Thursday, July 27, 2006

Heterogeneous Treatment Effects

There are many treatments in this world. I bet that reading this blog has a small negative treatment effect on your IQ while brushing your teeth has a positive effect on reducing cavities.

Many economists are attempting to measure such treatments because we care about what is "good public policy" and we care about causality. If deworming children in a LDC CAUSES an increase in school attendance, then this suggests that there are benefits from such a policy that may exceed the costs of the policy.

Treatment effects researchers face at least 2 challenges. First, they must establish that the treatment they are studying such as "attending Harvard" is not correlated with key variables that a researcher may not observe such as "intrinsic aptitude". If there is a positive correlation between the key causal variable and such unobservables, then a researcher will over-estimate the impact of the treatment (in this case attending harvard) because it partially proxies for high unobserved ability. The second challege a researcher faces is extrapolation. If a treatment works in one setting, how do you know that it will work in a different nation and a different economic environment. This raises the issue of "heterogeneous treatment effects" which Deaton talks about below.

Angus Deaton's quote

"But there are limits. Take Banerjee’s example of flip charts. The effectiveness of flip charts clearly depends on many things, of which the skill of the teacher and the age, background, and previous training of the children are only the most obvious. So a trial from a group of Kenyan schools gives us the average effectiveness of flip charts in the experimental schools relative to the control schools for an area in western Kenya, at a specific time, for specific teachers, and for specific pupils. It is far from clear that this evidence is useful outside of that situation. This qualification also holds for the much more serious case of worms, where the rate of reinfection depends on whether children wear shoes and whether they have access to toilets. The results of one experiment in Kenya (in which there was in fact no randomization, only selection based on alphabetical order) hardly prove that deworming is always the cheapest way to get kids into school, as Banerjee suggests."


Jim Heckman's recent work has pushed further on this point. If you would like to read some hard papers take a look at these:

If people differ with respect to what they will gain from taking a treatment and if THEY KNOW what their personal benefit would be, then there will be self-selection--- those who gain the most from treatment will be more likely to take treatment. A researcher who ignores this point will tend to over-estimate the gains from treatment for a random person assigned treatment from observing the treatment effects for the treated.

It will be interesting to see how radomization researchers handle the challenge that Heckman has posed. For example, If i know that i would gain greatly from treatment but I've been assigned to the control group in your trial will I pursue efforts to gain treatment? In this case, this would "contaminate" the experiment because you would code me as part of the control group while I really took treatment. In this case, the treatment effect estimate would under-estimate the true treatment effect because the control group represents a mixture of people who did not take the treatment and those (like me) who figured out a way to take it.

Wednesday, July 26, 2006

Berkeley, California Rules!

Did you know that every 8th vehicle in Berkeley is a Toyota Prius? Rather than attend the NBER Summer Institute Environmental meetings or the Real Estate meetings, I’m counting cars and getting some exercise walking the Berkeley Hills. The abundance of hybrids in Berkeley supports my core claims in new paper Do Greens Drive Hummers or Hybrids? Environmental Ideology as a Determinant of Consumer Choice. I will post that paper soon.

I would have enjoyed seeing my old friends at the summer institute but I must admit that I have trouble sitting still and listening to 45 minute presentations. Given how fast people can read, a better system would be to require that attendees read the paper ahead of time and then participate in a group discussion about the paper. My method would save time and people would learn more than in a setting where the paper presenter drones on barfing back the content of his power point slides.

I think the NBER summer institute’s great attraction is that it makes everyone feel like a young graduate student again. Even non-Harvard, MIT types can convince themselves that they are “Cambridge graduate students”, there to learn, mix and network. I attended one day of the economic history meetings and one day of the industrial organization meetings. Enough is enough.

San Francisco has offered a recent test of the law of unintended consequences. What happens when you offer free rides on BART, the rail transit system, on hot polluted days? The well meaning officials hope that car commuters substitute to BART and hence reduce urban smog pollution.

What really happened? The San Francisco Chronicle’s Matier & Ross report today that “The highly touted spare the air days spelled financial relief for thousands of bay area commuters and help for the environment but for BART, this summer’s experiment with 6 days of free rides also brought so much grief that the system’s police chief is calling for the program to be limited to just commuters in the future. The free rides brought an extra 25,000 patrons to the already taxed system during the spare the air days, an increase of 10% over normal. There were packed cars, blaring boom boxes, food and drink containers (which are banned) being tossed everywhere --- even reports of homelss people flocking in to beat the heat.”

