[sustran] Re: Comparing Costs of Modes

[Todd Alexander Litman]

I appreciate feedback on my publications, but I encourage Mr. Oster to more 
carefully read my papers before criticizing them.

It is wrong to say that the cost value estimates in "Transportation Cost 
and Benefit Analysis" are "pulled from the air." They are based on 
literally hundreds of studies that were reviewed and summarized in the 
book. How the values are estimated is discussed in detail in each chapter. 
You are welcome to disagree with the approach I used, but it is unfair to 
claim that they are not based on careful and transparent analysis.

I used a different approach than Mr. Oster to calculate the relative risk 
of bicycling. His approach underestimates total cycling mileage and 
overestimates the health risks of increased cycling to responsible adults. 
As discussed in the chapter, bicycles have a unique risk profile: 
significant portion of bicycle fatalities involve child or inexperienced 
riders; a significant portion of cycling is ignored or undercounted in 
conventional travel surveys (most do not include recreational cycling, or 
cycling by children); shorter nonmotorized trips often substitute for 
longer motorized trips, and nonmotorized travel provides health benefits 
that offset crash risks. Studies of cycle commuters indicate that 
responsible adults who bicycle seem to have about the same health risk per 
trip as automobile travelers, which is why total per capita traffic 
fatalities tend to decline in cities with high rates of bicycle commuting 
(see additional discussion in the  report "Safe Travels" 
http://www.vtpi.org/safetrav.pdf)

Mr. Oster misunderstands the concept of external crash costs, that is, the 
costs of crash risks and damages not borne by users. It is inaccurate to 
claim that all crash costs are internalized through insurance. If that were 
true than crashes would not be a problem since victims would be fully 
compensated. In practice there are significant uncompensated risks and 
damages, in fact there must be as an incentive for travelers to be 
cautious. This is discussed in the "Safety and Health Costs" chapter. 
Motorcycles impose relatively high external costs because they cause 
significant injuries and disabilities, a portion of which are not 
compensated by insurance, leaving hospitals to cover a portion of medical 
costs, government programs to cover a portion of disability costs, and 
families and businesses to cover a portion of lost productivity. All of 
these issues are well described in the literature and discussed in my book.

Similarly, Mr. Oster has misinterpreted motorcycle parking costs. While it 
may be true that  3 or 4 motorcycles *can* park in the place of one car, in 
practice, in the U.S. most motorcycles occupy one parking space, either 
because this is required, or due to motorcyclist preference. Only in large 
facilities where smaller, "motorcycle parking" spaces are provided do 
motorcycles use significantly less space to park. My analysis assumes that 
motorcycles use 25% less parking space, however, because motorcycles have 
lower load factors than automobiles, their parking costs per passenger-mile 
are about equal.

Similarly with congestion and roadway land values. Although in some 
jurisdictions motorcycles are allowed to share lanes, they seldom do in 
congested urban conditions. It is only possible when two motorcycles happen 
to be traveling side-by-side, and it is unsafe in stop-and-go conditions. 
While it is true that under certain conditions (narrow road, fast motorized 
traffic, slow cyclist) cyclists can cause traffic delay, those conditions 
are unusual, since they are unpleasant for cyclists, and cyclists are not 
allowed at all most urban Interstate highways were more than half of all 
traffic congestion occurs. See discussion in http://www.vtpi.org/nmt-tdm.pdf.

Mr. Oster is welcome to dismiss the costs of reduced diversity, barrier 
effect and negative land use impacts that result from more motorized 
travel, but I recommend reviewing the published literature on these 
subjects before calling them "bogus". Since I first included them in my 
analysis a decade ago they have become more widely recognized by planning 
professionals. See, for example,
David J. Forkenbrock and Glen E. Weisbrod, "Guidebook for Assessing the 
Social and Economic Effects of Transportation Projects," NCHRP Report 456, 
TRB, (www.trb.org), 2001; Louis Berger Inc., "Guidance for Estimating the 
Indirect Effects of Proposed Transportation Projects," NCHRP Report 403, 
Transportation Research Board (www.trb.org), 1998.

While these cost categories may not apply in every planning situation, they 
are often important to consider. For example, if a community is choosing 
between two options for addressing traffic congestion (e.g., widen highways 
or improve walking, cycling and transit service), I think most people would 
recognize that there are advantages to choosing the option that improves 
mobility for non-drivers, reduces barriers to nonmotorized travel, and 
encourages more compact development patterns. That is what these cost 
values are intended to represent. Mr. Oster may dismiss these issues but 
they are based on an extensive and growing body of literature and I'm 
pretty sure that they would be considered legitimate by most objective 
people who take the trouble to investigate them.


