Social Econ Blog

Given our recent focus on beliefs, it seems appropriate to link to the initial segment of "The Colbert Report" -- where Steven Colbert popularized the term "truthiness." The wikipedia entry describes truthiness as "the quality by which a person purports to know something emotionally or instinctively, without regard to evidence or to what the person might conclude from intellectual examination." That is, it describes how people justify their prior beliefs particularly when they are confronted with information which, unless ignored, would force them to update their beliefs.

For more satirical truthiness, see Colbert last night at the White House Correspondents' Dinner.

Today, we will discuss terrorism. Economics provides a great deal of insight into the problem of terrorism. The decision to commit a terrorist act is just that ... a decision. As such, we can identify the relevant tradeoffs and design policies that should reduce the probability of terrorist acts.

Terrorism has its roots in group level hatred. The suppliers of group level hatred are frequently the demanders in the market for terrorist acts (though they needn't be the same people), and the people who demand hatred are usually the people who end up actually carrying out (or supplying) terrorist acts. The whole thing is further confused by the fact that frequently demanders and suppliers of terrorism are the same people.

As such, if we hope to be successful in our "War on Terror", we need to focus on policies that dramatically change either the supply or demand of terrorists. And this ultimately requires reducing the supply and demand of hate.

To begin, focus only on the market for terrorist acts. A key component in our current strategy is negative deterrence. Gary Becker's classic model of criminal behavior suggests that criminals tradeoff the expected benefits (probability of success * value of success) of a criminal act against the expected costs (probability of being caught * penalty). Much of our current terrorism policy is designed to increase the probability of detection and increase the penalties associated with participating in terrorism. This effectively raises the price of terrorist acts by shifting the supply curve up.

The effectiveness of supply side polices is limited, however, by the amount of substitution available in the terrorism production function and the low price of terrorist acts. If we raise the price of attacking a government building, terrorists can attack shopping malls; if we profile all middle-eastern looking men, terrorists can recruit people who look different. Ultimately, the biggest cost faced by terrorists is probably the lives of any suicide attackers. Reducing the supply of suicide bombers would likely making committing terrorism more difficult. However, consumers of terrorist acts have done a good job of convincing their "employees" that there are substantial eternal rewards for their sacrifice. But even reducing the number of willing suicide bombers won't make a huge impact. Killing people in public places is unfortunately cheap. Benn Steil and Robert Litan estimate:

...acts of terror are cheap to carry out. The truck bomb used at the World Trade Center in February 1993 cost $400. The cost of training and support of the 19 individuals who participated in the 9/11 airline hijackings is estimated at only $500,000. Terrorist operations since 9/11 have been even cheaper: the October 2002 Indonesia bombings cost $50,000; the November 2003 Istanbul attacks cost less than $40,000; and the March 2004 Madrid bombings, carried out with dynamite and cell phones, cost less than $10,000. As Defense Secretary Donald Rumsfeld admitted in a memorandum leaked to the press in October 2003, "The cost-benefit ratio is against us! Our cost is billions against the terrorist cost of
millions."

(Source) Ultimately, there is not a whole lot that we can do to make the price of terrorist acts prohibitively high.

Even if we were successful at increasing the price of terrorist acts, the reduction in the equlibrium quantity of terrorist acts depends on the elasticity of demand. Peter Boettke and Christopher Coyne argue in their paper "Liberalism in the Post-9/11 World" that the demand for terrorist acts is highly inelastic (this approach is similar to Becker, Murphy, and Grossman's work on the failure of the "War on Drugs"). As such, supply-side policies are unlikely to reduce terrorist acts.

Reducing terrorism, then, requires fundamentally changing preferences for terrorist acts and prompting an inward shift in the demand curve. Unfortunately, this is extremely difficult. Some of the policies Glaeser outlines in the "Political Economy of Hatred" (e.g., increase ties between groups, promote hating the haters, etc.) will likely have some effect. However, it is important to note that blanket prescriptions like reducing poverty and increasing education are not likely to automatically improve things (Krueger and Maleckova document no relationship between income and education and individual decisions to be terrorists and only small relationships between country level poverty and education and terrorism). Indeed, some rich dude can always set up and fund his own terrorist group (see Al Queda and SPECTER).

So how does the war in Iraq fit into all of this? Initially, it was about reducing the probability of attack by raising the price of engaging in terrorism. The fear was that Saddam Hussein would provide WMD to terrorists, so we needed to intervene in order to "raise the price" of this type of terrorist attack. At the same time, we would "shock and awe" other states who might be willing to sell weapons to terrorists. Thus further raising the price of terrorist activities.

After this rationale turned out to be based in fantasy, the rationale changed to "bringing democracy to the middle east". Setting aside the question of why you would choose to promote democracy from the barrel of a gun in Iraq rather than working to increase democracy in semi-democratic places like Turkey or Egypt, the idea of reducing terrorism via democracy, I guess, works like this -- Creating a democracy in Iraq is supposed to generate more freedom and tolerance. This, in turn, should increase ties with the West and lead to less hatred of the West and more intolerance of terrorists and terrorist groups. This should reduce the supply of terrorists, reduce the demand for terrorism, and make it harder for those who do still demand terrorist acts to operate.

This is not a crazy theory in the abstract, but, in practice, there are a number of problems. The historical record for producing liberal democracies at will is not very good. Boettke and Coyne also consider this issue. They point out that of the 15 times the US attempted reconstruction efforts from 1898-1996 only four were democracies after 10 years. As they remind us, economics predicts that "conflict should not persist where gains from exchange exist." However, in most reconstruction efforts (and Iraq in particular) a number of barriers exist that prevent peaceful bargaining. I encourage you to check out their interesting arguments.

I used to have students present their term paper topics to the class during the last week of class. It always frustrated me a bit, though, because each student only was given 15 minutes or so to present and get feedback. This never seemed to be enough time.

