Nonprofit, nonpartisan journalism. Supported by readers.


Mortgage lenders should consider transportation costs

Courtesy of MnDOT
Commuting ten additional miles will cost $91,200 over the course of a 30-year mortgage. logo

Mortgage lenders turn a blind eye to the transportation costs of homebuyers, and that’s bad for borrowers, banks, and urbanism.

When a bank is deciding whether or not to lend you money to buy a home, they don’t really care whether you can afford to repay the loan, because they’re not holding on to your mortgage. It’s more likely that your small neighborhood bank plans to sell your loan to a larger financial corporation that will either sell it again or make mortgage sausage with it.

So what the loan officer is really worried about is whether your loan appeals to these big companies, that is, whether it meets Fannie Mae and Freddie Mac guidelines. These guidelines usually come close to approximating a borrower’s ability to repay, but they have a glaring flaw that’s also bad for cities: they don’t account for transportation costs.

The lender knows where the borrower lives. The lender knows where the borrower works. But the lender refuses to infer anything about the financial cost of the borrower’s regular commute between these two places. This doesn’t make any sense. Transportation costs are easily predictable and have a direct bearing on a borrower’s ability to repay the loan.

Housing developers have used this loophole to sell remote and inexpensive homes. In The End of the Suburbs, Leigh Gallagher relays the sales pitch of a developer in Wright County, Minnesota: “Drive 10 miles and save $10,000.” Simple math shows that it’s not worth it. AAA estimates that an average sedan costs 60.8 cents per mile to drive (to account for fuel, tires, depreciation, etc.), so commuting those ten additional miles will cost $91,200 over the course of a 30-year mortgage. Lending money to a borrower so they can pursue a financially-irresponsible lifestyle is bad for business: the foreclosure crisis has hit far-flung suburbs especially hard.

But the Wright County salesman’s pitch worked, and homebuyers have moved farther and farther from the core cities, ostensibly for affordability. I see two problems with this. First, for the homeowners, a long commute means that they can spend more than a quarter of their household income on transportation, easily offsetting their savings on the house itself. The Center for Neighborhood Technology’s Housing + Transportation map is a dramatic illustration of this, and they have a slick two-page PDF about affordability in the Twin Cities. Second, for the metro region as a whole, the phenomenon leads to sprawl and the concomitant hollowing-out of urban centers — with the all-too-familiar ills that follow.

This post was written by Scott Shaffer and originally published on Follow on Twitter: @streetsmn.

If you blog and would like your work considered for Minnesota Blog Cabin, please submit our registration form.

Comments (15)

  1. Submitted by Dan Landherr on 09/25/2013 - 10:54 am.

    A mortgage is 30 years

    Who can predict where they will work for the next 30 years?

    • Submitted by Todd Hintz on 09/25/2013 - 04:37 pm.

      30 Years hence

      That depends on what choice you make each time you move. Do you buy a house ten miles farther out or do you find a house that meets your needs that’s close to work?

      I’ve had plenty of coworkers who didn’t live ten miles farther out, but more like 60 miles because they “wanted to be out in the country.” They were a little put off when I pointed out that they would see more of their windshield than they would the fields out back of the hacienda.

    • Submitted by Jeff Klein on 09/26/2013 - 10:13 am.

      Nobody exactly, but…

      … clearly some places are likely to have lower transportation costs. You’re more likely to be commuting to work in 30 years if you live in Ham Lake vs. Minneapolis. All of these things are statsitically based anyway. You could make the same argument about your pay, for example – why include pay in a mortage approval when you don’t know what job you’ll have in 30 years? And the answer is that over large numbers of people you can make pretty accurate predictions.

  2. Submitted by Steve Titterud on 09/25/2013 - 11:05 am.

    To be consistent with your theory, ….

    …mortgage lenders should also consider additional factors as well.

    E.g., being a vegetarian or vegan and eating solely organic is vastly more expensive than eating the ordinary fare found in the supermarket.

    Take a look at the table of sampled prices of organic vs. non-organic, showing the organic prices are in the range of 40% – 133% HIGHER than non-organic, at – and this is from an organic farming advocacy organization !!

    So isn’t it obvious, by your logic, that people who eat organic foods should also be downgraded in mortgage underwriting with this negative factor ?

    Same goes for having anyone in the household who has no health insurance – couldn’t this wipe out the family’s ability to pay ? If so, shouldn’t it be an underwriting criterion ?

    Your thesis that you are equally concerned for borrowers and banks alongside “urbanism” seems very thin.

    Truthfully now, isn’t your main point of advocacy the “…bad for cities.” part – NOT the mortgage initiator’s poor underwriting practices nor the fallacies of the exurban buyer ?

