The True Cost of Driving and Travel Behavior

Steven Polzin's picture

Over the past few years a variety of documents ranging from contemporary media to more serious research efforts have addressed the cost of auto ownership and use.  These estimates are often used to address two important transportation issues, the household benefits of using transit in lieu of auto ownership and/or the consideration of household location decisions in the context of the total cost of housing and transportation.  Two often referenced sources of research on these issues are the Center for Neighborhood Technology's (CNT) initiatives in developing a housing and transportation affordability Index and the AAA updates on auto operating cost estimates[i].  These initiatives are noble efforts in our march toward holistic thinking about transportation and transportation policy.  Transportation costs – both financial and others – are important considerations in both household decisions and infrastructure planning and investment policy decisions. 

While estimating vehicle ownership and use costs can be of value in making decision makers – including individual travelers – aware of transportation costs that might otherwise not be transparent or accurately perceived, there are some risks associated with failing to more carefully explore the various estimates of transportation costs.  Citing something a lot does not automatically make it true or the most appropriate piece of information to use. 

The CNT transportation cost estimates are based on a model that produces household transportation costs for major metro areas in the US with estimated values typically ranging from just over $10,000 to in excess of $15,000 for personal vehicle ownership costs per household per year.   Alternatively, the Bureau of Labor Statistics Consumer Expenditure Survey estimates national average household spending on transportation to be under $8,000 per year.  And as anyone knows from driving down a residential street in America and in talking to their friends and neighbors, the vehicle transportation spending varies dramatically based on travel levels and consumer tastes and priorities as it relates to vehicle investment and use.  The AAA produces their annual "Your Driving Costs"  which produces an annually updated cost of vehicle ownership based on a five year ownership period and 15,000 mile per year travel rate and produces an annual cost of $8,946 or $0.596 per mile for 2012[ii].  The IRS allowed $0.555 per mile for business use of personal vehicles in the second half of 2011.   

Using the newest Consumer Expenditure Survey (CES) data from 2010, the average US household is reported to spend $7,667 per year on all travel of which $7,184 per year is attributed to vehicle travel[iii].  Based on national household travel survey data from 2008 – 2009, the average household had 2.1 vehicles, each of which was driven approximately 9,460 miles annually[iv].  This data produces an estimated per mile vehicle ownership and operating cost of approximately $0.36.  If divided by an average occupancy of 1.67 it results in an approximate $0.22 per passenger mile expenditure. 

Figure 1 shows the spending trend on vehicle travel by income quintile from Consumer Expenditure Survey data.  Costs aren't particularly high and are not growing – actually shrinking if adjusted for inflation - in spite of generally rising fuel and vehicle costs[v]. 

The disparity between various high estimates of auto ownership and operating costs and the measured average captured by the Consumer Expenditure Survey partially lies in the very meaningful differences between new car ownership and the reality that much of America isn't driving new cars with high depreciation levels.  Average vehicle retention has increased to 71 months meaning that households are retaining vehicles nearly 6 years, a trend attributed to both longer financing terms and longer warrantees and improved vehicle reliability.  The average age of autos has reached a record high 10.8 years as of July 2011.[vi]  In the U.S. light vehicles were noted in a 2006 study to be nearly 18 years old with approximately 170,000 miles on them when they were scrapped on average[vii].  Depreciation costs become insignificant at some point. 

It's easy to see how ownership cost get quite high with luxury and near luxury vehicle leases running above $500 per month plus equity investment, insurance, gas, parking, etc. and it's not hard to imagine cost per vehicle at or above the kinds of numbers cited by CNT or AAA.  On the other hand, the young person or low income worker who is driving aunt Martha's hand-me-down 14-year-old Ford or Toyota that they purchased for a few hundred dollars, maintain in their driveway, and drive with minimal insurance if any (various estimates suggest that as many as 30% of vehicles do not have automobile insurance) might be defining the other end of the household vehicle travel expenditure range[viii].  

Unless one chooses not to believe the Consumer Expenditure Survey data, Middle America has managed to control personal vehicle expenditure costs making personal mobility surprisingly inexpensive.  While this ignores the indirect costs and perhaps understates other costs (arguably we are under investing in transportation infrastructure and recent evidence suggests we've been depreciating our vehicle fleet asset as well) and there seems to be an increasing trend to use general revenue streams such as impact fees and property taxes to support transportation infrastructure, thus leaving some costs out of vehicle ownership and use cost estimates.  But  consumers are likely to make economic decisions based on the direct costs that they bear. 

