The Perfect and the Free

Are too many pedestrians dying?

I opened up the morning paper today to see a story about the Surface Transportation Policy Partnership's ranking of the "most dangerous metro areas for pedestrians."
 The local hook was how Raleigh (which is now a separate metropolitan entity from Durham) is the sixth-worst city for pedestrians according to the "pedestrian danger index."  I went and downloaded the report this morning.  Aside from providing a nice set of grisly statistics and anecdotes, the report argues that we ought to be spending more on pedestrian safety, particularly in newer Sun Belt cities that were built for the car and not the horse-and-buggy.  Here are a couple of quick factoids for you:

-- Large city with the worst pedestrian fatality rate: Tampa, with 3.52 deaths per 100,000 each year.

-- By comparison, the average murder rate across all metropolitan areas is 4.6 per 100,000.

-- The "pedestrian death index" is the pedestrian fatality rate divided by the percent of workers who walk to work.  So cities with a lot of walking college students (Raleigh would be one), tourists (Las Vegas, Orlando), or just plain leisurely strollers get a raw deal -- the miles traveled by pedestrians may be high, but if you're not walking to work, you don't count.  At least they are normalizing by something, I guess.

Any economist would naturally want to evaluate the organization's policy claim: that investments in pedestrian safety would save lives.  Let's do some back-of-the-envelope calculation here.  First, let's do some causal inference with n=2: compare two cities that spend very different amounts on pedestrian safety and see what difference it makes in terms of fatalities:

Richmond, VA: $0.15 per person per year on safety, fatality rate 1.25 per 100,000

Providence, RI: $4.01 per person per year on safety, fatality rate 1.35 per 100,000

So if every city switched from being like Richmond to being like Providence, we'd save 1 life per million people each year.  The cost, in terms of federal spending, would be $3.86 per person per year.  That amounts to $3.86 million per life saved.  Attaching a dollar value to saved lives is a controversial business, but most folks would agree that $3.86 million is a reasonable level of cost-effectiveness.

But wait!  We haven't considered all the costs yet.  The number one strategy for saving pedestrian lives is to slow down cars.  You can do this with "traffic calming" road designs, etc.  Suppose that among every 1 million people, 400,000 drive cars to work.  Suppose further that switching from Richmond-style pedestrian protection to Providence-style protection means that drivers spend 1 extra minute per day commuting.  Over 200 workdays, that's 80,000,000 extra minutes spent commuting.  Now, suppose that people will pay $6 to reduce the amount of time spent commuting each year by 1 hour (a reasonable estimate, most transportation studies assume that people will pay one-half their hourly wage to cut their commute time by an hour).  We then have to add $8 million to the cost of saving one pedestrian life.  A cost of $11.86 million per life saved sounds a little bit less like a good deal.  The same amount of money devoted to medical research or homicide prevention, for example, might accomplish much more.

We also shouldn't forget that idling cars (except hybrids) spew carbon dioxide and other noxious pollutants into the atmosphere.  A quick calculation reveals that the extra 80 million minutes of idling per life saved results in an extra 8,800 tons of carbon dioxide emitted each year.  

Ultimately, there can be little doubt that public investments could cut the number of pedestrian fatalities each year.  There is much more doubt, however, about whether such investments really represent the best possible use of taxpayer dollars.

My breakfast with Encarnacion

On Friday morning I was in Washington, DC, attending the APPAM annual meeting (for researchers who study public policy issues).  I tend to avoid staying in the conference hotel for events like this -- it's just nice to have an excuse to leave the building at the end of the day.  So, instead of the Marriott hotel on 22nd street, I was up at the Embassy Circle guest house, on R street just off Sheridan Circle.  Staying at the same guest house, unbeknownst to me, was Encarnacion Bail, a native of Guatemala who was apprehended by immigration authorities at the Missouri poultry processing plant where she worked in May 2007.  I was introduced to Encarnacion at breakfast by the innkeepers, who also introduced her lawyer.  Over a meal of toast, smoked salmon, fruit, cheese, cereal, and tea, the lawyer brought me (and several other guests) up to speed on her situation.

At the time of her apprehension, Encarnacion had a five-month old baby.  The baby, born in Guatemala and hence an illegal immigrant himself, was sent to live with relatives, who were themselves illegal immigrants.  These relatives enlisted the assistance of a local teacher, who had offered to look after the baby from time to time.

One day, the teacher reported to Encarnacion's relatives that the baby had been apprehended by U.S. Marshals.  The relatives were understandably upset, but owing to their own illegal status they feared the consequences of launching any kind of official effort to reclaim the boy.  In fact, the boy had not been apprehended at all.  The teacher had instead arranged an adoption.  According to Missouri law, adoptions are private matters and do not need to be supervised by courts.  An adoption can be "certified" by three different types of representatives of the birth parent: a doctor, a lawyer, or a clergy member.  The teacher (or a co-conspirant, I'm not quite sure) had obtained credentials as clergy and had fraudulently claimed to be Encarnacion's pastor.  