This year we will use this space to present your topics and get feedback. By next Thursday, I want each of you to post a brief description of your topic which looks similar to empex5 (hopefully you will have some at least preliminary results, but if you don't that isn't the end of the world) and includes:

1) A brief description of the hypothesis (hypotheses) that you are examining and a clear empirical prediction. What question are you trying to answer? What are the economics behind your argument/story?

2) What is your empirical strategy? This should include the basic equations you are estimating (e.g., Y = a + BX + cW + dZ + ... + e), a description of the data you employ (i.e., you should briefly describe your dataset), and, if necessary, details of any experiments you are running.

3) Any results you have obtained.

4) A basic description of any concerns with your approach. What might be biasing your results? Non-random sample? Measurement error? Omitted variables? Simultaneous causality? Any other problems which might affect what we can conclude from your analysis?

Keep the post on the main page brief. If you want to go into great detail (and I encourage you to do this), please send me a document version and I will post a link to it.

Please note that part of this assignment is commenting on your tutorial mates ideas in order to help improve their work. This can be done using the comment threads or email (just cc me on any email that you send). You obviously don't need to comment on everybody's ideas, but you should be sure to comment on at least 4 or 5.

Further note that you can post your ideas anytime you are ready. In fact, you have incentives to not post on Thursday. If everyone posts at the same time your ideas might get lost and not draw as many good comments. So you should think about posting earlier before others post in order to attract maximum attention.

Finally, I have longer to turn in grades than I thought, so the term papers will be due Friday, May 12 at 11:59 PM.

Varun's post below reminded me of an interesting paper which also put together a dataset of the full set of admissions decisions for students. This paper sought to create a ranking of colleges based on students' revealed preferences. It is pretty interesting, and it confirms what you all know (and what to hear -- except for you Tony V.) -- Harvard is number 1 (in every ranking they put together except for the one for students who plan to major in humanities). Here is the abstract:

We show how to construct a ranking of U.S. undergraduate programs based on students' revealed preferences. We construct examples of national and regional rankings, using hand-collected data on 3,240 high- achieving students. Our statistical model extends models used for ranking players in tournaments, such as chess or tennis. When a student makes his matriculation decision among colleges that have admitted him, he chooses which college "wins" in head-to-head competition. The model exploits the information contained in thousands of these wins and losses. Our method produces a ranking that, unlike rankings based on the matriculation or admission rate, would be difficult for a college to manipulate. If our ranking were used in place of these measures, the pressure on colleges to practice strategic admissions would be relieved. We show how to deal with tuition discounts, alumni preferences, early decision programs, specialty schools, and similar issues.

The following study suggests that the demand for a college is relatively inelastic to price changes in regional and national institutions while significantly more elastic in local institutions. This suggests that Harvard and other Ivy League institutions could continuosly raise their prices and see little effect in demand for their institution. If this is the case, what motive does Harvard have to not price gouge its students ?(some might already consider Harvard tuition price gouging but it has the opportunity to further raise prices.) At what point are we likely to see a reversal in what factors determine college choice? In other words, how much are students willing to pay for the increased selectivity and attached reputation of an Ivy institution?

This study examined the factors that influence the college choice decisions of high school seniors to matriculate in a cooperative education institution of higher education. A sample of 465 high school seniors was drawn from a database of 25,500 Pennsylvania high school seniors, who had their SAT scores sent to various cooperative education institutions for acceptance in the Fall of 1986. The results indicate that the two primary factors associated with the decision of high school seniors to attend a cooperative education institution of higher education are net price and the selectivity of institutions in their choice set. The findings suggest that mobile students (students applying to primarily regional and national institutions) see cooperative education as having high utility. These students are relatively less sensitive to price changes, such that a modest increase in the net price of the co-op institution will not drive them to choose one of the other alternatives in their choice set. The economic substitution effect is minimal, because these high school seniors see few public college or public university alternatives for cooperative education. They are willing and able to pay more to attend a cooperative education institution. Less mobile students (applied to primarily local and instate institutions) with only public alternatives in their choice sets appear to exhibit the opposite behavior. They are highly sensitive to price changes. Mobile students' college choice behavior is predicated upon the institutional attributes of price rather than the "cooperativeness" of the institutions in their choice set. The results suggest that less mobile students are price shopping for institutions, substituting higher cost institutions with lower cost institutions. The primary conclusion that can be drawn from this study is that a decrease in financial aid to mobile students with only public alternatives, would only modestly alter the probability they will attend a co-op institution. While, for less mobile students, with only public alternatives, an increase in financial aid modestly increases the probability they will enroll in a co-op institution.

Heres the link.

if this doesnt work the article is on EBSCO host under the title

The college choice process: An analysis of the factors that influence the college choice decisions of high school seniors to matriculate in a cooperative education institution of higher education.

In the second class, our discussion of the media turned to spin. I'll post more on spin later, but in the meantime here is a video that Patrick pointed us to in class of Triumph the Insult Comic Dog in "Spin Alley" after one of the presidential debates.

Also during the first class yesterday, we discussed the argument put forth in the Jerome Karabel's book "The Chosen" that group level hatred is responsible for the current Harvard admissions process. Some of you were unfamiliar with the argument. For those interested in learning more, here is Malcolm Gladwell's discussion of the book.

Yesterday before the first class, we were discussing plagiarism and intellectual property. Naturally, one of the examples that was discussed was drug research. At some point, I segwayed into Michael Kremer's idea to incentive companies to research drugs which are desperately needed in the developing world. The specific example I used was a malaria vaccine.

Today, my friend Dave E points out that yesterday was Africa Malaria Day. He provides some of the depressing facts about Malaria in his post (e.g., 1 million people, mostly children and mostly African, die of malaria each year). Further, as I mentioned in class, malaria is not just devastating because it kills; it also disrupts the economy and works to slow growth and development just by making lots of people very sick.

You don't need to have the money to fund a research competition in order to help. Dave also points to a very good Rick Reilly column about his effort to raise money to provide mosquito nets to affected areas and to the Global Fund to Fight AIDS, Tuberculosis, and Malaria.

When we see people expressing, what appear to us to be, "weird" beliefs or engaging in "bizarre" behaviors (which reflect some odd belief), we wonder what the heck is going on. How can someone believe something so totally strange?

Obviously, some "crazy" beliefs reflect people who have not been exposed to alternative views. There are plenty of people who have merely adopted the stock set of beliefs their culture, society, or dominant identity group without seeking out and considering alternative positions. It is hard to think of people with such views as "crazy." They are largely just isolated and uneducated. I am more interested in how people can, in spite of being aware of alternative positions, hold such massively different beliefs.

Applying economic tools can help us understand this issue. As usual, we want to ask what are the benefits and costs associated with having certain beliefs?

Two papers provide interesting models which help provide insights into this issues. First, Ed Glaeser's fascinating "The Political Economy of Hatred" develops a model to explain why groups hate each other. He breaks hatred down into demand and supply asking why would someone be willing to hate another group (which is costly to them) and what prompts people (particularly politicians, etc.) to supply hatred. Here is the abstract:

This paper develops a model of the interaction between the supply of hate-creating stories from politicians and the willingness of voters to listen to hatred. Hatred is fostered with stories of an out-group's crimes, but the impact of these stories comes from repetition not truth. Hate-creating stories are supplied by politicians when such actions help to discredit opponents whose policies benefit an out-group. Egalitarians foment hatred against rich minorities; opponents of redistribution build hatred against poor minorities. Hatred relies on people accepting, rather than investigating, hate-creating stories. Hatred declines when there is private incentive to learn the truth. Increased economic interactions with a minority group may provide that incentive. This framework is used to illuminate the evolution of anti-black hatred in the United States South, episodes of anti-Semitism in Europe, and
the recent surge of anti-Americanism in the Arab world.

The second paper, Laurence Iannaccone's "Sacrifice and Stigma," examines why rational agents willingly adopt strange requirements and beliefs as part of religious and social groups. He argues that having costly beliefs and behaviors is a way of preventing free-riders from lowering the quality of the religious of social experience. That is, people like to attend churches which provide a strong religious experience. However, the strength of the religious experience for everyone is diminished by the presence of a number of people who are only marginally into it. Thus, in order to keep out people who are only marginally religious, churches adopt rigid rules and beliefs in order to make it more costly to participate. Here is the abstract to his paper:

This paper presents an economic analysis of religious behavior that accounts for the continuing success of groups with strange requirements and seemingly inefficient prohibitions. The analysis does not presuppose any special motives for religious activity. Rather, religion is modeled as a club good that displays positive returns to "participatory crowding." The analysis demonstrates that efficient religions with perfectly rational members may benefit from stigma, self-sacrifice, and bizarre behavioral restrictions. The model also addresses sacrifice in nonreligious "social clubs": fraternities, communes, political parties, work groups, and families.

You should try and read at least the introductions to both of these papers.

A few years ago, Matt Gentzkow and Jesse Shapiro examined perceptions of the US among residents of nine predominantly Muslim countries in their paper, "Media, Education, and Anti-Americanism in the Muslim World." The document how beliefs vary across these areas focusing primarily on the relationship between media exposure and education and anti-American views. They find:

We first examine how individuals' overall quantity of media use and years of education relate to the accuracy of their beliefs, as well as to their favorability toward the United States. We find no effect of overall media use and at best a weak effect of total education. We then ask how these relationships differ among different media -- specifically the satellite networks Al Jazeera and CNN International -- and among education systems with different characteristics. We find that the effects of these specific information sources appear to be strong and widely divergent.

These results suggest that access to information itself is frequently not the source of variation in beliefs. How the information is presented and how that relates to what individuals "want" to hear plays an important role.

If you want to see these differences in less abstract, more "real" context, I highly recommend "Control Room." This film went behind the scenes at Al-Jazera during the lead up to and initial invasion of Iraq. It is a fascinating depiction of how people watching the same events can have massively different views of what is going on (and of the power of politicians and the media to shape those beliefs).

(It is also just interesting to watch a time capsule of that period from today's perspective.)

This really has nothing to do with our class, but it is not totally irrelevant because enough of your are taking macro. I really don't know what else to say (other than I am impressed by the quality of the production).

(Oh, and the back story here is that one of the people who was also considered for the fed chair was the dean of Columbia Business School -- Glenn Hubbard.)

This discussion from the Gary Becker and Richard Posner's blog is very relevant to our current topics. These are very smart people. You should check it out.

http://www.becker-posner-blog.com/archives/2006/04/the_economics_o_3.html

http://www.becker-posner-blog.com/archives/2006/04/on_the_economic.html

A taste:

Posner emphasizes two important reasons, habit and coordination costs, for why cultures, including those of nations and companies, often change very slowly. I agree with his arguments on why habitual behavior and coordination-network effects are important, and do not have much to add directly to his discussion. Instead, I will try to explain the seeming contradiction that many aspects of national cultures change drastically and often surprisingly quickly. In a nutshell what I claim is that major economic and technological changes frequently trump culture in the sense that they induce enormous changes not only in behavior but also in beliefs.

Today, Arnold Kling argues that too many students go to college:

I believe that too many college students are not emotionally ready for the experience. They go because their peers and their parents expect them to go. Their parents expect them to go because their parents' peers expect their children to go.

...

College is a highly unnatural setting. Students do not have to shop for food, worry about budgets for entertainment, or face consequences for behavior that is irresponsible if not downright illegal. Instead of learning life skills, they are learning escape-from-life skills.

This article in Forbes also tries to shift in the demand curve for college education. They focus on the many successful people (e.g., Bill Gates) who didn't compete college and various economic research that suggests that college only increases your earnings because of its value as a signal of ability:

In fact, there is plenty of evidence that what really matters is how smart you are, not where -- or even if -- you went to school. According to a number of studies, small differences in SAT scores, which you take before going to college, correlate with measurably higher incomes. And, according to a report from the National Bureau of Economic Research, the lifetime income of high-school dropouts is directly associated with their scores on a battery of intelligence tests.