    • Submitted by Alex Cecchini on 09/25/2013 - 12:19 pm.


      if banks really wanted to go all-in on some actuarial data mining to use lifestyle factors like insurance, family size (or likely family size), eating habits, etc and use it as a predictor for how likely a person is to pay back their mortgage over 30 years, I’m sure they could do this. How moral (and therefore legal) it is to use increasingly personal information like this to judge a person’s maximum loan amount is up for debate, I guess. And yes, an individual can change jobs (and therefore location) over the course of a 30 year mortgage, which would affect this. How is that any different than a career change that may have an affect on salary in the future, which a bank cannot/does not take in to account in their evaluation of borrowers? The point of the article is that a person or couple’s situation at the time of applying for a loan, including the location of the home and their jobs, is a relatively easy way to determine their monthly outlays due to transportation and housing. People should be doing this math on their own, but unfortunately they hear/see realtors explaining the savings by moving further out, are told by banks they can afford it, and may not be able to accurately predict the costs of dual+ car ownership and operation (including gas, insurance, maintenance, etc) over the next 10+ years. National policies, marketing, and prevailing beliefs about affordability keep people from realizing true costs.

    • Submitted by Scott Shaffer on 09/26/2013 - 09:44 am.

      No data-mining necessary

      I’m not saying the mortgage lenders should delve into our day-to-day spending habits, like on groceries. They shouldn’t, for a bunch of reasons. My grocery-spending habits can change from day to day. Maybe eating healthy is expensive in the short term but is cost-effective in the long term. Borrowers want privacy. Data-mining is expensive.

      However, every lender already has the relevant data needed to determine the borrower’s commute: the address of the primary residence and the address of the employer. So additional data-mining or privacy breaches are unnecessary. The habit (commuting) will not change as long as the borrower lives at address A and works at address B.

      There certainly are other, longer-term detriments to a long commute, like weight gain and increased crash risk, but they’re hard to quantify. So I don’t try to in this piece. I can’t think of any benefits to a long commute, unless you really really like Dave Ryan and his morning crew.

      So no, I don’t think my argument logically extends to surveilling prospective borrowers’ shopping habits.

  3. Submitted by James Hamilton on 09/25/2013 - 02:50 pm.

    Check my numbers

    10 miles x 2 per day = 20 miles
    20 miles x 5 days per week = 100 miles
    100 miles x 52 weeks per year = 5200 miles
    5200 miles x 30 years = 156,000 miles
    156,000 x $.608 = $94,848.00
    Monthly cost: $263.47

    $10,000 at 5% over 30 years:

    Monthly cost: $53.68
    Total cost: $19325.58 l

    Net cost: $75,522.42

    Transportation costs can be reduced by approximately 25% by driving a small sedan. ($.464 per mile). That still leaves a net cost of approximately $56,641.00

    The numbers look even worse if one factors in tax savings on deductible mortgage interest payments and on 401K or other tax deferred investments if the net cost per month was invested.

    On the other hand, the author doesn’t seem to take into account that the same money buys more house in the ‘burbs than it does in the city, the dramatic differences that can exist between schools, or dozens of other factors that come into play in choosing one’s home. But then, his goal here is to argue against urban sprawl. While I can’t say that every homebuyer crunches the numbers as I have above, it’s naive (at best) to assume that they don’t recognize the additional costs they will incur by a longer commute. As for lenders taking it into account: they don’t care where else you spend your money. Nor should they.

    • Submitted by Alex Cecchini on 09/25/2013 - 07:40 pm.

      The house in the burbs

      is only ‘bigger’ or ‘more’ for the same money because it doesn’t pay the cull cost of its infrastructure and environmental impact. Nor does it really take in to account the social cost thanks to driving as the only way to get around (namely car accidents, among others). Schools in inner and second-ring suburbs have proven to degrade (with few exceptions), proving that being new is the only thing that gives them a temporary advantage (also ignoring the role family and home-life plays in education and development).

    • Submitted by Janne Flisrand on 09/25/2013 - 09:50 pm.

      Full cost of driving

      James, did you include all the direct driving? Depreciation, tabs, insurance, etc. etc. etc.? I’d like to see where you got $.608 as the full per-mile cost.

      Here’s a list of what costs maybe should be included — you can quibble with the numbers, but they all need to be there.

    • Submitted by Scott Shaffer on 09/25/2013 - 10:36 pm.

      It’s about the ability to repay

      I was assuming two weeks of vacation (perhaps naively), which got me the $91,200 number. It doesn’t make much of a difference, $253.33 per month by my calculations, or your $263.47 for the poor souls who work 365 days a year.