Encarnacion's baby received a new first name and last name, and now lives with a new family in the same small Missouri town where his birth mother was apprehended.  The two-year old boy speaks English exclusively.  Encarnacion speaks no English.  As has been reported in the press, Encarnacion has been battling to reclaim her child.  The child's adoptive parents, understandably, are fighting to keep the child.

Encarnacion herself was in Washington to speak with officials at the Guatemalan embassy, and to discuss her situation with Senators and Representatives from various states.  She was represented by a Seattle-based lawyer (a securities litigator by training), who reported that his firm has donated about $650,000 in billing hours to assist with this case.

I wish there were a straightforward moral to this story, but one has eluded me so far.  One could take the stance that this is the sort of outcome Encarnacion should have been thinking about when she decided to cross the border illegally.  When you get right down to it, though, this is a case of a mother losing her child through the nefarious acts of others, and it's almost impossible not to sympathize with a parent who has suffered such a fate.  Caught in the middle is an innocent child, who now faces a conflict between his birth mother, who if victorious will have no choice but to return him to a life of relative poverty in Guatemala, and his adoptive parents -- who are really the only parents he could possibly remember.

Perhaps the moral of the story is this.  In theory there is a third way on immigration policy, and that is the status quo.  Liberalizing immigration would carry significant political and economic risks.  The complete elimination of illegal immigration through stricter enforcement would cost untold billions of dollars -- the Feds so far have tried to outsource much of this through 287(g), and local law enforcement agencies I've heard from really aren't happy with having to enforce somebody else's laws without compensation.  In theory, turning a blind eye to illegal immigration would allow us to avoid both the risks and the costs.

The path of least resistance, though, is the path of continued human tragedy.

Is the end of busing the end of the world?

The Raleigh News and Observer ran a front-page article today, discussing the likely consequences of axing the Wake County schools' busing policy.  I wrote about this a little while ago: opponents of busing are about to take over the school board.

My half-hour phone conversation with the reporter a couple of weeks ago resulted in the little quote you'll see in this article, about how Wake County schools are not going to end up looking like Detroit.  That is definitely true.  What you don't see in this article, however, is a more important part of the conversation.

We actually have startlingly little evidence to support the assertion that busing closes the black-white test score gap.  Yes, there's a logical argument to connect the two: if schools are integrated, you can't have disparities in the quality of teachers or instructional materials for students of different races.  There is also plenty of evidence showing that it's hard to attract and retain good teachers in high-poverty schools.

Unfortunately, though, American cities are not residentially integrated (Raleigh, incidentally, comes about as close as anyplace, which is a far cry from all the way).  So, to achieve school integration you have to relocate kids.  Which kids, in practice, get relocated?  In Raleigh, white kids are basically never forced to attend school in black neighborhoods.  The inner city schools are magnets.  Black kids, on the other hand, are indeed forced to travel long distances, at least in some cases.  So, the "costs" of integrating the public schools are not spread evenly across races.  Integration may be better than segregation, but it isn't clear (to me at least) that busing is a good substitute for true integration.  There are other arguments against busing too -- just google oppositional culture.

There are two really good academic studies of busing and black/white educational gaps.  Jonathan Guryan examined the consequences of school desegregation orders in the '60s and '70s.  Byron Lutz then studied the consequences of court orders mandating an end to busing in the '90s.  Both studies associated busing with lower dropout rates for black students.  Lutz's study found no effect in the South, however.

Moreover, while it's impossible to argue that dropping out of high school is a good thing, these results do not necessarily imply that younger black students achieve more under busing plans.  There have been many efforts beyond Guryan and Lutz to look for these types of effects.  For the most part, they have come up empty, attributing only small effects to school racial composition.

Folks at the N&O are fond of pointing out how segregated the Charlotte-Mecklenburg schools have become since a federal court stopped the busing program there in 2002.  Yes, segregation has gone up in Charlotte's schools.  Has black achievement suffered as a consequence?  I'll have a complete paper on the subject in a couple more months.  The preliminary answer: no.  I need to do a bit more poking around to figure out exactly what can explain the trends in Charlotte.  I'll post here as I learn more.

In the meantime, good residents of Wake county, don't fear the apocalypse.  As I said in the N&O, the end of busing will not be catastrophic.  As the N&O didn't tell you, and probably never will, there might not be any consequences at all.

World Series Special Edition: Why John Rocker Should Move to New York City

It's been quiet around here lately.  Hopefully that will change soon!  Here's a little piece from a couple of weeks ago that ran in the New York Daily News.  It's related to the new report on immigrant assimilation released last month by the Manhattan Institute.

Oops... apparently I've been away so long that I forgot that I linked to this already.  Don't worry, truly new stuff is on the way.