So is college a big waste? Could we do just as well with an alternative arrangement (e.g., some combination of suped-up community colleges and an ability certification service -- like the Harvard College Admissions Committee)?

Certainly, college is neither necessary or sufficient to guarantee individual success, and many students who enroll in college are not ready to be there and fail to get as much out of college as they could. Further, it is possible that we might be able to devise an alternative system which works better then our current one. However, when evaluating recommendations we don't ask are there some people who are made worse off by the recommendation. Rather, we try and figure out, on average, if there is a positive net effect. As yet, I unaware of any study which decisively estimates the causal effect of a college education (much less one that suggests that it is zero). So I am going to stick with the standard advice -- go to college. College, hopefully, provides you with general skills (writing, math, reasoning, and social) that are difficult to acquire outside of formal learning environments and without intensive study.

Very wealthy politicians fascinate me. Vast resources give such politicians the opportunity to do what they think is best. They don't have to spend enormous effort raising money and can avoid becoming beholden to fundraisers and/or special interest groups. They may not choose to do this, but, at least, they have the option.

Even the very wealthy, however, have obvious interests, and new research by Mara Faccio suggests that the market expects them to act to further these interests. Companies controlled, owned, or previously run by politicians "experienced a 2.3 percent increase in their share prices, on average, around the time of their electoral victories. If a politician entered the executive branch of government, the effect was larger -- up to a 12 percent lift for companies associated with new presidents, prime ministers and other top officials. " These effects were even larger in countries with high corruption.

This is an interesting way of measuring the returns to social capital. A set of entities (companies) are tied to an individual who rises to a position of influence. This shock increases their social capital. The market suggests that this one form of social capital is quite valuable.

It would be cool to be able to do something similar with individuals. Using longitudinal data on individuals social networks (like how many friends they have and how often they are socializing with them), see how much individuals' popularity increases when they experience a shock to the value of their social connections (e.g., their mom or dad wins an election or the lottery).

You can read a full summary of the research in this NYTimes article.

It is not a perfect experiment, but Katrina may provide an opportunity to examine the effects of moving to a "better" neighborhood on individual outcomes. The NYTimes reports today that many Katrina evacuees have ended up residing in neighborhoods with higher incomes then those the left behind. Obviously, Katrina represents a really big shock which may have direct effects in its own right (meaning that "treatment" isn't just "random" assignment to a higher income neighborhood -- it is surviving a hurricane, etc. and ending up in a higher income neighborhood). Regardless, I would expect to see some interesting papers over the next several years that exploit shocks from major natural disasters like the tsunami and Katrina.

Channel surfing while watching the NBA Playoffs (ok, I was also watching a few scenes from "What a Girls Wants" on Nickelodeon), I was reminded on one of the funniest things ever filmed ... David Hasselhoff's "Hooked on a Feeling" video. If you've seen it before enjoy it again. If you've never seen it, check it out.

Ed Glaeser has written several other pieces which describe differences in beliefs across space. Most recently, Ed and David Cutler produced a paper that argues different beliefs about the harmful effects of smoking explain much of the difference in smoking prevalence between the US and Europe:

While Americans are less healthy than Europeans along some dimensions (like obesity), Americans are significantly less likely to smoke than their European counterparts. This difference emerged in the 1970s and it is biggest among the most educated. The puzzle becomes larger once we account for cigarette prices and anti-smoking regulations, which are both higher in Europe. There is a nonmonotonic relationship between smoking and income; among richer countries and people, higher incomes are associated with less smoking. This can account for about one-fifth of the U.S./Europe difference. Almost one-half of the smoking difference appears to be the result of differences in beliefs about the health effects of smoking; Europeans are generally less likely to think that cigarette smoking is harmful.

Ed and Alberto Alesina produced a book "Fighting Poverty in the US and Europe: A World of Difference" that discusses why Europeans and Americans have such different welfare states. In part of the book they discuss the role for different beliefs about poor people (which in turn are related to differences in racial composition). Here is a paragraph from the Economist's summary:

NOTHING better encapsulates the different attitudes of America and Europe to the poor than a table towards the end of Alberto Alesina's and Edward Glaeser's remarkable book*, due to be published later this month. It compares the prevalence of three beliefs: that the poor are trapped in poverty; that luck determines income; and that the poor are lazy. The first is held by only 29% of Americans but by 60% of citizens of the European Union; the second, by 30% of Americans and 54% of Europeans; and the third, by contrast, by 60% of Americans and 24% of Europeans.

A paper forthcoming in the Journal of Economic Perspectives by Ed Glaeser and, well, me outlines the vast differences in beliefs across the U.S. (as part of a discussion of the popular Red State-Blue State framework). Below, I provide some of the relevant facts from the paper. Here is the abstract (and a link to an earlier version):

The division of America into red states and blue states misleadingly suggests that states are split into two camps, but along most dimensions, like political orientation, states are on a continuum. By historical standards, the number of swing states is not particularly low, and America’s cultural divisions are not increasing. But despite the flaws of the red state/blue state framework, it does contain two profound truths. First, the heterogeneity of beliefs and attitudes across the United States is enormous and has always been so. Second, political divisions are becoming increasingly religious and cultural. The rise of religious politics is not without precedent, but rather returns us to the pre-New Deal norm. Religious political divisions are so common because religious groups provide politicians the opportunity to send targeted messages that excite their base.