So what does this all mean in terms of transportation policy and planning?  Clearly, many American consumers of travel are not basing their travel decisions on vehicle ownership and use costs that are nearly as high as those referenced in various resources and research reports- several of which reflect an advocacy perspective.  Take for example somebody seriously thinking about using transit and perhaps giving up a household vehicle.  In all probability they would be giving up a lower cost/value vehicle – perhaps one that was fully depreciated or needed to be replaced.  There would be some savings in operating cost but a significant share of the household mileage would likely be shifted to the remaining household vehicles with increases in their operating costs and depreciation rate.  Persons sensitive to travel costs who might be considering this decision are likely to be individuals in low to moderate income categories where their household spending on vehicle travel might be below or well below the mean levels of the total population.  Interestingly, consumer expenditure data also show that even zero-vehicle households have significant spending on vehicle travel.  Giving up or not purchasing a household vehicle is more likely to result in the use of rental and/or borrowed vehicles or providing gas money to friends and family for providing trips.  Thus, presumptions about total savings from relinquishing a vehicle need to reflect the full reality of how travelers are behaving.  

Data indicate that consumers have made decisions to reduce auto ownership – or the economy has made it for them.  Auto ownership levels have declined.  While stories about home foreclosures are regular headlines, less noticed is the reduction in personal vehicle ownership from approximately 236 to 230 million registered vehicles nationally between 2008 and 2010[ix].

Assumptions about travel cost savings associated with various residential location choices also require review relative to other sources of household spending on travel.  Again, there can be large variations in context.  For example, 27.2% of US households have no workers in them – thus, their household travel costs can be more controlled by their own choice decisions regarding trip destinations for their travel.  Another few percent (and growing) have work-at-home members of the labor force.  Unfortunately, fixed costs associated with relocating a household particularly for home owning households, make optimization of household location with respect to employment or travel generally an unpromising proposition in an era of job mobility, multi-worker households, upside down home-equity, and significant home purchase/sale transactions costs.  However, recent data suggest a decline in homeownership – while a consequence of the economic conditions, this perhaps provides an opportunity to enable more households to more easily minimize commuting costs[x]. 

While pointing out potential auto operating savings might be an appropriate strategy for an advocacy entity, serious policy deliberations are misinformed if they don't use a far more nuanced and empirically based set of data on household travel expenditures and travel behavior as well as recognizing the full costs of accommodating the changes in behavior, be they increased transit demand and the corresponding costs or changes in population and employment location preferences.    

Consumers are generally rational creatures making what in their eyes are sound decisions on spending.  It is important that planners and policy makers appreciate the revealed behavior of various segments of the population when they make their location, vehicle ownership, and mode choice decisions.   Travel costs are a factor in vehicle ownership, mode choice and household location decision-making, but perhaps these costs are not well represented by the savings potential that some analyses are implying or contending.     

[iv] CUTR analysis of NHTS data.

[v] CUTR processing of CES data – a sample size of approximately 7500 households in recent years.

[vii] Vehicle Survivability and Travel Mileage Schedules, DOT HS 809 952, January 2006. 


[ix] Highway Statistics Series, Table VM-1, Light Vehicles excluding motorcycles.

[x]  Home ownership dropped from 68 to 62 percent from 2011 to 2012, according to an April 2012 Gallup Survey.


Steven Polzin is the director of mobility policy research at the Center for Urban Transportation Research at the University of South Florida.



Michael Lewyn's picture

Transportation is expensive for low-income drivers

If I understand Prof. Polzin correctly, he seems to be suggesting that driving is not a major inconvenience for low-income drivers, because used and older cars are not as expensive as new cars.

But according to the Bureau of Transportation Statistics, the median "working poor" driver spends 21 percent of their income on transportation.

That tells me that even with used or older cars, low-income drivers do suffer quite a bit financially from driving.