Best New Year wishes to all,
-Todd Litman






At 09:58 AM 12/30/2004 -0500, Daryl Oster wrote:
>Todd Litman,
>
>Your Transportation Cost Analysis (TCA) is a useful resource, and I agree
>that true cost comparisons must be made on a per passenger mile basis, I
>applaud your excellence and monumental efforts to do so.  Please do not
>interpret my suggestions for improvement in the negative.  I hope you
>consider additional categories for trains, planes, boats, and projected cost
>estimates for automated PRT, and ETT.  Thanks for your consideration of the
>following:
>
>There are several inequities readily apparent in your  TCA (table 4, average
>travel 1996 $/passenger mile ( http://www.vtpi.org/tca/tca.xls):
>*Internal crash cost of bikes, cars and motorcycles;
>*External crash cost of cars and motorcycles;
>*Internal and external Parking for motorcycle and cars;
>*Congestion of cars, motorcycles and bikes;
>*Land value of cars, motorcycles, and bikes;
>*Transport diversity
>*Barrier Effect of cars, motorcycles and bikes
>*Land use impacts
>*Water pollution of cars, motorcycles, bikes, and pedestrians
>*Waste of cars, motorcycles, bikes, and pedestrians
>It appears that many of the numbers were “pulled out of the air” to reward
>bikes and pedestrians, while punishing cars and especially motorcycles.  I
>will address each point as follows:
>
>
>*Internal crash cost of bikes relative to cars:
>http://www.bicyclinginfo.org/bc/index.htm shows that in 2001, there were 728
>bicycling fatalities and 45,000 bicycling injuries resulting from traffic
>crashes in the United States. While these numbers continue to decrease from
>year to year, bicyclist fatalities still account for 2 percent of all
>traffic fatalities as well as 2 percent of all traffic injuries.
>
>http://www.fhwa.dot.gov/environment/bikeped/study/index.htm shows that in
>2001 shows that bicycles accounted for 0.8% of the personal transportation
>trips.
>
>The Other bike trips - The 1995 National Personal Transportation Survey
>(NPTS) shows that bike to work trips comprise 8% of all bicycle trips.
>
>Average bicycle trip length ­ work - NPTS shows an average round trip
>distance for bike to
>work trips of 4.0 miles.
>
>Average bicycle trip length ­ other - NPTS shows an average round trip
>distance of 2.8 miles for all other bike trips.
>
>((4mile (0.08)) + (2.8mile(0.92))) = 2.9 miles average bicycle trip length.
>
>OAK RIDGE NATIONAL LABORATORY SUMMARY OF TRAVEL TRENDS 1995
>NATIONWIDE PERSONAL TRANSPORTATION SURVEY shows that the national average
>car trip length was 9.06miles.
>
>Therefore the safety of bikes relative to cars on a vehicle mile basis is:
>2%of fatalities  / (0.8% of trips * (2.9bike miles/9.06car miles)) = 7.8
>times the risk per vehicle mile for bikes relative to cars.  The cost of the
>internal risk of bikes relative to cars must therefore be adjusted
>accordingly.
>
>In tca.xls,  Todd Litman shows the internal crash cost of  cars to be 5
>cents per passenger mile; the internal crash cost for motorcycles is listed
>as 43.7 cents per passenger mile; and the internal crash cost for bicycles
>is listed as 5 cents per passenger mile ­ the same as cars.
>
>The factor of cars relative to motorcycles is 43.7 / 5 = 8.7 roughly
>corresponding to the internal crash death rate factor of motorcycles
>relative to cars.
>
>If Litman correctly applied this factor to motorcycles, why is it that he
>fails to apply it to bicycles?  If it were applied to bikes, it would
>increase the per mile cost of bikes to: 5 * 7.8 = 39cents per mile.
>
>Therefore the average per mile cost adjusted for this factor would show:
>Cars stay the same at 73.8cents/passenger mile, and bicycles would increase
>from 44.1cents to 78.1 cents per passenger mile from this factor alone.
>
>
>*External crash cost of cars and motorcycles;
>Litman claims the external crash costs for motorcycles are higher than cars,
>while insurance rates and physics prove the opposite.  Liability insurance
>is required for cars in most states, while for motorcycles it is usually not
>mandatory.  The liability insurance cost for cars is 3 to 4 times the rate
>for motorcycles, furthermore the liability insurance rate for bicycles is
>not zero, and when adjusted for mileage is often actually higher than cars.
>
>
>*Internal and external Parking for motorcycle and cars;
>Litman lists the parking internal and external costs for motorcycles as
>greater than or equal to cars, while listing bikes at less than 1/10 the
>rate of cars.  In actuality, one can park 3 or 4 motorcycles in the place of
>one car, thus the rate applied to motorcycles is not justified.
>
>*Congestion of cars, motorcycles and bikes;
>Litman incorrectly applies a higher rate to motorcycles than cars, while the
