year on average, mainly because real median income declined over that period. In marquee cities, the disparity between rent and income growth has been greater. In New York, for example, real rents grew by 7 percent whereas real renter income fell by 8 percent, a 15 percent differential.  Falling real income exacerbates rent growth to yield a greater decline in affordability than rent growth alone. 

However, today’s headlines about sky-high rents are merely the current manifestation of a long-run trend.  By the usual definition of affordability – do housing costs exceed 30 percent of income – most major cities have long been unaffordable for the lowest income households.  In the most expensive cities, such as New York and San Francisco, rents exceeded 30 percent of income for most of the lowest-income households many years ago.  Even for less-expensive cities, over the last 30 years rent-to-income ratios in excess of 30 percent have become pervasive.

In recent decades, this lack of affordability has been creeping up the income distribution as, in many metropolitan areas, growth in housing demand has exceeded growth in supply.  Indeed, in many cities, the shares of households in middle-income brackets paying more than 30 percent of their incomes in rent have been rising to levels comparable to the lowest-income categories.  Again, the most expensive cities were the first to experience declining rates of housing affordability among the middle class.  But other cities are exhibiting the same patterns, just a little later. 

These trends can be seen in Figure 2, which plots for a sample of U.S. cities the fraction of households over time in each of five different income categories who reported to the U.S. Census that they spent more than 30 percent of their incomes on rent.  Most notably, in many of the example cities in Figure 2, the lowest-income households have not experienced much of a decline in rental affordability lately – but only because rents in those cities already exceeded 30 percent of incomes for most of the low-income population.  Once the vast majority of the lowest-income households are paying more than 30 percent of their incomes in rent, there just is no room for the affordability rate to fall further.  (This is a limitation of the 30-percent-of-income definition of affordability.  Harvard’s Joint Center for Housing Studies notes that rents have increasingly exceeded 50 percent of low-income households’ earnings.[1]

Figure 2:  Households that spent more than 30 percent of their incomes on rent, 1980-2012

In all of the charts, the topmost blue line represents the lowest income category, those households making less than $20,000 per year in real (2010) dollars.  In most cities in Figure 2, rents have exceeded 30 percent of income for more than two-thirds of the lowest income households since at least 1980.  The fraction of such households paying more than 30 percent of their incomes in rent does increase between 1980 and 2012, but the increment often is small.  In Denver, for example, nearly 80 percent of such households paid more than 30 percent of their incomes in rent in 1980, and that fraction grew to 90 percent by 2012.  Similarly, little change is evident in New York, Philadelphia, Phoenix, and San Francisco.  In all of these cities, large fractions of the lowest income groups paid more than 30 percent of their incomes in rent as far back as 1980 and those shares are only slightly higher now.  By contrast, in Atlanta and Cleveland, the ratio increased from around 60 percent in 1980 to around 90 percent in 2012.

Instead, the greatest decline in affordability has occurred amongst low-to-middle income households.  In Atlanta, the share of households with incomes between $20,000 and $35,000 in year-2010 dollars who paid at least 30 percent of their incomes in rent rose from about 20 percent in 1980 to more than 80 percent in 2012.  In Boston, New York, and San Francisco, the affordability rate amongst households in that income category declined to the same rate as for households making less than $20,000. 

Even households in the $35,000 to $50,000 real income tier have experienced declining affordability rates, albeit not to the same degree.  For example, in Dallas in 1980, less than 10 percent of households in that income bracket spent more than 30 percent of their incomes on rent.  By 2012, the unaffordability rate had risen to more than 25 percent.

However, the highest income groups in the data – households making $50,000 or more in real terms –have experienced little decrease in affordability.  The share of such households paying more than 30 percent of their incomes in rent rose slightly in some cities – such as Boston, Chicago, New York, and San Francisco.  Nonetheless, on balance rents still are low enough in most cities that few high-income households spend more than 30 percent of their incomes on rental housing.  

Despite the recent focus on renters, the trend of rising housing costs is also evident in owner-occupied housing.  Many cities are exhibiting decreasing housing affordability, period.  It doesn’t matter whether the houses are owned or rented; in those cities, households of all stripes pay increasing shares of their incomes for housing.  Indeed, growth in rents has closely tracked growth in house prices for a very long time.  Figure 3 plots the average growth rate in real house prices between 1950 and 2000 against the average growth rate in real rents over the same time period for a host of U.S. metropolitan areas.[2] 

The thin line corresponds to the 45-degree line, where rent growth and house price growth is exactly equal.  The thick line is the fitted line between actual rent and house price growth.  It lies slightly above the 45-degree line – house price growth is about one-third of a percent higher on average than rent growth – but is almost exactly parallel.  That means that a city that averaged two percentage points higher house price growth, for example, also averaged two percentage points higher rent growth over the same 50-year period.

Returning to the question posed at the beginning of this section, instead of a crisis in rental affordability, we have a long-term erosion of housing affordability overall.  It is true that, in many cities, rental costs are higher now than they were ten years ago.  But ten years ago, rental costs were higher than they were ten years before that.  Likewise, ownership costs have followed the same pattern.  It is hard to conclude that there is an affordability cliff from whence we can step back from the brink.  Rather, the threat to housing affordability in this country is much more fundamental, and more economically pervasive.

Does building more apartments solve the problem?

One commonly-proposed policy to address declining housing affordability is to construct or preserve low-income housing.  However, such a policy would be, at best, a temporary panacea to a long-run issue.  Building new housing, if we can build enough, would ameliorate housing costs for the moment, but the benefit would soon be swamped by the rising tide of housing expense.

To understand why this is, we have to understand why housing costs are rising in the first place.  In recent research with Joseph Gyourko and Christopher Mayer, we discovered that a key factor is that naturally occurring growth in the number of households that wish to live in a given city – growth due to fundamental factors such as population growth or immigration – interacts with the limited availability of land in those cities to generate ever-higher housing costs.  This phenomenon is particularly pronounced in broadly-appealing cities – such as San Francisco and New York – where new construction is hampered due to regulation, expense, or topographical barriers.  The high housing costs deter lower-income households from living in these cities, making them enclaves of relatively well-off households (plus some very determined-to-live-there lower- and middle-income households).[3]

There are several implications of this research for rental affordability.  The first is that the economic process that generates rising housing costs is a fundamental, long-term one.  The driver of housing costs is not how many people actually live in a city, but how many want to live there.  For any city, that number rises steadily with population growth (and population growth tends to be a fairly reliable force).  Of course, if everyone who wanted to live in a city could easily do so, prices would not be bid up.  There must also be a constraint on new supply.  However, it is evident from 50 years of historical data that cities more-or-less “fill up” at some point – they transition from being a city with lots of new construction to one with very little – and they do not revert back.  That means that the ever-growing population is not met by ever-growing supply, at least in the desirable, supply-constrained cities, and so housing costs rise disproportionately there for a very long period of time.

The way these economic forces are manifested in the housing market is in ever-increasing house prices, especially in appealing cities with limited growth in housing supply.  The long-run house price growth across various cities is reported in Table 1, reproduced from Superstar Cities


Table 1: Real Annualized House Price Growth 1950-2000

San Francisco leads the list, with house prices that grew (on average) more than 3.5 percentage points per year faster than the rate of inflation for 50 years between 1950 and 2000.  Other areas with high house price growth include Seattle, San Diego, Los Angeles, and Boston.  By contrast, upstate New York cities such as Buffalo and Syracuse and rust-belt cities such as Cleveland and Dayton had house price growth that barely exceeded inflation.  It will come as little surprise, then, that the median home in San Francisco in 2000 was affordable only to a household that exceeded the 95th percentile on income in the U.S.  And, to be able to afford the 10th percentile (by price) home in San Francisco, one needed to have an income above the 85th percentile of the whole country.  By contrast, in Cleveland, the median home was affordable to a household in the 35th percentile of the national income distribution.[4]

These same economic forces are generating the patterns we see in declining rental affordability.  When housing demand first exceeds supply, the subsequent growth in rents makes a city unaffordable for the lowest-income households.  As the trend continues, the rising rent level prices out more low-income households, and then higher-income households.  It is not simply that more households want to live in some cities right now than there are places for them; rather, the gap between the number of households that want to live in the top cities and the available housing is ever-increasing. 

This means, however, that a one-time policy of increasing low-income housing supply is unlikely to be more effective than sticking a finger in one hole of many in a leaky dike.  Sure, Economics 101 says that rent growth will be attenuated if more rental supply is brought to the market.  But the realities of urban housing demand show that the new supply will be quickly filled and, after the brief correction, housing costs will revert to their relentless climb.  To ameliorate this pattern, policy would have to increase the long-run growth rate of low-income housing supply in the cities where historically it has been difficult to build.  That is neither an easy, nor cheap, task.  One could stop discouraging high density – local regulations often make it difficult or expensive to build – but the cost of such density is borne by existing residents, who in many cities have demonstrated an aversion to it.  Elsewhere, fundamental topographical constraints limit the amount of building by raising the cost of construction.  Compensating for that by subsidizing new building could be expensive.

In addition, concerns with housing affordability are no longer the sole province of low-income households.  Expanding the low-income housing supply will not ameliorate the growing problem of housing costs for middle-income households.

How can policy help with rising housing costs?

When housing costs rise unexpectedly, especially when incomes do not keep pace, renters may find themselves unable to afford to continue living in the cities where they already reside.  And, because house prices typically rise when rents go up, renters cannot escape growing rents merely by buying a home after the fact. 

