(Tom Rubin’s comments are identified by TAR: & END TAR)
Direct subsidies and costly externalities of motor vehicle use include: police,
fire and ambulance services, road construction and maintenance; property taxes lost
from land cleared for highways; subsidized parking; air, water and land pollution;
noise and vibration damage to structures; health damage from noise and pollution;
global climate change; petroleum supply line policing and security; petroleum production
subsidies; trade and infrastructure deficit; sprawl and loss of transportation options,
uncompensated auto accident costs; and congestion.
TAR: Well, a most interesting list. Let's first start with an overview, then go
through them one-by-one.
This list is a mixture of what is commonly known, in my business, as direct and indirect,
or "hard" and "soft" costs, with indirect/soft costs also called "social" costs.
"Hard costs" are the direct and most obvious, and easiest to tie to the mode, such
as road construction and maintenance. "Soft costs" are usually, but not always,
the same as indirect costs, such as "sprawl and loss of transportation options" from
above, and, by their nature, are far more difficult to measure, quantify, and express
in monetary terms with precision – or even to agree what they are.
The very obvious first problem with the list above is that is it totally "one-sided;"
what is presented is a list of costs – without any countervailing list of benefits.
In the hard cost analysis above, I measured subsidies by comparing expenditures of
funds by the public sector against the revenues that were directly generated by the
users of such services by the very act of using the service – no pay, no play. If
we are going to measure the "soft" costs, or social costs, or what other concept
is to be applied, it is vital that both the costs AND THE BENEFITS be considered;
failure to do so results in the rejection of the analysis as incomplete and inappropriate
on its face.
Failing the adoption of this preferred methodology for such analysis, a useful but
less desirable alternative is to set forth the "full" list of "soft" costs, including
ALL applicable ones, when comparing different modes.
For example, if the comparison is driving vs. transit for urban areas, the list above
appears to be rather comprehensive of the list of soft costs where it appears that
transit – or, alternatively, reduction of travel through land use changes and other
means – is represented to be superior.
For such an analysis to begin to be useful, there must also be discussion of such
goals, objectives, down to metrics which are important to individual travelers and
society where the auto could have an advantage.
For example, let us consider travel time and accessibility of destinations. With
few exceptions in the U.S., travel by transit is at a slower speed than driving and
also generally requires far more access time. (Yes, there are exceptions to this
and I am one of the people who has actually gone to a great deal of trouble to buy
homes where transit worked very well for me.) From detailed studies of census data
for the Bay Area (2006 ACS), the average drive time to work (drive alone and carpool)
was 24.6 minutes and, for transit, it was 42.2 minutes. For the round trip, that's
a difference of 35.2 minutes a day -- or about 2.44% of the 24 hours in a weekday,
even more of the time not spent sleeping, and even more of the controllable time
after adjustment for eating and normal activities like personal hygiene.
Also, because travel via transit is almost always at a much slower speed than travel
via auto, particularly when the access time is considered, travel by transit tends
to be very limiting, compared to auto travel, in both distance that can be covered
and destinations within any given travel time. When the common "30-minute" access
time test is done, it common for the "simple" "circle" analysis showing how far you
can go by car and by transit (which does not consider the problems of lesser-accessibility
of nearby sites by transit because of lack of nearby transit stops) for transit to
be under one-quarter of the size of that for auto users, and the more detailed site-specific
analyses which do consider transit accessibility can show transit doing even poorer,
compared to auto use.
In my opinion, these are factors that are extremely important to consider in any
competitive comparison of auto vs. transit modes and any analysis that does not do
so is deficient on its face – and these are most definitely not the only such factors
where the auto will generally have a significant advantage over transit.
Getting back to those factors listed above, we have, for example, among the "costly
externalities," "air, water and land pollution." I most certainly agree that is
a major concern and it is very valid to include this on this list, if such an analysis
is to be fairly and competently conducted. I also know that the current situation
in regard to, for example, air quality emissions, is far superior to what it was
even a few years ago and that emissions from rubber tire vehicles on the roads are
continuing to drop every year. Some of the former problems – such as lead – have
been almost completely eliminated, and every other one continues to get better in
almost every way in almost every community almost every year.
