Eight years ago, the United States economy was in the throes of its largest recession since the Great Depression. The U.S. housing market was in severe decline. In the years before the crisis, banks had facilitated the explosion of “subprime” mortgage lending whereby borrowers with poor credit history were offered mortgages they couldn’t dream of repaying. Eventually, the U.S.’s housing bubble burst and evictions, foreclosures, and prolonged unemployment rose across the country. Even Wall Street’s biggest firms were not immune to the fallout from the housing market’s collapse. Many banks had packaged these subprime mortgages together into collateralized debt obligations (CDOs) and traded them as securities. As soon as the real estate market took a turn for the worst, so too did the finances of banks holding large amounts of CDOs. When I was working at a Miami Beach Real Estate firm for a few years, this had happened in my career as well.
Eight years and a financial stimulus package later, I started to wonder how the U.S. housing market was doing today. As a soon-to-be college grad, I had heard plenty about the exorbitant monthly rents facing tenants in New York and San Francisco. I had even heard rumors that it might be less expensive to commute to San Francisco four times a week from Las Vegas than live in SF. From all anecdotal indications, the real estate markets in two major U.S. cities seemed to be booming. But what about the rest of the country? Even as recently as May 2015, the Detroit News wrote that the volume of unoccupied homes in the city was “absolutely terrifying.” In Las Vegas, the foreclosure rate still well exceeds pre-crisis levels. So was the real estate recovery only happening in select pockets with thriving economies? Or was the recovery more widespread?
To analyze the current state of the U.S. housing market, I decided to use data pulled from the real estate website, Trulia. Through its API, Trulia offers aggregated data about the median listing price of real estate in a given state, city, or even neighborhood. The data accessible through the Trulia API stretches back to July 2009, meaning that I could feasibly analyze housing market developments since the end of the financial crisis. After pulling in data from the API, I imported it to Tableau. The following data visualizations speak to a recovery in real estate that is progressing but far from over.
I started by looking at month-over-month changes in median listing price at the state level. Note that the listing price of a property is the price at which a seller offered the property on the Trulia platform, not necessarily the final price at which it was sold. Even so, median listing price provides a strong indication of housing market health. Sure, sellers often list their property at a price slightly higher than its true market value, in hopes of deriving a little surplus in the sale. But we should expect most sellers to take this approach across time. So long as this bias is equally prevalent at an aggregated level across months, median listing price should offer an effective way of tracking the directional trend of real estate prices.
The map below shows the trend in real estate prices by state since mid-2009. Note that the percentage change for a given month is calculated relative to July 2009—that is to say, if the median listing price for a given state was $300,000 in July 2009 and drops to $240,000 in a subsequent month, the change in median listing price will be reflected as -20%. For that month, the state will be colored red on the map. Darker shades of red indicate more significant drop-offs in median listing price; darker shades of green indicate larger increases.
A few interesting insights derived from this visualization: