Want housing that doesn't break the bank for lower-income residents? Get rid of some parking.

The National Low Income Housing Coalition  released their "Out of Reach 2013" report a few weeks ago, and it's a fascinating look at how unaffordable housing really is for low-income workers in this country. Most interesting is their "hours of minimum wage work needed to afford rent" chart, as seen below:

The metric is a bit contrived in that many minimum wage workers don't actually need a two-bedroom unit, and if they do they're more likely to be sharing that cost with a partner, but their conclusions are valid nonetheless: if you're poor in America, housing probably accounts for a huge portion of your income and leaves you little money for other essentials like food, transportation, and healthcare. And as Matt Yglesias at Slate points out, that's especially true in traditionally "blue" states:

More liberal states typically have higher minimum wages, but it's not generally the case that liberal states have a better housing affordability picture for low-wage workers. The least-affordable states—New York, New Jersey, Maryland, D.C., California, Massachusetts, Delaware, Virginia, Connecticut, New Hampshire—are a very disproportionately blue bunch. And the problem is that the impact of high regulatory minimum wages in many of these states is swamped by the impact of excessive restrictions on housing supply.

That's right on. And while ultimately agreeing with his message, the original intent of this post was to nitpick at Yglesias' conclusion regarding blue states--sure, housing is more expensive, but what about transportation? Transportation, after housing, is typically the second-highest cost for households, and blue states tend to do better on providing cheaper transportation in the form of buses, rail, and more opportunities for walking and bicycling.

So, while Massachussetts may require half again as many hours of work as North Carolina to afford a nice two-bedroom unit at 30% of total income, a transit-dependent Boston-ite might be spending just 10% of her income on getting around while the Raleigh-an is spending upwards of 20% to drive to work (and everywhere else) from the suburbs every day. Instead of working 40 extra hours, the transit user can just redirect their transportation savings toward housing costs.

This doesn't diminish the message of the Low Income Housing Coalition--it's clearly still unacceptable for someone to need to work even 80 hours a week just to survive, much less thrive--but it does change how we look at the numbers in each state. Where you choose to live has a huge impact upon the amount you end up paying for transportation, so to look at one or the other in isolation is misguided. Just look at the chart below--depending on location and transportation choices, the sum of housing plus transportation costs can range anywhere from 41% of income to 57%:

From Reconnecting America and the Center for Transit-Oriented Development's report, TOD101 [PDF].

Yglesias is certainly correct, however, that housing prices are out of control in many big, liberal cities. Just ask San Francisco, where the the average rent for an apartment now exceeds $2,700. Limitations on housing supply are the main culprit behind skyrocketing costs in high-demand areas, and those restrictions can take many forms. One of them is parking mandates, a topic recently picked up by the guys at Freakonomics, and also the subject of a study by Michael Manville of UCLA (via Alex Block). The abstract for his paper, "Parking requirements as a barrier to housing development: regulation and reform in Los Angeles," lays out the case:

Using a partial deregulation of residential parking in downtown Los Angeles, I examine the impact of minimum parking requirements on housing development. I find that when parking requirements are removed, developers provide more housing and less parking, and also that developers provide different types of housing: housing in older buildings, in previously disinvested areas, and housing marketed toward non-drivers. This latter category of housing tends to sell for less than housing with parking spaces. The research also highlights the importance of removing not just quantity mandates but locational mandates as well. Developers in dense inner cities are often willing to provide parking, but ordinances that require parking to be on the same site as housing can be prohibitively expensive.

As he notes, minimum parking requirements result in more space being dedicated to parking than is really needed; in a world of height limits, floor-area ratios, and endless other development regulations this necessarily leaves less space for actual housing. What really struck me, though, was the straightforward assertion that housing marketed toward non-drivers sells for less than housing with parking spaces. It's powerful, but it's also obvious: parking costs money to build, so of course buildings with less parking are cheaper. But to have research-driven data behind it adds force to the conclusion.

Right now, parking is usually required in most localities at a ratio of at least one parking stall per housing unit (often more), and in newer buildings it's mostly provided underground. Even though it's ultimately just a big slab of concrete, underground parking spaces cost between $30,000 and $50,000. Each. Sometimes more. Diggin' ain't cheap.

Developers aren't stupid, and they aren't interested in building parking spaces as charity, so they're going to recoup those costs one way or another. They could try to charge residents for the parking, a difficult prospect in some locales where curbside parking is abundant and cheap (or free). To break even, they'd have to rent out every space for every month for thirty years, for between $85 and $140 per month. Or they could just wrap the cost into everyone's rent and give everyone a free parking space. As you add more parking spaces, obviously the cost goes up.

In reality, it seems that the more common route is to charge a sub-profitable amount, like $50 a month, and to pass on the rest of the cost as increased rent. If you don't own a car you get to save a lot of money on a loan, gas, insurance, maintenance, and so on, but when it comes to building-provided parking you just get to eat the difference.

And these are low-ball figures. Most developers aren't going to wait thirty years to turn a profit, and project size can heavily impact per-parking-space construction costs. According to a developer in Portland, for him, the requirement to build parking was "the difference a $750 apartment and a $1,200 apartment." For a low-income worker, this is the difference between affordability and impossibility.

In the case of above-ground structured parking, it looks more like this: The average parking stall takes up 320 square feet (including circulation aisle, ramps, stairways, etc.), and we can estimate that the average 1,000 square foot apartment takes up closer to 1,300 square feet when taking hallways, storage rooms, etc. into account. Above-ground parking is cheaper to build than underground, but here we have to take into account opportunity costs. By being forced to provide parking instead of their preferred number of housing units, the developer is passing up one 1,000 sq ft unit per four parking stalls. As a 2-bedroom, an apartment of this size would go for a minimum of $1,400 a month in most big cities, often much, much more, so the developer would have to charge around $350 per month for each parking stall to make the same amount of money. No one would rent a space at that cost if they had a choice, of course, so some of that cost gets baked into the cost of rent instead.

The developer probably takes on some of that cost as well, which, at the margin, means they don't invest in otherwise appealing projects when excess parking prevents them from recouping their investment in an acceptable time frame. These are the "disinvested areas" that Michael Manville is talking about.

We all know that public transportation is often the difference-maker for affordability, especially in dense metropolitan areas where housing can be much more expensive than in the suburbs. But astronomical housing costs in the city aren't some immutable Law of Civilization. Those prices are affected by total supply, yes, but they're also strongly influenced by the type of housing that developers are allowed to build.

Allowing more parking-lite and parking-free housing will provide greater financial incentives for individuals and families to try less car-dependent lifestyles, leaving them with more money and everyone else with less congestion and pollution, as well as stronger local economies. But financial incentives alone won't convince people to get rid of a vehicle when there's no viable public transportation alternative, when a car is the only path to mobility. Transit and housing flexibility are Two Great Tastes that Taste Great Together™, and both, not just one or the other, should be promoted vigorously if we want a lasting solution to housing affordability for those who need it most.