How much does a one-bedroom apartment cost in Chicago, my hometown? A quick Google search comes up with an article claiming that median rent is $1,970, according to the real estate company Zumper.
But wait—according to real estate company Trulia, the median rent in Chicago was just $1,400 in January 2016, and that includes apartments with two or more bedrooms.
And the real estate company Zillow reported that the median rent was $1,647 just a few months ago.
And HUD says that “Fair Market Rent” in the Chicago metro area was just $922 for a one-bedroom apartment in 2015.
Not only can these different sources not agree on how much a “typical” apartment costs, they give very different pictures of how much (and in what direction) prices are changing. For example, Abodo claims that Boston is one of the cities where rents increased the most in February 2016—5 percent month over month. (By the way, that’s an increase of nearly 80 percent a year, which is by itself a red flag that something might be up with these numbers.) But in the same month, Zumper says that rents in Boston fell by 2.1 percent.
What’s going on?
Over the last few years, there’s been an ever-growing number of articles and lists of the most expensive rental cities in the county, the most expensive neighborhoods, and which cities are seeing their rents climb the fastest—and even where rents might be starting to fall. But what are presented as clear, objective findings are in fact coming from many different sources, most of which disagree with each other substantially. And while many sources are just fine for their original intended purpose—being a place to sift through real estate listings—they’re not so good as a statistical research database.
How, then, can journalists and readers know which sources to trust, for which questions, in what areas?
The first step is to acknowledge that none of these sources have the definitive, “correct” answer. All of them involve some guesswork and data that’s limited, biased, or somewhat out of date—and often all three. The advent of online listing agencies with massive databases has brought us the mixed blessing that is big data. While it’s a simple matter to query your big database and compute the median value of rental listings, there are a lot of reasons why every organization’s particular collection of listings is an incomplete and statistically biased sample. Unlike home purchases, landlords don’t have to report their rent figures to to any public agency, so collecting broad, representative data on rents can be extremely challenging. As we wrote back in November:
For a number of reasons, just taking the average of all the listings you can find is likely to produce extremely skewed results, with numbers much higher than true average home prices. For one, many apartments, especially on the lower end of the market, aren’t necessarily listed in places that are easy to find—or at all. Instead, landlords find tenants with a sign on a fence or streetlight pole, local (and not necessarily English-language) newspapers, or just word of mouth. On top of that, if you have two homes of similar quality but even slightly different prices, you would expect the cheaper one to rent or sell more quickly. As a result, it would spend less time listed than the more expensive home; any given sample of listings, then, would tend to over-represent those more expensive, harder-to-rent homes. (If this doesn’t make sense, read the “visitors to the mall” example here, explaining a similar statistical problem with attempts to measure prison recidivism.)
But that doesn’t mean you have to throw up your hands in rental-data nihilism. There are important differences in the reliability of rental price statistics from different sources—and, just as crucially, different sources are best for different questions.
Tomorrow, we’ll publish a sort of field guide to rental statistics, meant to help both journalists getting a barrage of press releases, as well as readers trying to wade through the swamp of contradictory rental figures that are published on a regular basis.