Some of the findings (numbers represent the fraction of the state's respondents agreeing with the statement in the combines PEW Values Survey 1987-2003):

1. State	N	Schools should have the right to fire homosexual teachers.
Massachusetts	430	0.23
District of Columbia	74	0.26
Connecticut	272	0.26
Maryland	449	0.27
New Jersey	588	0.29
West Virginia	230	0.54
Oklahoma	261	0.56
Tennessee	514	0.60
Arkansas	226	0.61
Mississippi	283	0.65

2. State	N	It is okay for blacks and whites to date.
Kentucky	339	0.35
West Virginia	230	0.40
Tennessee	497	0.41
South Carolina	322	0.43
Alabama	382	0.46
Oregon	240	0.77
California	1860	0.77
Delaware	58	0.79
Maine	124	0.81
District of Columbia	74	0.88

4. State	N	The best way to ensure peace is through military strength.
District of Columbia	77	0.36
Vermont	52	0.40
Oregon	257	0.42
Delaware	62	0.42
Minnesota	418	0.47
Idaho	122	0.66
Oklahoma	265	0.68
Mississippi	281	0.69
Arkansas	230	0.70
South Carolina	330	0.73

5. State	N	When something is run by the government, it is usuall inefficient and wasteful.
District of Columbia	77	0.45
Mississippi	292	0.51
Delaware	63	0.57
Nevada	87	0.57
South Carolina	339	0.58
Montana	113	0.72
Nebraska	189	0.72
Arkansas	242	0.74
Oregon	262	0.74
South Dakota	71	0.77

And a brief discussion:

The country doesn’t just display remarkable difference in beliefs about religious things like the devil; beliefs about foreign policy related facts also differ significantly across space. For example, a CBS/New York Times poll of April 2004 asked respondents, “Do you think Saddam Hussein was personally involved in the September 11, 2001, terrorist attacks on the World Trade Center?” Of the South Central region respondents (Texas, Oklahoma, Arkansas, Louisiana, and New Mexico), 45 percent said yes to this question, but only 25 percent of the Pacific Southwest respondents (California, Nevada, Arizona, and Hawaii) shared this belief. In the same poll, 60 percent of the South Central region respondents and 62 percent of the Mountains and Plains respondents (Colorado, Utah, Wyoming, Montana, North Dakota, and South Dakota) said that they think that “Iraq probably does have weapons of mass destruction that the United States has not found yet?” Only 43 percent of the Pacific Southwest and 40 percent of the Pacific Northwest respondents (Alaska, Washington, Oregon, and Idaho) shared this view.

These differences in beliefs within the United States drive home a central point about how politically relevant beliefs are formed. People in different states have been exposed to quite similar evidence through national media outlets, but they have come to radically different conclusions, and continue to hold these conclusions despite being aware that others disagree. This disagreement requires either different prior beliefs or some other deviation from Bayesian reasoning. One natural alternative model is that people base opinions mostly on the views of those around them. As such, local interactions are critical, and these provide plenty of possibility for wide geographic variation (as in Glaeser, Sacerdote and Scheinkman, 1996; Murphy and Shleifer, 2004).

Update -- a longer excerpt and many of the figures from the paper can be found at Economist's Voice -- the blog run by Mark Thoma (who was my freshman year Honors College macro prof at UO).

Next week, we will examine how beliefs are formed. Particularly, we will focus on building an economic model which will help us understand why certain beliefs become more common among certain people at certain points in time.

Before we get there, I thought I would spend some time outlining various findings on differences in beliefs across time and space.

Recall, at the start of the semester, I described the Bayesian view of belief formation which is the standard way economists assume that beliefs are formed. Bayesians argue that each individual starts with some sense of the probability that their beliefs are correct (their prior distribution). They go about their lives and receive new information that either confirms their existing beliefs or suggests that their beliefs are wrong. Based on how trustworthy the new information is thought to be, individuals update their prior belief distribution. Differences in initial beliefs, in access to new information, and in perceptions of the trustworthiness of the source of the new information generate differences in beliefs across time and space.

Bryan Caplan provides more discussion of the use of Bayes' Rule in economics in this post.

Yesterday, I talked about incorperating identity into the economic framework. The bulk of the work on this topic has been done by George Ackerlof and Rachel Kranton. The original paper can be found here. Here is the abstract:

This paper considers how identity, a person'’s sense of self, affects economic outcomes. We incorporate the psychology and sociology of identity into an economic model of behavior. In the utility function we propose, identity is associated with different social categories and how people in these categories should behave. We then construct a simple game-theoretic model showing how identity can affect individual interactions. The paper adapts these models to gender discrimination in the workplace, the economics of poverty and social exclusion, and the household division of labor. In each case, the inclusion of identity substantively changes conclusions of previous economic analysis.

A popular summary from the NYTimes can be found here. A taste:

Conforming to one's self-image, to social expectations, or to a combination of the two, is one of the important ways people seek to be happy - to "maximize utility" in economic jargon. But while economic models have no theoretical reason to exclude the concept, economists have rarely considered it.

"The standard methodology in economics is to say that people care about objective things, and they don't care about subjective things" like how a good woman or a good man or a good Hawaiian should behave, the economist George A. Akerlof said in an interview. Economists may consider goods that money cannot buy, like having friends, but they tend to ignore attitudes or states of mind.

Yet social norms "are things that people really have very,veryy strong views about," Mr.Akerlof said. "Not only do they think that they should do certain things and not do other things, but they also think that their friends and associates and relatives should do some things and not do other things. That's what you don't see in economists' utility functions."

From the "that research must have been really fun" file:

It seems that the more macho a man is — at least according to his hormones — the more the sight of an attractive woman will affect his judgement.

Researchers at the University of Leuven in Belgium asked men to play an ultimatum game, in which they split a certain amount of money between them. High-testosterone men drove the hardest bargain — unless they had previously viewed pictures of bikini-clad models, in which case they were more likely to accept a poorer deal.

The sight of flesh had less effect on the bargaining tactics of low-testosterone men.

I really love economics.

(P.S. for Dave B., the study was done by Bram Van den Bergh and Siegfried Dewitte. Does Jan know these guys? Did he get to particiapte in the study?)


A few other interesting findings in the same vein (all archived here):

Psychologists Margo Wilson and Martin Daly at McMaster University in Hamilton, Canada, asked students whether they would prefer, say, $19 tomorrow or $25 next week. They then showed them the faces of ladies whose attractiveness was ranked on the website 'How Hot Am I?'.

After eyeballing pictures of pretty women, men were more likely to want immediate monetary gratification than to wait for a bigger bonus. Women's calculations, on the other hand, were unaffected by male eye candy, as were men's after ogling plainer women or fancy cars.