I am not riding the BART. Instead, when I’m not hiking around I’m writing. Sunshine without humidity allows one to think clearly and I’m functioning!

Thursday, July 20, 2006

Is Boston the Epi-Center of Academic Economics?

All University of Chicago graduates know the answer to that question! With the NBER summer institute in high gear, plenty of important economists are dragging themselves to humid Boston this month. Tomorrow, I will go attend the NBER industrial organization meetings in an attempt to sit still and learn something.

Boston certainly has a great set of economists when you add up all of gang at Harvard, MIT, BU, BC, Tufts and Brandeis, Clark and some other joints in town. Perhaps St. Louis is the new epi-center? Or will you argue that the buildup at Columbia and NYU puts New York City back on top?

My wife and I are about to run a small field experiment. We will move to Berkeley for a few weeks pretty soon and then we are planning to be in Los Angeles in early 2007. Will our productivity drop? Is sunshine a complement of leisure or research output? I will let you know.

Sunday, July 16, 2006

Green Cities: Urban Growth and the Environment will be published in August 2006 by Brookings Institution Press

I'm proud that Brookings will publish my book, Green Cities: Urban Growth and the Environment, next month. I promise to send out copies of this book to bloggers who are willing to read it! You can e-mail me at

What is a green city? What does it mean to say that San Francisco or Vancouver is more "green" than Houston or Beijing? When does urban growth lower environmental quality, and when does it yield environmental gains? How can cities deal with the environmental challenges posed by growth? These are the questions Matthew Kahn takes on in this smart and engaging book.

Written in a lively, accessible style, Green Cities takes the reader on a tour of the extensive economic literature on the environmental consequences of urban growth. Kahn starts with an exploration of the Environmental Kuznets Curve (EKC)—the hypothesis that the relationship between environmental quality and per capita income follows a bell-shaped curve. He then analyzes several critiques of the EKC and discusses the implications of growth in urban population and surface area, as well as income. The concluding chapter addresses the role of cities in promoting climate change and asks how cities in turn are likely to be affected by this trend.

As Kahn points out, although economics is known as the "dismal science," economists are often quite optimistic about the relationship between urban development and the environment. In contrast, many ecologists and environmentalists remain wary of the environmental consequences of free-market growth. Rather than try to settle this dispute, this book conveys the excitement of an ongoing debate. Green Cities does not provide easy answers complex dilemmas. It does something more important—it provides the tools readers need to analyze these issues on their own.

Matthew E. Kahn is professor of international economics at the Fletcher School of Law and Diplomacy at Tufts University. Kahn has published widely in the fields of environmental, urban, and international economics, including papars on sprawl, public transit, and the costs and benefits of environmental regulation. He also blogs on environmental and urban economics at

Friday, July 14, 2006

The Quality of Capitalist Products: The Role of Reputation

In Arthur Miller's play "All My Sons", a World War II military contractor makes shoddy planes in order to earn a fast buck. When these products prove to be defective and his shriking is revealed, he commits suicide. Does this "story" apply more broadly to the quality of capitalist products?

The recent Boston Big Dig disaster would say "yes". In response to my post yesterday, one smart reader posted: "is there any reason to suppose that the problem is unique to public works? If Boeing (to choose a random company that does big contracts) had these problems, how would we know anything about them?"

IN a public works project such as the Big Dig, I am guessing that there were over 30 firms working together to do the big project. Multiple firms makes it harder to establish who is responsible for each task in a complex project this can raise accountability issues when "bad things" happen.

