Are There Any Affordable Cities Left in America?

When you factor in both housing and transportation costs (H+T) as a percent of income, the car-dependent cities in the right column expensive. But are DC, SF, and NYC that much more affordable, even if you count the benefits of transit? Source: Citizens Budget Commission
When you factor in both housing and transportation costs (H+T) as a percent of income, the car-dependent cities in the right column are especially expensive. But are DC, SF, and NYC that much more affordable, even if you count the benefits of transit? Source: Citizens Budget Commission

Are Washington, San Francisco, and New York the most affordable American cities? A new report from the New York-based Citizen’s Budget Commission [PDF], which made the rounds at the Washington Post and CityLab, argues that if you consider the combined costs of housing and transportation, the answer is yes.

But a closer look at the data casts some doubt on that conclusion. Between the high cost of transportation in sprawling regions and the high demand for housing in compact cities with good transit, very few places in America are looking genuinely affordable these days.

The CBC report uses a better measure of affordability than housing costs alone. Transportation is the second biggest household expense for the average American family, and looking at what people spend on housing plus transportation (H+T) can upend common assumptions about which places are affordable and which are not. Regions with cheap housing but few alternatives to car commuting don’t end up scoring so well.

There are some problems with the CBC’s methodology, however. While abundant transit is absolutely essential to keeping household transportation costs down, and it provides a lifeline to low-income residents of major coastal cities, the report still tends to exaggerate overall affordability in these areas.

According to the report, for example, New York City ranks third in affordability among 22 large cities. A “typical household” in New York City, the CBC finds, spends 32 percent of its income on housing and transportation combined. Part of the reason New York comes out looking good, though, is that CBC used a regional measure of income but looked at typical rents only in the city itself. Because the region’s median income is higher than the median income in the city ($62,063 vs. $51,865, respectively, according to 2008-2012 Census data), NYC appears more affordable than it really is.

Another issue, flagged by Michael Lewyn at his CNU blog, is that by looking at average rents, which in some cities include many rent-stabilized units, the calculation doesn’t necessarily capture what someone searching for shelter is likely to pay. If you’re trying to find an apartment in New York now, getting a place for the average rent would probably be extremely difficult.

What really stands out in the CBC report isn’t that New York, San Francisco, and DC are affordable — it’s that car-dependent areas that may have cheap housing turn out to be so expensive once you factor in transportation.

A commonly accepted definition of “affordability” is when housing and transportation costs consume no more than 45 percent of income. For moderate income families in several cities, including Jacksonville and San Diego, combined H+T costs exceed this threshold. (In these areas, the central city tends to account for a large share of the whole metro region, so the region-vs.-city issue that skews the NYC data is less of a problem.)

While the report didn’t include Midwestern cities known for both low housing costs and lower transportation costs, like Minneapolis and Pittsburgh, the overall picture is alarming.

There’s no good evidence that in places with good transit, low transportation costs are enough to compensate for the rising cost of housing. Meanwhile, many places that do have cheaper housing saddle low-income people with mandatory car ownership and high transportation costs.

66 thoughts on Are There Any Affordable Cities Left in America?

  1. The real enemy isn’t opposition to the bike infrastructure; it’s out-of-control property speculation that leads to people moving way outside the zone where they’d be able to have a reasonable commute distance.

    That’s part of the problem, but even if it were solved people are still going to want to take trips between boroughs for a variety of reasons. That’s where a bike highway network would come in.

    And yes, given NYC’s horrible track record on any bike infrastructure which involves more than laying thermoplast I’m not holding my breath waiting for this. It will take a sea change in the way this city is governed, plus another sea change in how we approach major infrastructure projects, to get this done. Right now if NYC embarked on building out a bike highway system, it probably still wouldn’t be finished in my lifetime (I’m 51) given the glacial pace other construction projects move at. By then the first parts might start falling into disrepair, but of course NYC wouldn’t repair them until they actually fell apart. Even then, it would most likely be a half-assed repair which would need to be redone constantly, as they’ve done with that sinkhole on the West Side Greenway.


    Alright, this is more realistic than AAA. One thing to note is how much costs decrease after year one. If you were buying a used car just one-two years old, you wouldn’t pay nearly as much for depreciation, and would not get hit with massively elevated operating costs. And if you really want to be cheap, if you look you can find a car that is full of dents from getting stuck in a hail storm, but otherwise in perfect condition, that you can get for a bargain. Keeping a car for 8-10 instead of 5 years is perfectly feasible and cuts costs significantly as well.