>rate for bikes is less by a factor of 18!  In fact many states allow
>motorcycles and bikes to split lanes, therefore the congestion factor for
>motorcycles should be less.  Bikes actually cause congestion at a higher
>rate than cars due to their slow speed compared with motorized traffic, they
>typically operate at less than 1/3rd the speed limit ­ causing much more
>congestion than their physical volume.  Pedestrians crossing streets also
>contribute much more to congestion than is reflected in the chart.  The time
>at traffic lights devited to pedestrian crossing is a major cause of
>congestion.   The per mile congestion rate must compare physical volume,
>time, speed, and flow issues.
>
>*Land value of cars, motorcycles, and bikes;
>Litman charges cars at 17 times greater rate than bikes.  Litman incorrectly
>charges motorcycles a higher rate than cars, when cars occupy 3 to 4 times
>more space. Bikes usually get very little use, so they have increased
>external costs for storage for a much higher percentage of the time.
>
>
>*Transport diversity.
>This is a totally bogus factor ­ without an merit ­ it appears just an
>invented way to unjustly penalize cars and motorcycles relative to walking,
>trains, busses, and bikes.
>
>*Barrier Effect of cars, motorcycles and bikes.
>Litman claims bikes at zero ­ the came as pedestrians, he claims cars at
>.007, motorcycles at .009, this again is incongruent with reality.
>Motorcycles should be rated between bikes and cars, and bikes should rate
>higher than pedestrians.  Is this just another reward for walking and bikes?
>It depends on the perspective, pedestrians and bikes could be as justifiable
>viewed as barriers to cars and motorcycles due to their slow speed.  Train
>tracks constitute huge barriers, what of this?
>
>*Land use impacts
>Again Litman assigns zeros to walking, bikes, trains and busses, while
>placing a large factor on cars, and even greater factor on motorcycles.
>This again is a matter of perspective ­ there is a lot of rail ROW that is
>exclusive ­ no other use possible, where cars, bikes, motorcycles, and even
>pedestrians use roads.  Motorcycles and bikes both are able to use narrow
>paths along with pedestrians. ­ once again the bias for trains, busses,
>bikes and walking is apparent in Litman’s work.
>
>*Water pollution of cars, motorcycles, bikes, and pedestrians
>Most water pollution is the result of erosion. Pavement for cars,
>motorcycles bikes, and pedestrians causes high velocity runoff that induces
>high erosion rates.  Another major contributor is foot and bike traffic that
>kills ground covering vegetation.  To assign zeros to pedestrians and bikes
>is unjustified.
>
>*Waste of cars, motorcycles, bikes, and pedestrians.
>Physical evidence shows that pedestrians and cyclists are more likely to
>litter than motorists.  Pedestrians and cyclists wear out their clothing and
>shoes at a faster rate than motorists.  Tires on bikes wear out 30 times
>faster than car tires.
>
>Some other notes / observations / suggestions:
>
>Litman also places the cost of vehicle ownership for motorcycles higher than
>cars 25.2cents/mile motorcycles, to 14.vcents per passenger mile.  While
>this may be true in North America (where per year use is low), it is not
>representative of the rest of the world.  Litman assigns an average
>occupancy of 1.00 to motorcycles; actual is closer to 1.2 as many ride
>double, especially those who tour and put on high miles.
>
>Also, if everyone used telecommuting, and all transportation were automated,
>would not all much of cargo impacts and costs be attributed to
>telecommuting?
>
>To correctly calculate the true impacts and costs of travel there are
>several additional things that must be considered, please see the attached
>spreadsheet that shows some of the biggest cost and impact reasons that show
>why cars are so popular.
>
>Happy holidays,
>
>
>Daryl Oster
>(c) 2004  all rights reserved.  ETT, et3, MoPod, "space travel on earth"
>e-tube, e-tubes,  and the logos thereof are trademarks and or service marks
>of et3.com Inc.  For licensing information contact:    et3 at et3.com ,
>www.et3.com  POB 1423, Crystal River FL 34423-1423  (352)257-1310
>
> > -----Original Message-----
> > From: Todd Alexander Litman [mailto:litman at vtpi.org]
> > Sent: Friday, December 24, 2004 9:47 AM
> > To: et3 at et3.com
> > Subject: RE: Comparing Costs of Modes
> >
> >
> > Please read my paper more carefully before you criticize it.
> >
> > If automobile insurance totally covered accidents society would be
> > indifferent to crashes and there would be no need for traffic safety
> > programs. But by its nature insurance can only cover a portion of total
> > costs (if crash damages were fully compensated some people, those who