With ladies fresh in their minds, men want money immediately because they might use it to impress and woo them, Wilson suggests. Such a trait might have been evolutionarily advantageous: men with more to offer might have been better able to get the girl.

Scientists reported last week that male rhesus monkeys will 'pay' to check out pictures of female monkey bottoms or images of socially dominant members of their species.

The insight into monkey urges was arrived at by researchers at Duke University in North Carolina who gave male macaques the choice of looking at images on a computer screen of either a female's posterior or of a socially-dominant monkey. They found that the monkeys would take a cut in their fruit juice allowance for glimpses of either alluring vision. As they report in Current Biology magazine, the pleasures of pay-per-view television appear to be shared by more than one species. However, when the monkeys were offered visions of a social inferior monkey they refused to look unless they were paid extra rations.

Since we discussed social conventions and the creation of social trends I thought I would send out a link to what will soon be the newest trend in America. So as many of you were probably not aware today was National High Five Day. http://www.nationalhighfiveday.com (I was two days off from my social experiment in high fiving actually being relevant.) I encourage you all to participate in this wonderful event so that we can swing the social trends in favor or national high five day being nationally recognized.

Update from Bryce -- The social experiment Varun refers to was his high fiving everyone on his walk from the River to the Quad on Tuesday. Unfortunately, on actual National High Five Day, he was asking people their favorite color.

Tyler Cowen's column in the NYTimes yesterday was about the interesting new field of neuro-economics. These researchers try and understand what's going on in the brain while people make decisions. While it is still questionable whether or not these studies are externally valid (i.e., do people think the same in "real world" situations), I still think this stuff is interesting. Here's a taste:

For instance, when humans are in a "positive arousal state," they think about prospective benefits and enjoy the feeling of risk. All of us are familiar with the giddy excitement that accompanies a triumph. Camelia Kuhnen and Brian Knutson, two researchers at Stanford University, have found that people are more likely to take a foolish risk when their brains show this kind of activation.

But when people think about costs, they use different brain modules and become more anxious. They play it too safe, at least in the laboratory. Furthermore, people are especially afraid of ambiguous risks with unknown odds. This may help explain why so many investors are reluctant to seek out foreign stock markets, even when they could diversify their portfolios at low cost.

If one truth shines through, it is that people are not consistent or fully rational decision makers. Peter L. Bossaerts, an economics professor at the California Institute of Technology, has found that brains assess risk and return separately, rather than making a single calculation of what economists call expected utility.

At the end of the article, Cowen suggests scanning voters brains. I don't know if that has been done, but there is an interesting study which shows that partisans don't rationally evaluate information which challenges their side. You can read about these fascinating (and troubling) results here. The test:

Researchers asked staunch party members from both sides to evaluate information that threatened their preferred candidate prior to the 2004 Presidential election. The subjects' brains were monitored while they pondered.

The results:

"None of the circuits involved in conscious reasoning were particularly engaged," Westen said. "Essentially, it appears as if partisans twirl the cognitive kaleidoscope until they get the conclusions they want, and then they get massively reinforced for it, with the elimination of negative emotional states and activation of positive ones."

Answering this questions if obviously very complicated. Two papers from the Journal of Economic Perspectives provide a few insights into the process.

The first, "The Economics of Convention" discusses the evolution of conventions like driving on the left or right side of the road. That is, it describes how, out of a set of potential choices without an obvious preferred alternative, societies improve welfare by converging on one standard convention.

The second, "Learning from the Behavior of Others, Conformity, Fads, and Informational Cascades" presents a model for conformity which explains how fads arise.

Following up on the earlier post on the fact that monetary incentives might lead to worse outcomes, a paper by Uri Gneezy and Aldo Rustichini further confirms that monetary incentives can actually reduce performance. However, they also find that larger monetary incentives do increase performance. The paper can be found here. Here is the abstract:

Economists usually assume that monetary incentives improve performance, and psychologists claim that the opposite may happen. We present and discuss a set of experiments designed to test these contrasting claims.

We found that the effect of monetary compensation on performance was not monotonic. In the treatments in which money was offered, a larger amount yielded a higher performance. However, offering money did not always produce an improvement: subjects who were offered monetary incentives performed more poorly than those who were offered no compensation. Several possible interpretations of the results are discussed.

Their results are based on two experiments. In the first experiment, students answered questions on a test which largely tested effort. All students were promised a fixed payment for participation; however, some students were paid an additional sum based on the number of questions they answered correctly. One group was given 10 cents per correct answer, another $1 per correct answer, and another $3 per correct answer. The group getting 10 cents per correct answer performed significantly worse than the group not getting any additional payment (and this group did worse then the $1 and $3 groups).

In the second experiment, the researchers introduced incentives to a public service day which occurs regularly in Israel. Apparently, certain days during the year students take to the streets to collect donations for various causes. Three groups of students were studies. In one group, no additional incentives were introduced. In another group students received 1 percent of what they raised. And in the other group they received 10 percent of what they raised. In both of the treatment groups it was made clear that the money was being paid by the researchers and was not coming out of the money that was being raised. Again, the no treatment group out performed the small incentive group, and in this experiment they equaled the high treatment group.

Over at Econball, my good friend Dave B posts the latest scheme to lower gas prices circulating via the internet:

"For the rest of this year, DON'T purchase ANY gasoline from the two biggest companies (which now are one), EXXON and MOBIL. If they are not selling any gas, they will be inclined to reduce their prices. If they reduce their prices, the other companies will have to follow suit. "

Quiz for economics students, what assumptions must be true for this strategy to actually work?

In class yesterday, we discussed how dense social networks promote cooperation (and discourage free-riding) by increasing observability and by increasing the threat of punishment (because people caught violating norms and/or free riding stand to suffer greater losses to their social capital when social networks are more dense). Ernst Fehr and Simon Gaechter are two of the big names in this area. A summary piece describing the results of their various public goods experiments that show that observability and enforcement promote cooperation in public goods experiments can be found here. As the abstract below points out, the paper also describes some additional experiments which show that formal contracts can reduce output by crowding out cooperative motivations (leading to a situation where employers actually prefer incomplete contracts).

