« Intellipedia | Main | The Kerry Joke »
Should Californians walk more?
I just read that the external social cost of having another driver in California -- due to accidents and not even citing congestion -- ranges from $1,725 to $3,239 a year.
The number seems high to me (I couldn't spot any problem in the paper), but I learned a new argument for market failure.
An externality arises from the difference between the average and marginal costs of accidents. Each driver pays, on average, half the cost of an accident involving two people. But in Coasean terms, the "marginal product" of either driver was to cause the full accident (this seems to ignore that on some days, certain drivers will crash with someone else, virtually no matter what, but of course this does not cover every case either).
So should you go drive in Idaho instead? Or can the argument be flipped to create a comparable uninternalized benefit from driving? When you drive to someone or somewhere else, you will bring some consumer surplus or producer surplus to the other person. Or is the "complementarity of the match" more potent on the downside, with accidents?
That is all from Aaron Edlin and Pinar Karaca-Mandic, in the October 2006 Journal of Political Economy. The authors also claim that an optimal Pigouvian tax on driving could raise $66 billion in the state of California, more than the other state taxes put together.
Here is an earlier version of the paper. Guest blogger Eric Helland also covered the Edlin, Kraca-Mandic paper before it was published. He suggested toll roads were the solution.
Posted by Tyler Cowen on November 3, 2006 at 06:44 AM in Economics | Permalink
Comments
Biased argument. As congestion increases, speeds slow, and accidents become less costly. Thus Massachusetts has the highest accident rate in the country, but the lowest death rate from accidents. I would think it is unclear whether adding drivers has any externalities at all aside from congestion and delays.
Posted by: pj at Nov 3, 2006 7:13:57 AM
Yowza. What does it say that one Pigouvian tax could swamp all other taxes in CA? All the deadweight losses of other taxes would make substituting such a tax a massive efficiency improvement. Then add other recommended Pigouvian taxes, and you have more than enough money to optimally fund goverment AND make a massive wealth transfer back to the people of CA, all while cleaning up congestion, air pollution, etc. I understand there are tax administration issues, but with the kind of welfare at stake those issues become small. If they're right, they begin a radical critique of CA political economy, no? For that matter, the US.
Posted by: Jonathan at Nov 3, 2006 7:33:30 AM
the "marginal product" of either driver was to cause the full accident
No, the marginal product of one driver was to cause the full accident. Over time, that one driver will internalize the cost of his tendency to cause accidents by facing higher insurance costs. You only derive an externality if you assume that accidents are caused by happenstance rather than individual fault. As usual, externalities disappear when you introduce property rights and the corresponding duties - in this case, the duty to drive safely.
Posted by: eddie at Nov 3, 2006 8:31:05 AM
I was confused by the marginal cost vs. average cost calculation, and figured it was standard economics. Guess not. Saved the paper, anyways--traffic is an interesting subject, and the treatment is almost like something out of a physics textbook.
Surely the $66B Pigouvian tax must be a pipe dream though... elasticities and all.
Posted by: Chi at Nov 3, 2006 8:50:21 AM
I see that my point has already been considered by the authors:
A damage allocation system can provide adequate incentives for careful driving, but it will not provide people with adequate incentives at the margin of deciding how much to drive or whether to become a driver (Vickrey 1968; Green 1976; Shavell 1980; Cooter and Ulen 1988).
I won't look up those other papers and try to refute them, but I have to say I'm skeptical. The authors do make the argument directly themselves:
An average person pays the average accident cost r either by paying an insurance premium or by bearing accident risk. The accident externality from driving results because a driver increases traffic density and thereby increases accident risks and costs for other drivers. Although the increase in D from a single driver will affect r only minutely, when multiplied by all the drivers who must pay r, the effect could be substantial. The driver does not pay under any of the existing tort systems for exerting this externality.
An easy counter to this: assuming that increased density does mean increased risk of causing an accident or of being in an accident that is not your fault, drivers will choose to take trips along routes with known densities and thus known risks, and by doing so their insurance premiums and/or born accident risk will increase or decrease accordingly. The relevant question is not whether an additional driver on a route adds risk to everyone else in excess of their own costs - the relevant question is whether or not individual drivers know what the risks of driving on a route will be. And they do. It's no surprise to anyone that the same routes that were congested yesterday and congested last week are congested today.