> > place
> > a relatively low value on their own injuries, would have an incentive to
> > cause crashes and be injured). The costs in my analysis reflect
> > uncompensated crash costs.
>
>Todd, In actuality, what you describe is the one of the leading insurance
>frauds.  In fact, my wife and I were the victim of such an attempt to
>defraud our insurance company.  A driver intentionally swerved across 2
>lanes to sideswipe the front corner of our vehicle, and then claimed we ran
>into him causing great injury.  Fortunately I had a digital camera, and took
>several photos.  It was also fortunate that we had the same insurance
>company!  The insurance company attorney noticed that the injuries claimed
>were on the opposite side of the body from the crash! The court case was a
>slam dunk - he got zip, and his attorney was severely reprimanded.  This
>abuse runs up the rates for all of us - and is not a reflection of true
>risks and damages.
>
>
> >
> > Again, I recommend that you learn more about multi-modal economic
> > evaluation, which accounts for all relevant costs. That is the only way
> > you
> > can really justify a new mode.
> >
> >
> > Best holiday wishes,
> > -Todd Litman
> >
> > At 12:27 AM 12/24/2004 -0500, Daryl Oster wrote:
> >
> > > > --Original Message From: Todd Alexander Litman
> > [mailto:litman at vtpi.org]
> > > >
> > > > For more discussion see "Transportation Cost and Benefit Analysis"
> > > > (http://www.vtpi.org/tca) and the "Comparing Transit and Automobile
> > Costs"
> > > > section of "Evaluating Public Transit Benefits and Costs"
> > > > (http://www.vtpi.org/tranben.pdf).
> > >
> > >
> > >Todd,
> > >I scanned your work when it was announced a week or two ago, (it looks
> > like
> > >you have gone to a lot of effort); and it appeared to me that some of the
> > >costs you are claiming for cars are added in more than once, while some
> > of
> > >the other modes do not receive the same treatment.  One area is risks and
> > >accidents, this cost is covered in insurance cost, yet you add it again.
> > >For bikes, the death risk is greater - AND the cost is mostly paid by
> > auto
> > >insurance - not reflected in the graphs. I will be happy to discuss
> > further
> > >if you want to.
> > >
> > >
> > > > I think it is generally a mistake to criticize a particular mode as
> > being
> > > > inefficient or unsustainable. A better approach is to recognize that
> > > > nearly every mode can play a role in an efficient and sustainable
> > > > transportation system, including walking, cycling, public transit,
> > inter-,
> > > > city rail highways, and perhaps some new modes yet to be developed.
> > The
> > > > key is to determine which is most cost effective for a particular
> > > > situation, taking into account all benefits and costs.
> > >
> > >We definitely agree on this, and I applaud your efforts to move in this
> > >direction.
> > >
> > >
> > > > I cannot say how Evacuated Tube Transport costs compare with other
> > modes
> > > > because we lack operating examples. It would be interesting to perform
> > a
> > > > comprehensive analysis.
> > > > Best holiday wishes,
> > > > -Todd Litman
> > >
> > >
> > >We do have examples of travel in an evacuated environment 109 gigameters
> > per
> > >hour multiplied by 6.1B people - continuously 24-365 without fail. That
> > >amounts to 5,824,524,000,000,000,000,000 (5.8X10ee21) passenger
> > kilometers
> > >per year, and the energy use is immeasurably low.
> > >
> > >And there are plenty of examples of the costs required to approximate
> > those
> > >"perpetual motion" transportation conditions on earth's surface in tubes.
> > >We have performed comprehensive analysis, some is presented on
> > www.et3.com ,
> > >I invite any criticism you may offer.
> > >
> > >
> > >
> > >Daryl Oster
> > >(c) 2004  all rights reserved.  ETT, et3, MoPod, "space travel on earth"
> > >e-tube, e-tubes,  and the logos thereof are trademarks and or service
> > marks
> > >of et3.com Inc.  For licensing information contact:    et3 at et3.com ,
> > >www.et3.com  POB 1423, Crystal River FL 34423-1423  (352)257-1310>
> >
> >
> > Sincerely,
> > Todd Litman, Director
> > Victoria Transport Policy Institute
> > "Efficiency - Equity - Clarity"
> > 1250 Rudlin Street
> > Victoria, BC, V8V 3R7, Canada
> > Phone & Fax: 250-360-1560
> > Email: litman at vtpi.org
> > Website: http://www.vtpi.org
> >
> >
> >
> > --
> > This message has been scanned for viruses and
> > dangerous content by Netsignia Online, and is
> > believed to be clean.
> >
> >
>


Sincerely,
Todd Litman, Director
Victoria Transport Policy Institute
"Efficiency - Equity - Clarity"
1250 Rudlin Street
Victoria, BC, V8V 3R7, Canada
Phone & Fax: 250-360-1560
Email: litman at vtpi.org
Website: http://www.vtpi.org