Steven Polzin's picture

Transportation is expensive for low-income drivers

I did not suggest that "driving is not a major inconvenience for low-income drivers" I simply suggested that forgoing driving will not save them the amount of money many have reported. Driving is a major expense for low income persons. It is also a major factor in their ability to thrive economically in many environments.

Michael Lewyn's picture

Point taken

In that case, we agree more than I thought at first glance.

Sanger-Galton urban planning

Add to what I say above, the point that any household with children, and both parents working, is absolutely unable to afford to locate efficiently to TWO jobs, and schools, and shopping, and other amenities required by them; and walk, cycle and use public transport for all of their requirements.

The urban elites who promote "car-free living" are little more than social engineers engaging in a little de facto eugenics. How dare these proles irresponsibly introduce more CO2 emitting climate criminals onto the planet? We won't introduce a "one child" policy or compulsory sterilisations, we will just use urban planning to make it too expensive for anyone but multimillionaires to breed. Margaret Sanger and Francis Galton must be smiling down upon these people.

Sprawl = Parents' Paradise

I have seen people who live in Wodehouse's parents' paradise of suburban sprawl. They need two incomes to get by; they leave the house at 7AM to start their long commutes; they phone the spouse when they get to work to decide who will pick up the children at day care at 7PM and who will pick up dinner at McDonalds.

As a bonus, they get to spend hours per day fighting traffic on the freeway, and they emit enough CO2 to leave a less livable world for their children.

Charles Siegel

"Options" and out-of-touch urban elites

You never actually engage with my point.

What were these people's OPTIONS?

In free market, low land cost cities, it is FAR CHEAPER to "move closer to any one job".

In your utopian, planned cities where crack shacks cost $1,000,000 (refer "Crack Shack or Mansion") it is prohibitively expensive to live in the LEAST desirable/efficient location (as per your example) and THE COST OF THE VARIOUS OPTIONS ("housing plus transport") GOES UP FROM THERE.

Your assumptions are simply absurd: you make these people out to be stupid or negligent or "lovers of automobiles", when they are mostly victims of the circumstances that utopian planning has imposed on them. Do you actually ever talk to such people and discuss real life options with them? Oh, sure, they should save on transport costs by living in a condo with 20 square foot per household member, that still costs several times more than their sand suburb home, and that has nowhere safe for their kids outside the home, and "chained" trips to school, shopping and work that will involve them in the region's worst congestion. Oh, they should use transit for all these trips. Yeah, right. Your type are simply the covert inheritors of the Sanger/Galton tradition. Why don't you just come right out with it and advocate "one child" policies and compulsory sterilisation?

One of the most Orwellian arguments of all time, and one hears this all the time these days, is that the planners are "increasing housing choice".

In the process, they have caused a fringe McMansion to rise in price from $150,000 to $450,000 (NONE of that being in extra building quality or size) and condos near centrally located office jobs, from $150,000 to $1,000,000.

How "Smart" is THAT? This is "increasing housing choice". Yeah, on the planners planet Orwell.

Meanwhile, in any affordable, "sprawling" city that smart growthers love to hate, anyone changing their job to anywhere in that city, will have a high chance of being able to afford to move closer to their new job. The lower the urban land rent curve is from fringe to CBD, and the more dispersed employment is, the LESS "the cost of land" prevents efficiency-gaining location decisions.

Consider the high costs of suburban sprawl

If you're concerned about high costs, Wodehouse, consider that the half of America that lives an automotive lifestyle and that tends to live in the suburbs consumes 90% of transit dollars, while the half that doesn't drive and that tends to live in an urban center only gets 10%.

Does that sound even remotely fair? What if those spending percentages were reversed? I don't think you'd like that at all.

America, meanwhile, spends some $300 billion a year on road construction and maintenance, not to mention $100 billion a year on oil wars.

Do you think there are things that we can spend our money on that are just a tad bit more important than your precious suburbs?

You know? Like education for our kids? Health care for our sick and elderly? Mass transit that helps the working poor become productive? Emergency services that keep us all safe?

As for housing costs in the urban center, building more housing is the only way to stabilize prices. Not building housing for all these decades is what caused the shortage and subsequent high prices in the first place.

You call THOSE, "facts"?

The costs of automobility that ARE covered by drivers include the purchase, depreciation, insurance, fuel, repairs, maintenance and licensing of their own vehicle. Drivers also pay petrol taxes that contribute to the cost of roads.