Furthermore, it's hard to take seriously a paper that outlines a framework for accident frequency by saying:
Consider the risk that a given driver i faces. With speed held constant and under the assumption that drivers make independent decisions about where to drive, the chance that another driver is in the same location as i will be proportionate to the amount of driving that these other drivers do and inversely proportional to the amount of roadway L over which this driving is distributed. [emphasis added]
That seems a lot like assuming a spherical cow.
Posted by: eddie at Nov 3, 2006 9:22:01 AM
It seems to me that the two drivers, especially in a congested area, may together pay far far less than the actual costs of the accident, when you include the lost time of the drivers delayed by the accident. That's an externality that neither driver pays for. Although you may need to subtract from that the entertainment value derived by rubber-neckers.
Posted by: Grant Goodman at Nov 3, 2006 9:36:59 AM
Tyler,
I am either missing something fundamental, or the whole idea of an insurance externality is a non-starter.
If the total insurance cost increase is $2500 for an increase from 1 million drivers to 1 million + 1, then, in the normal course of events, premiums for all drivers will increase by a quarter of a cent, as opposed to being attached to a single, unidentifiable marginal driver.
If the premiums are not increased, then the insurance companies will either see their underwriting profits reduced or their losses increased, a generally unstable state. There is no unpaid insurance externality.
Regards, Don
Posted by: Don Lloyd at Nov 3, 2006 10:32:35 AM
Speaking of externalities ... I happened to see your post right after this one:
Posted by: odograph at Nov 3, 2006 10:53:44 AM
I think we should start by killing the economists. It seems that a vast number of otherwise sane people have been seduced by the social engineering possibilities available with Pigouvian taxes.
Save the world by taxing humanity to death!
Posted by: Foobarista at Nov 3, 2006 1:08:47 PM
"market failure"
Given that the road system and drivers licenses
are publicaly provieded, how can this be a
"Market Failure?" Maybe it is a Government Failure?
P.S I know its a generalized mechanism that could be applied to other cases.
Posted by: Robert Sperry at Nov 3, 2006 1:36:33 PM
Social costs from more people in California? This can not be true. More than 100% of the population growth in California is from immigration (Americans are net leaving). This means that immigration has negative externalities. Not exactly PC thinking. The broad point is that population growth has strong negative externalities associated with it that the Open Borders crowd refuses to admit. California is now one of the poorest states in the US. The accident costs mentioned here may or may not be correct. However, the direction is clear.
Posted by: Peter Schaeffer at Nov 3, 2006 2:44:14 PM
How close are we to having the technology for private roads?
Posted by: josh at Nov 3, 2006 3:54:25 PM
josh: since the required technology consists of paving equipment and toll booths, I'd say we're pretty much there now.
Posted by: eddie at Nov 3, 2006 4:10:15 PM
If the total insurance cost increase is $2500 for an increase from 1 million drivers to 1 million + 1, then, in the normal course of events, premiums for all drivers will increase by a quarter of a cent, as opposed to being attached to a single, unidentifiable marginal driver.
If the premiums are not increased, then the insurance companies will either see their underwriting profits reduced or their losses increased, a generally unstable state. There is no unpaid insurance externality.
I think insurance companies already take this into account, as you surmised: when my family recently moved across town our car insurance rates went up.
Posted by: bob montgomery at Nov 3, 2006 5:13:24 PM
"An externality "is a cost or benefit arising from an economic transaction that falls on a third party and that is not taken into account by those who undertake the transaction."
In crowded conditions every member of the crowd imposes a cost(c) on every other member. The n+1 member costs the crowd nc but they cost him nc. There is no thrid party involved where where the cost is not taken into account. The cost of accidents are only a special case. Would we have the government impose a Pigouvian tax every time we got into a line at a movie theater or went to sold out rock concert?
Posted by: joan at Nov 4, 2006 12:55:18 AM
There is a deffinate need for carpooling to work and finding a faster more efficient
way to commute. There may have been more accidents, but its the fuel we are really
worrying about at this time. Less drivers on the road seems to be the best solution.
We are then conserving fuels, road space by carpooling and cheaper ways of
transportation. Efficiency will come with alternative roads.
Posted by: renee at Nov 8, 2006 8:45:39 AM
Weight loss is easier than you think!
Posted by: adipex at Mar 4, 2009 10:10:51 PM
Hello!
very nice post... enjoyed it very much.
Thank you
http://www.best-tutor.com
good site
Posted by: tutor at Jun 3, 2009 10:02:51 AM