Abstract: This paper shows that reciprocity has powerful implications for many economic domains. It is an important determinant in the enforcement of contracts and social norms and enhances the possibilities of collective action greatly. Reciprocity may render the provision of explicit incentive inefficient because the incentives may crowd out voluntary co-operation. It strongly limits the effects of competition in markets with incomplete contracts and gives rise to noncompetitive wage differences. Finally, reciprocity it is also a strong force contributing to the existence of incomplete contracts.

This is fascinating. Canadian Kyle MacDonald has, over the course of a year and several trades, managed to trade a red paper clip for one year of housing in Phoenix. His goal is to ultimately get a house. The internet really is remarkable. Honestly, I am supposed to understand markets, but I am not sure how to think about this. Do we believe that he could have sold the red paper clip on eBay, taken the earnings and bought something else (not on eBay), sold that on eBay, and so on and reached the same point? If not, what is going on?

In our discussion of social capital trends on Thursday, I failed to ask the two main questions that should have grounded our assessment of Putnam's arguments. I never explicitly asked whether or not the returns on individuals' investments in social capital have fallen and whether or not the costs of investing in social capital have risen? That is, we failed to completely discuss the trends in marginal benefits and marginal costs. Assessing these two factors is essential to determining social capital trends.

Putnam argues that several factors raised the cost of investing in social capital. Suburbanization and sprawl increased distances between people and raised travel costs associated with meeting others. Television, higher wages, and higher female labor force participation also substantially raised the opportunity costs associated with social engagement. (He also argues that a substantial portion of the decline stems from generational factors which I am not sure how to fit into the framework. I guess maybe different shocks changed the benefits associated with social engagement across cohorts.)

The trends Putnam discusses almost certainly affect social capital investments in the way he hypothesizes; however, observing trends which we expect reduce social capital does not imply that social capital declined overall. Other trends might offset these trends by lowering the costs or raising the benefits associated with investments in social capital.

In class, we discussed some of the ways in which communications technology changes the costs associated with certain social decisions. And while technology changes the composition of social networks and how people are tied together, it seems very likely that some of the increase in the cost of investing in social capital discussed by Putnam is offset by declines in the cost stemming from the use of technology.

There have also been significant changes in the expected returns to social interactions. Growth in service employment has likely raised the returns to social skills (and other forms of social capital) in the workplace. Specifically, office work requires large investments in social ties with one's co-workers and clients that needn't occur on an assembly line.

Further, to the extent that friendship is a normal good, we should expect that individuals demand more of it as their incomes rise. Also, if social interaction is complementary to other normal goods (like restaurant meals) we should expect higher incomes to increase the demand for social interaction.

Higher incomes, however, might reduce the need for favors and thus the need to invest in certain relationships (particularly with one's neighbors -- e.g., it used to be that Ed had the lawnmower and John had the leaf blower and they would borrow from each other, now Ed and John each buy their own stuff and needn't interact). Other economic changes like 24-hour markets and easily accessed emergency services also reduce the need to call upon neighbors for help (and thus the expected returns from interacting with them).

Returns from social networks also decline as a result of the development of more formal information signals. Investments in college degrees and high credit scores provide information that used to be informally stored and passed through social networks. When one's reputation (at least for certain things) is externally certified the need to make sure people know and trust you diminishes. Further, people may invest differently in their social capital because exploiting social networks to obtain jobs or capital has become semi-taboo (although it is still done extensively).

There are a number of other trends to consider, but the basic picture is starting to emerge. Social capital is more important, on average, in one's work and professional life than it used to be. However, in a number of contexts, market institutions have replaced social institutions substantially changing the returns to many forms of social investment and thus the composition of social capital. Such changes are clearly apparent in individuals' neighborhoods. Today, people rely on their neighbors for a smaller set of favors, and thus people are less inclined to "grin and bear" socializing with the random people who happen to live nearby. Instead, they choose to socialize with people who they know they like and who, because they have a great deal of shared interests, will involve them in very enjoyable (high return) activities and experiences.

These changes don't imply that individuals' stocks of social capital are smaller. People are likely to (and do) spend as much or more time engaged in social activities as they used to. They are likely to have as many or more social ties then they used to. However, who they are tied to, what ties them together, and the social skills necessary to make and maintain useful social ties have probably changed. Specifically, we have fewer ties to those who randomly share the same space. Thus Putnam may be right to worry. Many problems require collective action among neighbors and the coordination necessary to solve these issues is hindered by an absence of social ties. Specifically without norms of trust and reciprocity people are unwilling to compromise, skeptical of others motives, and lacking familiarity with certain social skills and manners that help govern the process.

While this specific trend may be undesirable, it is not obvious whether or not society as a whole is worse off. These bad effects may be more than offset by the good of being able to spend one's social time with people that you really like, etc. More empirical work is necessary to help clearly frame the full set of tradeoffs.

On Thursday, we discussed a variety of ways that technology changed social interaction. One of the technologies that we discussed was the cell phone. The Pew Internet and American Life Project provides some basic facts about cell phones and their effect on social connections in this report. Here are a few of the relevant findings:

The cell phone has become an integral and, for some, essential communications tool that has helped owners gain help in emergencies. Fully 74% of the Americans who own mobile phones say they have used their hand-held device in an emergency and gained valuable help.

Another striking impact of mobile technology is that Americans are using their cell phones to shift they way they spend their time. Some 41% of cell phone owners say they fill in free time when they are traveling or waiting for someone by making phone calls. And 44% say they wait to make most of their cell calls for the hours when they do not count against their “anytime” minutes in their basic calling plan.

The PEW center has a number of interesting data sets that might provide interesting information to examine for your term papers.