Seeing drivers and their families are an overwhelming majority of members of society, there is quite a good "fit" between the externalities they impose, and those who suffer from those externalities.

But the costs of those externalities even when compiled by the most anti-automobile of analysts, are never much more than 15% of the costs that ARE explicitly covered by drivers. But what proportion of the total costs of transit, do riders cover? Even the benighted Weyrich and Lind admitted in their "myth busting" papers that many transit systems cover only 10% of their cost in fare revenue i.e. the "subsidy" is 90%, versus 15% max. in the case of car drivers. (Maybe in the most efficient rare cases of transit systems, as much as 50% of the cost is covered)

This is a "level playing field"? In what fantasy land?

Transit has externalities too, such as community severence and accidents. Why are railway lines so strictly fenced? Ans: because rail is b-----y dangerous. And why should the costs of grade separation and level crossings be allocated to something other than rail as they frequently are?

Transit's energy use is mostly now so inefficient per passenger mile, that the argument about military protection of resource availability is hardly an argument in favour of transit. It is also noticeable that the extraction and utilisation of the evidently more and more abundant energy resources in North America itself mysteriously seems to be opposed by the very same people as oppose automobiles, freedom, and American exceptionalism.

But I am the first to agree that the US bears far too much of the cost of policing the world, and that the ingrate Europeans, for a start, should be left to look after themselves and stop their bludging on US Defence efforts. Ron Paul for President.....!!!!!!!

"Time to exercise" is what any individual wants to make it. The freer the urban economy, the lower the cost of urban land, the more dispersed employment is, and the more affordably any individual can choose his location relative to work and amenities.

"Vehicle Ownership and Income Growth, Worldwide: 1960-2030" by Joyce Dargay, Dermot Gately and Martin Sommer (2007) points out that Germany and the UK are outliers among the European nations due to punitive impositions on drivers and car owners (far in excess of anything necessary to correct for externalities and the cost of roads). There is a very interesting "S-shaped" relationship between incomes and automobile ownership that holds very constant for almost all nations in history. France and Italy, for example, appear to be on track to achieve vehicle ownership "saturation" at a level close to the USA. The same goes for Canada, Australia, and New Zealand.

Developing countries are tracking along the same "S" curve; they are just still closer to the bottom of it.

The fact that such a high proportion of people in Germany still actually drive even with punitive costs something like treble the necessary level to correct for unpriced externalities, is evidence of very large "consumer surplus" for "automobility"; this is prima facie evidence for a very high economic value of automobility.

No such "consumer surplus" or prima facie evidence for a very high economic value for transit, exists.

The former USSR did not fail "in spite of the advantage" of having everyone forced to live in apartments and catch trains. This was an inherent part of "the reasons why" their economy was so inefficient.

Good luck trying to make "housing" affordable by building taller and taller apartment buildings. All the evidence actually points to "building OUT" as being far superior to "building UP", for "costs of space" (for households and businesses) AND for economic productivity. Karl Marx and Frank Lloyd Wright both pointed out the correlation between the "rentier" class capturing the major share of increases in the fruits of labour, and the restriction of ownership of land to small "urban footprints". Marx's "solution" was of course the nationalisation of land. Good luck with that, too.

Frank Lloyd Wright's solution was automobile based, low density, development on rural land, the supply of which was so abundant as to guarantee genuine competition between land vendors and maximum democratisation of land ownership. To this day, his genius on this point has not been recognised.

We live in the Reality-Based Community.

But the costs of those externalities even when compiled by the most anti-automobile of analysts, are never much more than 15% of the costs that ARE explicitly covered by drivers.

I'd ask for reality-based evidence, but I know that won't be forthcoming. So I simply call BS.

Wodehouse has scuttled back, so our BS meters will be ringing.



"Forgoing Driving" Will NOT Save Them Money - EXACTLY

Too busy to take much part here lately, but what Steve is saying is very important.

"......forgoing driving will not save them the amount of money many have reported......"

In fact, the urban land price curve always results in the price of "location" rising faster than the price of "travel". The only exception is for people who trade off a LOT of living space. The Contrarian Planner above may have been exaggerating about the $785,000 condos, but his point was totally valid. And in the bubble markets, his $785,000 is probably NOT even much of an exaggeration.