      I think you’re wrong when you write, “it’s naive (at best) to assume that they don’t recognize the additional costs they will incur by a longer commute. As for lenders taking it into account: they don’t care where else you spend your money. Nor should they.”

      Taking a look at the recent foreclosure and poverty spikes in the far-flung exurbs, it’s obvious that many homebuyers did not rationally evaluate their financial risk, and it seems that huge transportation costs might have been what sunk their budgets. That developers encouraging their customers to move farther from population centers in the name of affordability (not better schools or bigger homes) supports this view.

      And as far as banks not caring about where you spend your money, my main argument is that banks should care about their borrowers’ ability to repay the loan. If there are two borrowers who earn the same salary and want the same amount of money to buy a home, and everything else about their credit profiles is equal, but borrower A wants to live 50 miles from work and borrower B wants to live 10 miles from work, it would be smart to lend to borrower B before lending to borrower A. Because transportation isn’t free, and that’s a significant and durable financial obligation borrower A is undertaking. Right now, lenders make no distinction based on transportation costs.

      Notice this holds even if both their jobs are in Eden Prairie, and so my argument isn’t inherently anti-suburb.

      • Submitted by Dan Landherr on 09/27/2013 - 01:44 pm.

        There’s something here but you may have it backwards

        Housing has more intrinsic value if it is closer to employers, less if it is farther away. One reason we saw more foreclosures in outer ring suburbs is the rise in transportation costs dropped the value of the house so the borrowers were underwater. The problem was the original valuation of the property. If the property holds value then it doesn’t matter if the borrower can’t afford it due to job loss because they can sell the property later.

        It isn’t just transportation costs that affect the value of real estate. Tax policy and interest rates arguably have more direct influence. There are many factors that determine the value of real estate and you’re focused only on one of them.

    • Submitted by Jeff Klein on 09/26/2013 - 10:18 am.

      Where you spend your money

      But transportation costs aren’t just a choice you can change, like spending a lot of money on expensive wine or some such thing. Nor are they, like a commenter above suggests, like eating vegan to save money. Transportation costs are inextricably linked to where you choose to live, and can’t be avoided. Transportation will always cost more for residents of Plymouth than residents of Minneapolis (and just because you can point out anecdotal exceptions doesn’t change the average). By not taking into account they’re allowing the same person to get more mortgage in the suburbs than in the city, unfairly subsidizing (in yet another way) sprawling suburban development.

      I’m honestly kinda surprised to see the blowback on this relatively simple argument on Minnpost, of all places.

  4. Submitted by Alex Stenback on 09/26/2013 - 10:27 am.

    Why stop there? Why start there?

    Here’s a list of things underwriters do not consider in evaluating a borrowers ability to repay a loan.

    – Child care expenses (with the exception of the VA – they look at this)
    – Any form of insurance expense aside from homeowners (auto, life, health, etc.)
    – Pre-tax deductions/witholding from payroll.
    – Cell phone
    – Utilities (gas water electric)
    – cable/sat/streamed media expenses.

    All of them easier to measure (one can quantify, not apply an IRS formula) and neither more nor less transitory than distance from home –> Job.

    • Submitted by Scott Shaffer on 09/26/2013 - 12:11 pm.

      Here’s why.

      Maybe it would be worth it to build a more robust ability-to-repay profile that accounts for all the things you mentioned, but there are at least three problems to compiling and relying on this information. First, it would mean the lenders need to spend more money to discover and verify the data. Second, borrowers might be scared off by the further invasion of their privacy. Third, some of the measures could be bypassed by a borrower choosing to temporarily cancel their phone, satellite, internet, and buy life insurance for a few months (health insurance and auto insurance are mandatory, thank goodness). After they got the loan, the borrowers could resume their irresponsible financial habits, and all the investigation and extra diligence would have been for naught.

      None of these downsides applies to including transportation costs in the underwriting process. The lender already has the information about the commute, the borrower won’t feel violated, and I don’t see how the borrower could game the system by getting a nearby job just for the purpose of qualifying for a mortgage, and then moving to a remote workplace. Jobs (and commutes) are more permanent than that, if not everlasting.

    • Submitted by Jeff Klein on 09/26/2013 - 12:14 pm.

      MIssing the point

      I don’t know why this is complicated, or why people think the inherent transportation costs of different locales are somehow like an optional expense like a cell phone or cable. As you point out, child care expenses are looked at by the VA, and including untilites is fine with me. Auto insurance is part of transportation costs, life insurance is optional, and I’m fine with including healh insurance for the sake of consistency.

Leave a Reply