Which is certainly not to say that we have reached the point where no further action
is needed, but let's realize the progress that has been achieved – and where we are
reaching the point where the cleanest auto's are about as good as it is possible
to get them, to the point where spending any more money to improve is wasteful, compared
to what can be achieved from other types of expenditures. In particular, for the
same amount of money applied to automotive emission reduction, you get FAR more band
for the buck by taking junkers off the road than by trying to get a vehicle that
is already 98% cleaner than base model (early 1970's) vehicles to 99% cleaner.
Also, when one is making such analyses, it is vital to review the alternatives, and
what has gone before. In this case, what was the condition in America prior to the
automobile and the truck?
The answer is, quite literally, pretty close to being a cesspool. Prior to internal
combustion power plants providing mobility in American cities, we had animal power
– chiefly horses, mules, and oxen. The bodily waste of these animals was a massive
daily insult to the quality of life. Moreover, when the animals died, the common
way of disposal was simply to leave them where they expired – in many cases, FOB
middle of the street, where they would remain, generally for hours, sometimes longer,
until the carcasses were hauled away.
Any rational comparison of the pre- and post-automotive era pollution/public health
status cannot fail to note that, even when rubber tire vehicles were at their dirtiest,
there was a massive improvement over the previous case, and the qualitative improvements
in vehicles since have already eliminated the vast majority of the problems of the
motorized vehicles from the bad old days, with further improvements already guaranteed
as older vehicles are taken out of service and replaces with even more superior technologies.
Further, it is important to note that the automobile and the American road system
have produced an improvement in American mobility by orders of magnitude – something
that is totally lacking in the analysis methodology presented above.
Let us consider another criterion, "health damage from noise and pollution." Yes,
there is no doubt that the internal combustion road vehicle, as well as other modes,
has damaging impacts on the human body and on other life. Again, we see a realization
of this, which has resulted in changes that have already produced huge reductions
in health-damaging aspects with continued improvements as time goes by.
Now let us look at the benefit side of the picture. The mobility provided by the
automobile has, in and of itself, very significantly improved the provision of health
care in the U.S. In the old days, there was the family doctor – who, out of necessity,
was a generalist. With his (and, to the then very limited extent, hers) mobility
options so minor, the geographic area of the practice was very limited. Specialization
was simply not very possible for most physicians at the time because, in most cases,
there were not enough patients with the specific ailments that could reach him.
At the present time, I have specialized health care providers that I see on a regular
basis that are over 40 miles apart, and my particular case is in no way unusual.
Let us consider another very significant health care improvement that is solely due
to the auto and the road system – well-trained paramedics in well-equipped mobile
medical vans. Their ability to get to people with life-critical medical emergencies
within minutes of getting notified, and then rapidly and safety transport them to
emergency rooms, saves many thousands of lives every year.
While we are on safety, the annual highway death toll rate in the U.S. has been roughly
in the low 40,000's for several years; let's round up to 50,000, which would give
us a fatality rate of a bit over one out of every 6,000 people in the U.S. each year
(current U.S. population, a bit under 306 million).
At the height of passenger rail travel, the decades just before and after the turn
of the 20th century, annual railroad deaths were approximately 10,000 per year –
and, very arguably, the counting was far less comprehensive than it is now for road
fatalities. With about 76 million Americans back that, that's about one fatality
for every 7,600 people – which is pretty darn close to the number we have for auto
The difference of course, is that today's Americans are FAR more mobile by auto than
our great grandparents were, both locally and nationally. Back in the day, passenger
rail travel was of three main types: streetcars, what we now call "commuter rail,"
and inter-city; all were very dangerous in different ways and all offered far less
mobility than the average automobile user today.