    You can get a cheap car if you need one. Plenty of people do.

  3. Well, that link ought to be enough to tell you Alon’s averages are terrible.

    But the studies jive with each other quite well. CR uses 12,000 miles as a baseline, which is a survey average. and AAA uses 15,000, so of course AAA will be a bit higher. It may actually be more realistic when considering suburban areas too. I selected a vehicle based in part on @anon_coward_12:disqus’s scenario, mind you, probably being charitable to his argument. I even said I was cherrypicking (like him).

    Anyway, even going by the cheapest vehicle listed there, you still need to net thousands of dollars/year over housing savings over transportation. And both those studies still make troubling assumptions, like treating vehicle storage as free, which is certainly a factor when comparing urban metro areas.

    And your assumption about getting lightly used cars may work sometimes, but it’s hardly a panacea for macro-level transportation policy given good lightly used models are probably the hardest used models to get.

  4. Vehicle storage is free in most places, at least to the parker. These are also prices for new cars. Used cars aren’t cheaper than new because they are cheaper to own, there is a several thousand dollar psychological penalty for buying a new car, they also have better amenities etc…Maintenance and fuel are more expensive for older cars, but not by as much as you think, depreciation is a major expense for new cars, not so much for older cars. 12k miles a year is plenty in much of the country. Leaving NYC doesn’t mean heading to an exurb. Plenty of inner suburbs with short commutes are affordable, and in smaller metro areas all there really is. Total cost of a used car can be well under 5k. And remember, the average age of a car on the road is more than 11 years. Most cars aren’t new. That new cars are more expensive than the average doesn’t mean the average isn’t much less than 9k.

  5. Storage can range from cheap to expensive depending where you are, but it’s never free and not accounted for at all in any of those methodologies.

    Of course total cost of a car can be extremely low, especially after the loan is paid off, but it requires light use, which probably is impractical in a car-dependent area. So maybe good transit has potential to make cars more affordable!

  6. Maybe we got sidetracked, but when looking at housing + transportation costs it’s probably already included in the cost of housing in much of the country, so you would calculate the cost of a car without it.

  7. Sidetracked? I’m not even clear what you’re trying to point out. If you’re going to make comparisons, model what people do spend, not what they could spend. We could all move to an apartment next to work and walk and bike everywhere (theoretically). Your low expense car anecdotes obviously occasionally happen, but to model this you still need to multiply the cheapest realistic ownership scenario you can come up with by the number of household members* who drive to have a realistic floor on household transportation costs with car dependency.

    As far as storage costs go, that’s a fair point with regard to the Citizen’s Budget Commission report. But it isn’t considered in these “cost of car ownership” analyses, and I’d think it would at least be relevant when comparing urban cores. Even a $5 garage fee at work every day is $1,300/year multiplied over 260 working days. Even if you call that small, it’s non-trivial.

    * And, of course, there are potential hybrid scenarios, like a single-car family where one spouse drives to work and the other depends on transit when the car is not available.

  8. The underlying problem is that the 99% haven’t seen any income gains in decades.

    This is really quite straightforward to fix. You take money from the billionaires (through taxation) and you give it to everyone else. It is also quite obvious why this hasn’t happened yet — the billionaires can afford to buy Congressmen and state legislatures.

  9. The age and condition of NYC housing is another matter, of course. But NYC is weird and an outlier.

  10. I don’t know about that. 100+ year-old brownstones have held up at least as well as 60yo suburban housing stock.

  11. It seems to me that car ownership costs are very individual. Do you own an older, smaller fuel efficient car or a big Hummer? How does the methodology
    look at the variations in car ownership costs between two people? Who works on their own car? Who owns a car that is expensive to maintain? I am just wondering if someone could explain how the studies generalize about the cost of car ownership. An older Honda and a new SUV are very different cost-wise.

  12. There are several inherent problems with all studies that allege that there are savings in “housing plus transport” costs, to be had from “smart growth” type policies, and that unaffordable housing in many urbanists favourite cities is counterbalanced by “lower transport costs”.

    It is basic land economics, that transport cost savings capitalise into site values anyway. If there are savings to be made by living at location A instead of location B, then location A will be priced higher than location B to an extent that reflects the savings. In fact the premium is always HIGHER than the transport costs saved, probably reflecting perceived value of time savings.

