January 12, 2003
The Future of Economics

Every year at about this time I get extremely hopeful about the future of economics. You see, I read the dissertations-in-progress of those who hope to land assistant professor jobs this spring.

Here are three who look like the very, very best:


Ericq Field: Over the past decade, the Peruvian government has issued property titles to over 1.2 million urban households, the largest government titling program targeted to urban squatters in the developing world. This paper examines the labor market effects of increases in tenure security resulting from the program. In particular, I study the direct impact of securing a property title on hours of work, substitution of home for market work and substitution of adult for child labor. To isolate the causal effect of ownership security I make use of differences across regions induced by the timing of the program and differences across target populations in the level of pre-program tenure security. My estimates suggest that titling results in a substantial increase in labor hours, a shift in labor supply away from work at home to work in the outside market and substitution of adult for child labor. For the average squatter family, granting of a property title is associated with a 17% increase in total household work hours and a 47% decrease in the probability of working inside the home. Furthermore, for households with children, urban property titling is associated with a 28% reduction in the probability of child labor.

Hoyt Bleakley: Disease and Development: Evidence from Hookworm Eradication in the American South. Disease and Development: Evidence from the American South. August 2002. Corporate Dollar Debt and Devaluations: Much Ado About Nothing? (The IADB has placed a "bounty" on these results.) Maturity Mismatch and Financial Crises: Evidence from Latin American Corporations September 2002. Language Skills and Earnings: Evidence from Childhood Immigrants. New Data on Worker Flows During Business Cycles. Shifts in the Beveridge Curve, Job Matching, and Labor Market Dynamics.

Roland G. Fryer Jr.: ...This paper introduces a psychological notion of categorization into economics and derives its implications for economic decision making. We show, using a tractable model of social cognition, that a decision maker in (efficiently) assigning past experiences to categories, will sort experiences of interaction with larger (majority) groups more finely than experiences with smaller (minority) groups. We then apply the model to understand simple forms of discrimination and social identity. It is shown that discrimination in hiring can result from such cognitive processes even when there is no malevolent taste to do so and workers' qualifications are fully observable. The model also provides a framework that is equipped to investigate the social psychological concept of identity, where identity is viewed as self-categorization.

Posted by DeLong at January 12, 2003 08:29 AM | Trackback

Email this entry
Email a link to this entry to:


Your email address:


Message (optional):


Comments

In regard to the dissertations, not tim.

Huh!

Posted by: Tom Strong on January 12, 2003 09:18 AM

http://www.nytimes.com/2003/01/09/business/09SCEN.html

Study Looks at Squatters and Land Titles in Peru
By ALAN B. KRUEGER - NYT

An estimated 400 million to 600 million people worldwide are squatters, living on land they have no legal right to occupy, usually on the outskirts of cities. Urban squatting poses a growing economic problem in less-developed countries.

Hernando de Soto, a Peruvian economist, has long advocated reforming property rights for squatters as the path to economic development for poor countries. Now hard evidence suggests there may be something to Mr. de Soto's argument, although for reasons he did not anticipate.

The new research was done by Erica Field, a Ph.D. student in economics at Princeton University, and is available from www.irs.princeton.edu. For the last three years, she has studied land title reform in Peru, which, at Mr. de Soto's prodding, began the world's largest program to provide property titles to urban squatters. From 1995 to 2001, more than 1.2 million households — 6.3 million people — received title to the properties they were inhabiting.

Before the reforms, it could take years for squatters to obtain a title. The reforms, started in different neighborhoods at different times, enabled squatters to get titles quickly, for a fee of $20 to $50. To be eligible, squatters needed only to establish that they had lived in their residences before 1996.

Without a title, Ms. Field hypothesizes, squatters devote extra time and effort to protecting their residence from intruders. This typically requires someone to stay home and defend the property, reducing the opportunity to work in the formal labor market. Cutting the opposite way, however, is that home ownership increases a squatter family's wealth; all else being equal, wealthier families work less.

Ms. Field exploits the natural experiment created by the staggered introduction of title reform in different areas of Peru to test whether granting land titles increased or decreased the amount of labor supplied by squatters....

Posted by: on January 12, 2003 01:23 PM

Wonderful thesis....

She first looks at squatters eligible for a title at the start of the reform in 1995. For this sample, she compares the number of hours worked per family in 2000 depending on whether the family lived in a neighborhood that had already introduced title reform or in another neighborhood in the same city that had yet to introduce reform.

The neighborhoods where title reform was introduced early were quite similar to those where it was introduced late, as far as poverty, literacy and access to sewage and electricity were concerned. Still, as an additional layer of control, she subtracts the difference in hours worked by those in previously titled families between the two sets of neighborhoods.

She performs the analysis using a sample of 2,750 households surveyed in eight Peruvian cities in 2000. About a third of the households were in neighborhoods where title reform had already been introduced.

Her results strongly suggest that labor supply increased as a result of land titling. Two to three years after title reform came to a neighborhood, families that were in untitled houses before the reform worked 20 more hours a week, on average, than those from untitled houses in neighborhoods that had not yet had title reform. By contrast, among households titled before the reforms, family work hours were a statistically insignificant four hours greater in areas where reform was introduced early than in those where it was yet to be introduced.

These results suggest that introducing title reform caused hours of work to rise by a net 16 hours a week — or 17 percent — among those who initially lived in untitled residences. And this figure understates the effect among families that actually obtained a title, as only about 70 percent of squatters had obtained a title two to three years after reform was begun in their neighborhood.