Instead, to avoid the risk of housing costs outpacing one’s income, one has to be a homeowner in the first place.  Because home owners lock in their house price at the time of purchase, when rents rise, a homeowner’s annual housing cost is unchanged.  In addition, the initial savings from renting are illusory.  Over decades, total rental costs often exceed the cost of owning.  In the cities where house prices tend to be much higher than rents, rents tend to grow more rapidly, closing the gap over time.

However, given the recent history of housing markets in this country, it is important to recognize that there are caveats when touting home ownership as the safe way to guarantee oneself a place to live.  It is important that any potential home owner avoid the two behaviors that were the biggest problems in the most recent housing crash: Don’t spend more on housing than one can easily afford, and don’t use too much debt to purchase a house.

Unfortunately, the current subsidy to home ownership – the mortgage interest deduction – fails to spur home owning, as recent research by Christian Hilber and Tracy Turner shows.[5]  Rather, it encourages buying more expensive houses and the use of mortgage debt.  In addition, because the value of the subsidy is tied to the marginal income tax rate, the mortgage interest deduction is most valuable to high-income households – and, as we have seen, they can still afford housing in most cities.  Instead, economists tend to favor a more targeted subsidy, such as a capped refundable tax credit for first-time home buyers.[6]    

Nonetheless, even a subsidy for home buying will neither make the top cities miraculously affordable to low- and middle-income households nor prevent the steady erosion of middle-income housing affordability.  However, there is no easy permanent policy solution to protect low- or moderate-income households from rising housing costs.  At its core, something that this country seems to view as an entitlement – the ability to live in whatever metropolitan area one wants – is becoming very expensive to provide.  Because housing cost growth in some cities has outstripped inflation for so long and by so much, an ever-smaller fraction of the population can afford to live there unless subsidized.  And those subsidies would have to be quite large.  Already, to make the rent of the typical apartment in San Francisco in 2013 as inexpensive as a comparable unit in San Antonio would require a subsidy of nearly two-thirds of the rent – and that presumes that rents in San Francisco would not rise in response to such a subsidy.

Even so, home owning can help.  Housing costs are rising because the incomes of households that could live in a city are growing faster than the incomes of households that already live there.  By buying a house up front, and locking in housing costs at a level one can afford, a home owner can continue to live in a city even as housing cost growth exceeds their own income growth.  That is why many long-time home owning households could not afford to purchase their own houses at their current incomes if they did not already own them, and why low- and middle-income households are disproportionately crowded out of high-price cities unless they bought their houses years ago.  Ironically, policies that promote renting when it is temporarily more affordable can inadvertently expose households to the risk of being priced out of their cities.

Already it is no longer the case that someone can automatically afford to live in the city in which she grew up.  A debate needs to take place about the value our society should place on whether a household should have an unlimited right to choose where to live, and how much they should be insulated – if, at all – from the differences in cost.  There are ways households can adapt to differences in housing costs.  Already, low-income households appear to compensate for rising rents by sharing apartments with more roommates.  Alternatively, households can commute further – and more effective transportation networks could make less expensive areas more appealing.

Lastly, in the U.S., land is plentiful and thus there is plenty of inexpensive housing – just not necessarily where people want to reside.  The time may have come to accept that many households cannot afford to live in the existing high-cost cities and, job-permitting, are instead better off buying homes in up-and-coming cities that are currently cheap – thus insuring themselves against those cities becoming the expensive cities of the future.

About the Author

Todd Sinai

Todd Sinai is an Associate Professor of Real Estate and Business Economics and Public Policy at the Wharton School of the University of Pennsylvania. He also is a Faculty Research Fellow of the National Bureau of Economic Research.  His research focuses on commercial real estate and REITs, real estate and public economics, risk and pricing in housing markets, taxation of real estate and capital gains.

Previously, Professor Sinai has been a Visiting Scholar at the Federal Reserve Bank of Chicago and the Federal Reserve Bank of Philadelphia.  He serves on the editorial boards of the Journal of Urban Economics and Real Estate Economics.


[1] Joint Center for Housing Studies of Harvard University, “America’s Rental Housing: Evolving Markets and Needs,” 2013.

[2] This figure is reproduced from Todd Sinai, “Feedback between Real Estate and Urban Economics,” Journal of Regional Science, vol. 50, number 1 (February 2010), pp. 423-448.

[3] Joseph Gyourko, Christopher Mayer, and Todd Sinai, “Superstar Cities,” American Economic Journal – Economic Policy, vol. 5, number 4 (November 2013), pp. 167-199.

[4] See Sinai, “Feedback between Real Estate and Urban Economics” (2010).

[5] Christian Hilberg and Tracy M. Turner, “The Mortgage Interest Deduction and its Impact on Homeownership Decisions,” Review of Economics and Statistics, forthcoming.

[6] Richard Green, “Testimony of Richard Green to the Senate Finance Committee,” October 6, 2011.