Steve, a focus on "actual" costs of HOUSING is just as wrong as a focus on "actual" costs of car travel. (As you say, cheaper-than average travel is easily obtainable). But there is also a major difference between the "actual" costs of housing, and WHAT THE COSTS WOULD BE FOR SOMEONE BUYING A GIVEN PROPERTY TODAY.

Of course the "drive to qualify" suburbs had the most foreclosures, because the residents were disproportionally people who had bought their first homes AFTER the prices started inflating. AFTER the prices started inflating, there was NO WAY any of these people could afford to buy in an EFFICIENT location. The "drive to qualify" location was barely affordable, in fact "unaffordable" and risky for most.

But for people like Todd Litman and Charles Seigel to say that the buyers of these suburban homes "should have bought a home somewhere where they could have saved on transport costs" is what I call Marie Antoinette Planning Syndrome. It was unwise to buy a $450,000 sand suburb house and have to spend $10,000 a year driving, but it was NOT the "answer" as Litman et al allege, to buy a $1,000,000 house or a $785,000 condo in an efficient location - and these literally ARE the prices in a bubble situation. (See "Crack Shack or Mansion" online).

In a non bubble market like in any of the 200 odd cities in the USA that the Demographia Reports show to have had low and stable housing prices, the "choices" are more like, between a $150,000 fringe McMansion and a $150,000 house in a mature, efficiently located suburb; and the condos might also be $150,000 or even less. Utopian planners driving up the urban land price curve across its entire length and talking about "increasing choice" are talking out of the wrong orifice.

Michael Lewyn might be "getting it" about the above reality too, going by his comments. There is nothing like being young and not yet a property owner, to give one a grasp of reality.

But, but, but

Todd Litman has told us ad nauseum how it is so much *cheaper* to buy a $785,000 condo in the city and give up your car. Could it really be possible that individuals actually understand their own trade-offs and do what is in their best interest? Perhaps some additional links to Todd's many "studies" would help :-)

In all seriousness, it would be far more fruitful to make driving more expensive to account for its full costs rather than try and convince everyone they are crazy and stupid to live cheaply in a suburban hell.

Michael Lewyn's picture

Now that's what I call attacking a straw man

My condo doesn't cost $785,000! (I think I'll let Mr. Litman speak for himself on that one...)

Instinctive bias towards transportation over housing costs

I'm not so sure consumers are generally rational. There seems to be systematic and measurable forms of irrationality in the marketplace, and many of these are at play in the difference between housing and transportation costs. We tend to weigh upfront cost more heavily than an equivalent cost that is spread out over a long period of time. When a household is looking to purchase a home, the housing costs must be confronted from the start while transportation costs are incurred only after the decision is made and not all at once. Even so, many people will choose to irrationally accept a mortgage that is higher and riskier than they can handle, which is why banks don't let them do this (at least now they don't). There is a system of checks and balances that helpfully constrains households in this way, but there is nobody even advising home-purchasers of the transportation costs they are assuming with the transaction.

Secondly, we tend to measure costs relative to what we see others paying around us. If I'm in a social circle in which everyone else spends a similar portion of their income on transportation, it is naturally going to fade into the background like breathing air. Unless I am confronted with viable alternatives in a clear way, I'm not going to give much consideration to this cost.

Thirdly, we tend to overvalue sunk costs. Most households looking to purchase a home already own at least one automobile, so the fixed price component, which is the most important, may not be factored into the transportation costs associated with a particular location. In reality, anyone can sell a vehicle at market rate at any time, so the fixed costs should be figured in. We just tend to avoid going back on decisions we have already made, even if it is more rational to do so.

Fourthly, we tend to weigh costs that are known more highly than the same cost that is unknown. A mortgage usually yields a fixed monthly rate that is given upfront usually with taxes and insurance neatly bundled into an escrow. On the other hand, it's hard to tell what transportation costs will be until you live in the home for a while, and they may change with new needs. And, of course, gas prices are volatile and basically unknowable. In human decision-making, fixed costs like mortgages are weighed more heavily than an equivalent, more diffuse, cost measured through probabilities (especially when its hard to tell what the probabilities are).