In a free market setting, when you get on an airplane --- the producers of that airplane such as Boeing may have stronger incentives to build a quality plane. Nancy Rose has documented that plane crash deaths per mile of travel have fallen sharply since the 1950s. Did safety regulation cause this progress? I do not know. I do believe that reputation and fear of loss of a "good name" causes companies to not shirk and to not scrimp on providing costly quality. Suppose that Boeing knows that there will be a .2% chance of a crash if they build a crappy plane and a .0005% chance of a crash if they build a high quality plane, the rational company's decision to build the high quality plane will depend on:

1. how much future business they will lose if a plane crash occurs (as airlines substitute and purchase planes from other domestic businesses or from abroad)
2. how costly it is to build a higher quality plane
3. how much extra liability they will face from a plane crash

Returning to the Big Dig, my question is whether #1 matters in this specific case. If companies do not fear losing their reputation, then tney have more of an incentive to produce shoddy products in the short run (the Arthur Miller case). If the Big Dig hired "small businessmen" rather than big corporations to do many of the jobs then an unintended consequence is to weaken the reputation incentive and this leads to a shoddy product.

So my point here is that big corporations greatly value preserving their good reputation and will take on extra costs to preserve it. We see Dell computer today very concerned that it is earning a rep for having an unhelpful "help desk". This is just the tip of the iceberg.

Thursday, July 13, 2006

Trends in Corruption in Public Works Projects: Is the "Big Dig Disaster" an Anomaly?

Two leading Economic Historians wrote an important "long run trends" paper that is relevant for thinking about the recent Boston Big Dig construction problems. Engerman and Sokoloff examined cost over-runs in major public works projects over a long time period. To my surprise, % cost over-runs have increased over time.

I had thought that "muckraker" press would now do a better job monitoring the public sector to provide public goods more efficiently. This is an interesting case of asymmetric information. If for profit firms can sell cheap low quality cement for the project at a high price and not get caught, then the overall quality of such projects will be low. It is possible that the modern public works project is so complicated (such as the Big Dig) that the journalists cannot possibly monitor the contractors and would get roughed up if they really tried. Complexity increases monitoring costs and exacerbates the principal-agent problem.

So to return to the Big Dig, was the recent disaster just "bad luck" that took place on a high quality project or is this a classic case of shoddy government services due to fundamental contracting problems in the face of high monitoring costs?

Digging the Dirt at Public Expense: Governance in the Building of the Erie Canal and Other Public Works

Stanley Engerman, Kenneth L. Sokoloff

NBER Working Paper No. 10965
Issued in December 2004
NBER Program(s): DAE

---- Abstract -----

The Erie Canal was a mammoth public works project undertaken largely because the scope of the investment was beyond what a private firm could manage during the early 19th century. As with most public works, there were ample opportunities for public officials to realize private gains from the effort, and many did. On the whole, however, the construction of the Erie Canal (and most other major public works projects of the era) appears to have been well conceived and executed; it not only paid off more than its costs through tolls, but also generated substantial welfare improvements for the residents of the state of New York in the form of producer and consumer surplus and a wide range of positive externalities. Although there was obviously some fraud and mismanagement, the public authorities carried out the work at costs relatively close to those projected at the point of authorization. In an effort to try to place this episode in a broader perspective, we compare the ratio of actual expenditures on construction relative to the estimated costs at the time of authorization for the Erie Canal, to those for a range of other public works over American history up to the present day. It is our contention that this measure, albeit quite narrow in focus, is informative about the quality of governance of public resources. We highlight how, by this standard, the governance of public resources during the canal era stands up well in comparison with what we have seen since. Indeed, the cost overrun ratios have risen sharply over the last half-century, coinciding with both a marked increase in the relative size of the government sector as well as sustained economic growth. These patterns suggest how important it is that better measures and other means of systematically studying how the prevalence and effects of corruption vary across different contexts be developed.

Wednesday, July 12, 2006

When the Rich Get Richer, what happens to the poor?

If Matt is a jealous guy, then as Bill Gates grows even richer Matt's envy may cause him to suffer a welfare loss. But if the middle class and poor are not envious, are they made worse off as income inequality increases?

It depends. If the super-rich use some of their $ to finance nice musuems and art galleries and Buffett-Gates their money to solve public health challenges then society could be improved along some dimensions.

Alternatively if this crew uses their money to purchase 100,000 square foot homes in desirable areas then the price of land will skyrocket and the middle class will be squeezed out.

Below I report on a new NBER working paper on this subject. My only serious question here relates to the supply side. How do for profit firms respond to income inequality? Joel Waldfogel keeps making the point that in many markets there are fixed costs to producing a variety. If Matt wants a purple shirt, the average fixed cost of producing such a shirt would be a lot lower if there were lots of other people who also want purple shirts. Joel's point is that there are preference externalities. If the middle class shrinks, will they have more trouble finding their ideal varieties as it isn't profitable for suppliers to make them.