So, when railroads ruled for American transportation, the chances you would die due
to something happening involving a train was about the same as it is in regard to
an auto today – but, with autos, there is greater mobility, that I would estimate
at one to two orders of magnitude, with all the benefits that comes with that.
To sum up this segment, if you want to try to compute the social costs, or externalities,
of auto use, well, go ahead if you want, but I will warn you that this is a very
difficult proposition to accomplish with any degree of precision and confidence.
To do so with any degree of fairness, the criteria must reflect not only the areas
where you think your favored modes will perform best against the ones you do not
favor, but also those that show where other modes have advantages.
If you do not also attempt to do a comparable job of calculation the social and externality
benefits of the auto and the road system and other modes, you will perhaps understand
why the only reason I will refer to it is to discuss why it is not worth considering.
Now let's get to the specific items on the list:
1. Police, fire and ambulance services -- I don't have a great deal of confidence
in the data on this particular road expenditure line item as to its completeness
– if someone else does, I'd be interested in reviewing it.
The FHWA stat's I mentioned above showed total nation road expenditures of $14.5
billion on "Highway Law Enforcement and Safety" for the 2006 reporting year, or about
9% of total road expenditures. This is split almost 50:50 between state and local
governments. My problem is, I haven't been able to get good answer from anyone at
FHWA as to what is included; if this supposed to be ALL emergency services for roads,
The FY09 Budget for California Highway Patrol, which is primarily, although not
exclusively, a roads safety service, is about $1.9 billion. There are about 12,000
CHP officers, or about 2% of the U.S. total number of police officers – which would
appear to indicate that the $14.5 billion above could be on the low side. That works
out to a bit over $700 for each U.S. licensed driver. This leads me to believe that
the FHWA may not include a real good accounting for city/county sworn officer patrol
and related work devoted to traffic safety et al and there may be other issues, as
However, the FHWA data does not include any fines or other revenues – which are
hard to get estimates on, but appear to be somewhere in the $10 billion range, nationally.
In many locations, this is a major revenue source for the "general" government functions
and is getting to be more so, but there are costs in running the courts and even
administering the processing of violators who just mail in a check.
In short, I'm having a hard time getting high quality data on the costs, and the
revenues – and, thus, the net taxpayer subsidy – of police, fire, and other emergency
services for roads. From the numbers I presented above, I'm not sure what the number
might be – any help?
2. Road construction and maintenance – as discussed above, there does not currently
appear to be any governmental "general fund" subsidies for roads construction and
maintenance as a whole; in fact, a stronger case can be made that road users are
subsidizing other governmental functions, notably public transit.
3. Property taxes lost from land cleared for highways – while not entirely invalid,
this is not properly presented, as it ignores the significant increase in property
taxes for property that has superior transportation access; I don't think there can
be any question that, overall, the construction of roads has significantly increased
the value of land in the U.S. and, therefore, the property taxes on such land. Also,
it must be considered that roads existed long before the invention of the automobile.
Interestingly, this is a variation of long-standing debates on the subsidies granted
to transcontinental railroads in the form of land along the right of way. Some argue
that these subsidies should be valued at their value after the rails were in place;
however, this is generally rejected in favor of costing these subsidies at the their
pre-railroad construction values – generally a small fraction of the "after" values
– because the increase in land value was, for the most part, CAUSED by the construction
of the rail line.
Transportation guideways can have impacts on land values in various ways to various
users. For a trucking terminal, being near a freeway on/off-ramp is very valuable;
for a homeowner, being next to a newly built freeway is generally a major negative
(in general, residential real estate values are optimized close to, but not TOO close
to, a freeway, all else equal). A heavy rail station in the basement of a high-rise
condo complex would generally positively impact the value of the units; a single-family
detached home located near an at-grade intersection without a station, where there
were a dozen or more trains each hour blowing 85 dBA horns, from 5:00 a.m. to midnight
each day, will generally have a negative impact.