Based on these observations alone, I think we would expect to see households weighing housing costs more heavily than the transportation costs associated with the choice and making decisions in the marketplace that reflect this bias. We might see them moving further from the metropolitan core based on these human instincts in cost valuation. This is before even considering public policy implications.

Your point is well taken that the nationwide averages for transportation costs in the CNT index really should be more closely aligned with the CES averages. Hopefully, the model can be adjusted to get to a closer match, perhaps by considering used cars as you've mentioned. That being said, the power of the index lies not so much in the absolute values attributed to transportation but in the locational variations in transportation costs. I think these variations would still remain, and we even get a hint of this in the CES itself. Households in "Center City" pay less in transportation, both in absolute terms and as a percentage of income, than "Other Urban" or "Rural." It's not as stark as the HTA index, but there is some reality to it. This is the counterpoint to the traditional drive-til-you-qualify narrative, and I'm glad that CNT is starting to tell this story.

Bias Toward Initial Over Lifecycle Costs

Agreed, and it doesn't just apply to housing and transportation. Consumers generally give more weight to initial costs than to total lifecycle costs. This is why energy efficiency standards for appliances save consumers money: without these standards, consumers tend to buy appliances with lower initial costs and to ignore their higher long-term energy costs.

Charles Siegel

Why Are Vehicle Travel Costs Going Down?

"Figure 1 shows the spending trend on vehicle travel by income quintile from Consumer Expenditure Survey data. Costs aren’t particularly high and are not growing – actually shrinking if adjusted for inflation - in spite of generally rising fuel and vehicle costs."

The figure shows that costs have declined for low and middle income groups since 2007 or 2008. There are many causes of this decline, but two obvious ones are:

-- The housing bust his the most remote suburbs the hardest. In the worst sprawl areas, where people drive the most, many houses have been foreclosed and remain vacant. This seems to undermine the claim that sprawl does not impose high transportation costs.

-- Unemployment has gone up. People drive much less when they stop commuting to work - but they will have to drive more when they get jobs again.

Charles Siegel

Steven Polzin's picture

housing bust and sprawl

While I have not explicitly researched the distribution of home foreclosures as a function of the transportation costs of residents, I would caution analysts to more fully explore the nature of the housing foreclosure trend before jumping to the assumption that transportation costs were a significant contributor to geographically differential rates of foreclosure. Foreclosures were more prominent in homes purchased more recently relative to the housing crash. These new home purchasers were more often highly leveraged, had little equity in their home, and in many cases younger workers with less job seniority and more susceptible to layoffs. In addition, in fringe areas that had been growing there was a high concentration of homes all purchased recently. Thus, new growth areas were more susceptible to both foreclosures and the cascading effect of home depreciation spreading based on nearby foreclosed properties.

In a new suburb a young financially extended family may lose their job, have no equity in the house and quickly lose their house. Its depreciated value is soon reflected in adjacent appraisals cascading the stress throughout relatively fragile neighborhoods. On the other hand in established neighborhoods only a relatively small share of the homes changed hands near the peak of the building bubble. Thus, many of those homeowners had far more equity in their home and perhaps more job seniority and security enabling them to whether a housing downturn. In addition, the diversity of home ages and types and the less frequent occurrence of foreclosed properties will control the pace at which home value depreciation will cascade through the neighborhood.

If commuting cost was as big a contributor to suburban fringe foreclosure rates then one would have expected downtown condominiums to weather the housing bubble. In many locations like Florida large clusters of new downtown residential properties suffered the same rapid depreciation as did suburban fringe areas. The concentration of new units seemed to be more critical than the location.

I would encourage folks to explore more complex sets of phenomenon that have resulted in variable home depreciation rates across geography. I offer the comments above as a potential hypothesis worth exploring. Similarly, one would need to disentangle the impact of investment purchasing of homes (reported to be well over 20% during the peak of the bubble in many locations) from purchases associated with homeowner/resident decisions. Things are most often more complex than the superficial assumption that geography relative to a downtown is a dominant determinant of travel costs and economic condition of a given area.

If anyone has done any research on this I would like to see it.

Re: Housing Bust and Sprawl

You misunderstand my point, which I may not have made clearly enough.