Despite the overall rise in work hours, the amount of labor supplied by children was an impressive 27 percent lower for families that were initially squatting in neighborhoods that introduced title reform early. Although school enrollment did not rise, this may be because many children in Peru already attended school while they worked. Still, the reduction in child labor should allow children to devote more time to school.

Ms. Field's results further suggest that too many squatter families try to run a small business out of their house so they can safeguard the homestead at the same time. Title reform enables them to run a business or work outside the home.

Lastly, she finds that those who were initially squatters and live in areas that underwent title reform are more likely to say they consider their dwelling very secure from eviction or invasion than are squatters in areas that have yet to undergo reform. This reinforces the notion that the shifts in work effort are indeed a result of the additional security provided by land titling.

Mr. de Soto has argued that giving title to squatters fosters growth by enabling entrepreneurs to leverage their capital; squatters can use their homes as collateral to obtain a loan to start a business or renovate a house, or sell a home and buy a new one.

The benefits that Ms. Field documents, however, relate to the improved allocation of work effort from not having to devote labor to protect the home and possessions. Although important, especially for the families involved, these benefits are unlikely to increase a country's long-run growth rate greatly because they represent a small proportion of nationwide work hours.

Preliminary studies by the Peruvian government have not found much evidence that access to credit has increased for families that gained titles. But Ms. Field notes that increased access to credit may come at a later stage, as has been the case for government loans for housing materials. The irony may be that although title reform has been touted as a boon for capital markets, at least in the short run its greatest benefit is in the labor market for poor workers.

Posted by: on January 12, 2003 01:29 PM

So, you have Field, who appears to be hard at work rediscovering the Hawthorne Effect on a macro scale, Bleakley, whose list of papers looks like a World Bank job application and Fryer, who appears to want to be to sociology what Krugman was to geography; a lunkish outsider who won't read the literature.

Forgive me my lack of enthusiasm; it is early on a Monday morning.

Posted by: dsquared on January 12, 2003 11:14 PM

Mr Squared,
does your summary of Ms Field's work capture your assessment of the whole Institute for Liberty and Democracy project?

Posted by: jack on January 13, 2003 06:05 AM

DD

The research on Peruvian land reform strikes me as most important. Land reform is a critical issue in so many poor countries. Especially so in southern Africa. Why are you not more impressed with this thesis?

Posted by: on January 13, 2003 10:41 AM

Ya really wanna know? Because I read the .pdf working paper attached.

1) It's got a completely superfluous assumption of a twice-differentiable, continuous and monotonic utility function in it, which adds nothing to the discussion of the actual problem.

2) An unanalysed concept of "security" is just an input to that utility function. This actually doesn't cause to many problems in the empirical model because the unanalysed concept is just instrumentalised as the titling reform in estimation, but it's irritating that the author has seemingly made no attempt to address about a hundred years of sociology literature.

3) It's an event study across a period which saw a huge policy action without any attempt to consider the Hawthorne Effect.

4) Most seriously, though there is a certain attempt to conceal this with modish language, the guts of the model is an OLS regression with no time structure. This is really serious because this sort of model cannot distinguish endogeneity in the titling reforms.

For example, a plausible alternative hypothesis might have been that the Peruvian government decided to roll the titling program out to the areas as and when they judged the conditions to be most propitious; it would indeed have been perfectly sensible to do this. Obviously this would bias the results, because to an extent the titling program would have been brought in in response to the improving conditions rather than vice versa. The model used cannot pick this effect up. I am frankly shocked that Brad was impressed by such a Stone Age statistical framework.

It's not that the Fields paper was *bad*; it actually looks quite good in that the author is tackling an interesting problem and has had the gumption to seek out a new dataset. But it's got all the faults of most modern economics, and it's got them in absolutely stereotypical ways. I don't understand how it can give anyone "new hope" for the profession.

Posted by: DD on January 13, 2003 01:42 PM

Ya really wanna know? Because I read the .pdf working paper attached.

1) It's got a completely superfluous assumption of a twice-differentiable, continuous and monotonic utility function in it, which adds nothing to the discussion of the actual problem.

2) An unanalysed concept of "security" is just an input to that utility function. This actually doesn't cause to many problems in the empirical model because the unanalysed concept is just instrumentalised as the titling reform in estimation, but it's irritating that the author has seemingly made no attempt to address about a hundred years of sociology literature.

3) It's an event study across a period which saw a huge policy action without any attempt to consider the Hawthorne Effect.

4) Most seriously, though there is a certain attempt to conceal this with modish language, the guts of the model is an OLS regression with no time structure. This is really serious because this sort of model cannot distinguish endogeneity in the titling reforms.

For example, a plausible alternative hypothesis might have been that the Peruvian government decided to roll the titling program out to the areas as and when they judged the conditions to be most propitious; it would indeed have been perfectly sensible to do this. Obviously this would bias the results, because to an extent the titling program would have been brought in in response to the improving conditions rather than vice versa. The model used cannot pick this effect up. I am frankly shocked that Brad was impressed by such a Stone Age statistical framework.

It's not that the Fields paper was *bad*; it actually looks quite good in that the author is tackling an interesting problem and has had the gumption to seek out a new dataset. But it's got all the faults of most modern economics, and it's got them in absolutely stereotypical ways. I don't understand how it can give anyone "new hope" for the profession.

Posted by: DD on January 13, 2003 01:44 PM

Well, DD, when Fields' work is published, you have a readymade opportunity to publish a "Comment" and add a line on the old CV.

Posted by: Michael Harris on January 13, 2003 09:08 PM
Post a comment
Name:


Email Address:


URL:


Comments:


Remember info?