    Some studies allege to prove something different to this, and one of the most common flaws, is that they use the “average cost of housing” of actual households in place today; which includes households who bought their first home decades ago and have only a small mortgage remaining, and some paid off altogether. In an unaffordable-housing city, recent buyers with large mortgages will all be found at the locations where houses are NOT so unaffordable, weighting those locations with very high H+T costs. What is actually needed, are studies that analyse the options of people deciding on their location now, not people who have been on the urban housing ladder for a long time already.

    Another approach which is not quite so bad, is the use of “rents” rather than mortgage costs; however this is still flawed because a feature of unaffordability and bubble markets, is that rents and actual house prices are misaligned. Rents remain less inflated than prices. So we need to be clear in such cases, that we are foregoing the benefits of ownership versus renting. Also, rents are skewed downwards relative to the housing market per se, by the fact that rental accommodation is generally much smaller than owned accommodation; and this is much more so the more centrally located the housing is.

    Another common flaw in these H+T studies, is the use of costs of automobility that are estimates, or derived from averages, that include the significant discretionary purchases of new cars that depreciate rapidly, and of expensive and thirsty vehicles, by better-off people who can afford them. Generally people who are constrained by their ability to pay for location, will also choose their vehicles carefully. Anthony Downs points out in “Still Stuck in Traffic” (2004) that the “lowest cost” option for automobility in real terms has steadily fallen for decades; the cheapest option in, say, 1975 might have been a 1960 Ford “subcompact”; the real cost now of buying and running a 1999 Asian made 1300cc hatchback will be VERY much lower.

    People who regard these rigged H+T cost analyses as credible, are simply out of touch with real-life young couples making their decisions of location, house type, and so on. It was my acquaintance with these realities, years ago, that motivated me into a long crusade ever since, against utopian Marie Antoinette type elitists peddling lies as part of a systemic swindle of younger people and the poor.

    There is also other databases of living costs that do not at all reflect the alleged advantage of more compact but expensive-housing cities: eg

    “The C2ER Cost of Living Index for 307 Metropolitan Areas”.

  13. As per ITDP guidelines, all Transit Oriented Development must have 10% affordable income housing minimum.

    Also, let’s look at how much you can save by not owning a car. An economical car costs $8000 a year. I’ve seen about $6500 for a Honda Fit (most economical car according to Consumer Reports) driven 12,000 miles a year, but that fails to include parking, tolls, and that the average car commuter drives more like 15000 miles per year just commuting (assuming a 20 day working month, 12 working months a year, and a 30 mile commute), and probably drives even more than that in the course of his/her daily life. What’s more, car commuters lose time to servicing and fueling the accident and causes enough pollution to create a negative loss for all of society per mile he/she drives.

    Now let’s look at the cost of commuting in NYC on the MTA. That would cost just $1320 per year, assuming you work 20 days a month (after weekends, holidays, sick days and so on). Assuming this commuter (like most NYC households) does not own a car, that’s a savings of $6680 on commuting alone over the car dependent person assuming that the car commuter drives an economy car (like the Honda Fit), and doesn’t include the health benefits of walking to and from the stations, the savings from not having to drive to the store, to your child’s (or children’s) school(s), to the doctor’s office, and any other place you may go. That’s very likely enough to pay for the extra cost an apartment, condo, or co-op unit in Brooklyn, Queens or the Bronx near the Subway and grocery stores if not more, and can help to substantially reduce the extra cost of equivalent units in Manhattan. It is very likely a similar situation in Washington DC and San Francisco. If I were to bet, this effect is strongest in Philadelphia and Chicago.

    Just my two cents.

Leave a Reply

Your email address will not be published. Required fields are marked *


The True Cost of Moving to Cheaper Suburban Housing

Today Streetsblog Network member The City Fix reports on the "cost of place" in the Washington, DC, area — the way that the price of housing and transportation stacks up for people in the urban core and the suburbs. According to a report recently released by the Urban Land Institute, Photo by ehpien via Flickr. […]

The Cost of Sprawl on Low-Income Families

Via the Manhattan Institute’s new blog, Streetsblog learns of a pdf-formatted report entitled A Heavy Load: The Combined Housing and Transportation Burdens of Working Famillies, which looks at the housing and transportation expenses paid by lower income families in a number of cities. The report, published by the Center for Housing Policy, a K Street […]

High Transportation Costs Make a Lot of HUD Housing Unaffordable

Rental assistance from HUD isn’t enough to make the cost of living affordable when the subsidies go toward housing in car-dependent areas, according to a new study by researchers from the University of Texas and the University of Utah. The study evaluated transportation costs for more than 18,000 households that receive HUD rental subsidies, estimating that nearly half of recipients have to spend more […]