Do Rising Tides Lift All Prices? Income Inequality and Housing Affordability

Janna L. Matlack, Jacob L. Vigdor

NBER Working Paper No. 12331
Issued in June 2006
NBER Program(s): AP EFG

---- Abstract -----

Simple partial-equilibrium models suggest that income increases at the high end of the distribution can raise price paid by those at the low end of the income distribution. This prediction does not universally hold in a general equilibrium model, or in models where the rich and poor consume distinct products. We use Census microdata to evaluate these predictions empirically, using data on housing markets in American metropolitan areas between 1970 and 2000. Evidence clearly and unsurprisingly shows that decreases in one's own income lead to less housing consumption and less income left over after paying for housing. The effect of increases in others' income, holding one's own income constant, is more nuanced. In tight housing markets, the poor do worse when the rich get richer. In slack markets, at least some evidence suggests that increases in others' income, holding own income constant, may be beneficial.

Monday, July 10, 2006

The New York Times Discusses New York City

There are two intriguing articles in today's NYT that focus on New York City. Paul Krugman examines the puzzle that corporate headquarters are moving back to New York City. Information technology has allowed corporations to split into at least two spatial pieces. Put bluntly in the new economy, the bosses work in high density expensive real estate so that they can network with clients and work with lawyers and accountants on deal making while the "worker drones" are exciled to cheap suburban corporate towers.

For a serious urban economics paper fleshing out this important point, see Rossi-Hansberg, Esteban and Pierre-Daniel Sarte and Raymond Owens III Firm Fragmentation and Urban Patterns.

Krugman's article does not delve into whether executives want to work downtown because of "consumer amenities" (i.e good restaurants and culture) or because downtown raises their productivity. Clearly, you can't ask the executives this question because they would claim the latter.

The second article in the paper today is here:

In the New York State energy market there is the potential for the gains to trade with upstate New York selling its cheap power to New York City. As a physicist would tell you, there is a catch. Power lines are needed to "export" the power to New York City. You don't have to be Meryl Streep to know that people don't love power lines. All sorts of NIMBY issues are breaking out.

This is an interesting example of the Coase theorem and O-ring technologies. Will the power lines be routed in crazy non-linear ways if there are rural upstate towns
who absolutely refuse to have powerlines run through their town?

Will Hilary Clinton side with "the people" or with the NYC rich?

Monday, July 03, 2006

Will China's Cities be "Green" in 2050?

I've joined a group of scholars who are starting to write a longer policy paper on China's cities. In particular, I'm thinking about what are the key lessons to take away from the U.S and Europe's urban experience with respect to achieving the "win-win" of economic growth and keeping cities clean and green?

When I look at China's cities today, I see a couple of things;

1. Urban air pollution (particulates) in major cities such as Beijing is too high and with the scale of driving continuing to increase --- urban smog will also grow worse

This pollution is supplied from multiple sectors; industry, transport, electric power generation and perhaps home heating. Some reports claim that Beijing is importing dust from nearby rural areas where desertification has taken place.

2. Water demand and supply ---- urban areas demand more water and have the money to pay for it. Similar to the U.S , much water (perhaps too much) goes to agriculture. What is the efficient allocation of this scarce resource? How can market signals (i.e prices) be harnessed to signal the value of this scarce resource?

3. The ecological footprint and greenhouse gases; environmentalists love calculations such as; "if all billion people in china buy a car and if a car has 2 tons of steel, then 2 billion tons of steel will be needed --- the aggregate demand here vastly swamps aggregate supply of steel (

such aggregation logic ignores price signals and behavioral responses to those signals ---- but the issue remains for non-market goods such as greenhouse gases how many tons will China produce in 2050? Does urban growth in China help mitigate or exacerbate this global externality? As I discuss in my book's chapter 2; the ecological footprint can certainly be shrunk when there are market pricing signals. The challenge is how to get a government to sign on to present resource consumers with these signals!

So, I have 30 pages to cover all of these topics in my report! If you think I have omitted any other key Chinese urban issues. please post here and I will actually read what you say!