My point was that foreclosures in sprawl-suburbia contributed to the decline in driving, not that high transportation costs in sprawl-suburbia caused foreclosures.

At the top of my comment (above) I quote the statement that I was responding to.

Charles Siegel

understated costs

while you admit that your analysis may understate some travel costs, i think perhaps you underestimate the indirect costs associated with auto travel. You mention infrastructure investment but do not mention the cost of "free" surface parking for each vehicle in multiple locations throughout the day. Far more hidden than the payments spread over time are the higher prices paid for everything else by everyone (even transit users)in order to provide easy access and free parking.

Todd Litman's picture

Vehicle Costs and Savigns

Steven, I agree with your basic point, that often-used vehicle cost estimates do not necessarily reflect the actual costs facing specific categories of motorists, and therefore do not indicate the actual savings from reductions in vehicle ownership and use. In particular, often-cited estimates published by automobile associations reflect newish cars (i.e., the first six years of a vehicle's lifespan) with full insurance coverage, which tends to exaggerate depreciation and insurance costs and underestimates repair costs. Similarly, average annual vehicle travel and expenditure data do not reflect those factors for lower-income households and second cars. These issues are discussed in detail in the "Vehicle Costs" chapter of my report "Transportation Cost and Benefit Analysis" ( ).

However, the higher vehicle ownership and travel required in automobile-dependent, sprawled communities certainly does impose many direct and indirect costs, or put more positively, living in a more accessible, multi-modal community (called smart growth, transit-oriented development or new urbanism) can provide a variety of direct user savings and benefits, including vehicle cost savings, parking cost savings, reduced accident risk, and improved health. The additional costs of automobile dependency are particularly burdensome to lower-income households, which spend a greater portion of their income on transport, and often drive older, less efficient and reliable vehicles, and so are more likely to experience financial stresses when fuel prices spike, or their vehicle fails or crashes.

When we owned a car, which we purchased new, maintained carefully and drove low annual miles, we still spent about $1,000 annually on major repairs (electrical system failure, new tires required, muffler failure, a minor crash, etc.). Similarly, a fuel price spike can increase the cost of driving by hundreds or thousands of dollars a year, as indicated in the ABOGO Model ( ). For many lower-income households these present critical financial problems.

From a household wealth generation perspective it is far better to choose higher-housing-lower-vehicle-costs than lower-housing-higher-vehicle-costs when purchasing a home, since vehicles tend to depreciate while housing tends to appreciate over the long run. As a result, after a decade a household that pays a $25,000 annual mortgage and spend $5,000 on vehicle expenses gains about $100,000 in equity compared with a household that pays a $15,000 annual mortgage and $15,000 on vehicle expenses.

In conclusion, although the details may vary depending on specific circumstances, the basic principle that automobile dependency significantly increases residents' transport costs, and improving transport options can provide significant user savings and benefits is valid, and these benefits are likely to increase in the future due to aging population and rising future fuel prices.

For information see:

Yingling Fan and Arthur Huang (2011), How Affordable is Transportation? An Accessibility-Based Evaluation, CTS Report 11-12, Transitway Impacts Research Program, Center for Transportation Studies (; at

Housing + Transportation Affordability Index (

Todd Litman (2009), "Where We Want To Be: Home Location Preferences And Their Implications For Smart Growth," Victoria Transport Policy Institute (; at

Todd Litman (2009), "Memo From Future Self: Hope For The Best But Prepare For the Worst," Planetizen (

Todd Litman (2010), "Affordable-Accessible Housing In A Dynamic City: Why and How To Support Development of More Affordable Housing In Accessible Locations," Victoria Transport Policy Institute (; at

Asha Weinstein Agrawal, et al. (2011), Getting Around When You’re Just Getting By: The Travel Behavior and Transportation Expenditures of Low-Income Adults, Report 10-02, Mineta Transportation Institute (; at

Todd Alexander Litman
Victoria Transport Policy Institute
"Efficiency - Equity - Clarity"

Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.
Planetizen Courses image ad

Looking for the perfect last-minute gift?

A one-year subscription to Planetizen Courses is the gift that will continue giving all year long.
Wood necklace with city map

City Necklaces

These sweet pendants are engraved on a cedar charm with a mini map of selected cities. The perfect gift for friends and family or yourself!