You're less likely to engage in criminal behavior:

Of course, seeing a correlation between fatty acids and nonviolence doesn't necessarily prove that fatty acids inhibit violence. Bernard Gesch, a senior research scientist at Oxford University, set out to show that better nutrition does, in fact, decrease violence. He enrolled 231 volunteers at a British prison in his study; one-half received a placebo, while the other half received fatty acids and other supplements. Over time, the antisocial behavior (as measured by assaults and other violations) of the inmates who had been given the supplements dropped by more than a third relative to their previous records. The control group showed little change.

Over at Marginal Revolution, Tyler Cowen muses about how the availability of TV on the internet will affect TV shows. Here's a taste:

1. Individual episodes are more complex and less likely to be self-contained. To watch only one show of Lost or BSG leaves you baffled. But who can make sure he catches every episode? What if you want to leave the country for a while? Now if you have missed a show, you can use the Web to keep in touch with the longer and more integrated story. You will do this even if you, like I, find web viewing distasteful and inconvenient. Not everyone can afford TiVo, and some of us still need Yana to operate the remote and indeed the service itself. This mechanism will raise the intellectual quality of TV.

2. Perhaps the time lengths of programs will vary more. Has The Sopranos gone on a nearly two-year hiatus? How about a fifteen-minute web shortie to keep us interested?

I think he has some interesting ideas. However, I am more interested in how the availability of shows on the internet and the availability of DVRs affect the decision to engage in social interaction. Putnam argues that television is one of the main causes of the decline in social capital. While, as we discussed in class, I am not sure the decline in social capital is as big as Putnam argues, TV certainly raised the opportunity costs of social engagements (particularly during prime time). It seems quite plausible that there is some non-trivial fraction of the population who would pass on the opportunity to do something social because they didn't want to miss an episode of "24". Thus, the introduction of DVRs (and now shows on the internet) reduces the opportunity costs of social activities during your favorite shows and thus should have an effect on the amount of social interaction we observe.

If you are interested in this topic, it might make for a good term paper. I am not sure what data is available on DVR use, particularly DVR use matched to information on social interaction. I also don't know, off the top of my head, what kind of exogenous variation in DVR use is available to exploit to produce cleaner causal estimates. We can look though. Further, if looking specifically at DVRs is not feasible, I think that a paper on social interactions and TV can still be written using the American Time Use Studies and TV re-run schedules (the basic idea being to infer the effect of new TV shows on social decisions by comparing time spent in social activities during weeks with mostly re-runs to weeks with new shows controlling for a variety of things).

Does media affect voting? Activists on both sides of the aisle claim that the media is biased, but does exposure to biased media affect how people vote? Another of my first year TF's, Stefano DellaVigna, and a co-author, Ethan Kaplan, examine the effect of the introduction of Fox News on voting for Republicans. DellaVigna and Kaplan argue that the timing of the introduction of Fox News in communities is largely random. This allows them to use a difference-in-differences approach to estimate the effects of exposure to Fox News on voting. That is, they compare the change in voting in communities that were "treated" with Fox News to the change in voting in "control" communities. Here is the abstract for their paper:

Does media bias affect voting? We address this question by looking at the entry of Fox News in cable markets and its impact on voting. Between October 1996 and November 2000, the conservative Fox News Channel was introduced in the cable programming of 20 percent of US towns. Fox News availability in 2000 appears to be largely idiosyncratic. Using a data set of voting data for 9,256 towns, we investigate if Republicans gained vote share in towns where Fox News entered the cable market by the year 2000. We find a significant effect of the introduction of Fox News on the vote share in Presidential elections between 1996 and 2000. Republicans gain 0.4 to 0.7 percentage points in the towns which broadcast Fox News. The results are robust to town-level controls, district and county fixed effects, and alternative specifications. We also find a significant effect of Fox News on Senate vote share and on voter turnout. Our estimates imply that Fox News convinced 3 to 8 percent of its viewers to vote Republican. We interpret the results in light of a simple model of voter learning about media bias and about politician quality. The Fox News effect could be a temporary learning effect for rational voters, or a permanent effect for voters subject to non-rational persuasion.

Update -- If you find this interesting, there may be a potential term paper here looking at the effect of the introduction of Fox News on campaign contributions to Republicans (and maybe extending things to look at the effect of the introduction of CNN). The empirics would be fairly simple (largely just doing the same stuff DellaVigna and Kaplan do), but the data work might be a bit tricky. I have worked with the FEC data, so I could provide you with much of the code that you would need. But there would likely still be a fair amount of data work.

I've been meaning to blog about this for awhile. Ray Fisman, Sheena Iyengar, Emir Kamenica, and Itamar Simonson set up a speed dating experiment among grad students at Columbia to examine dating. They test two broad sets of hypotheses. First, they test a variety of hypotheses about which traits people find desirable. They essentially regress whether or not I say, "yes, I'd like to see you again" on your attributes (like how attractive you are) in order to find out what traits, on average, people find desirable. Second, they exploit variation in how many partners individuals face to examine the effect of more choices on selectivity. That is, they regress the "yes rate" on the number of partners individuals had. Here's the abstract to the main paper:

We study dating behavior using data from a Speed Dating experiment where we generate random matching of subjects and create random variation in the number of potential partners. Our design allows us to directly observe individual decisions rather than just final matches. Women put greater weight on the intelligence and the race of partner, while men respond more to physical attractiveness. Moreover, men do not value women's intelligence or ambition when it exceeds their own. Also, we find that women exhibit a preference for men who grew up in affluent neighborhoods. Finally, male selectivity is invariant to group size, while female selectivity is strongly increasing in group size.

Fisman is interviewed about applying economics to dating here. A taste:

4. Do you think speed-dating is more efficient than traditional search methods?

In some sense, it’s efficient: there are all these slice studies on how 10 seconds’ worth of observation is as predictive of your experience with a professor as a semester’s worth, and they’ve reduced it to 2 seconds and that’s just as good; and they’ve reduced it to just a photo and that’s pretty good, too. So you learn a lot in four minutes, perhaps as much in four minutes as you do in a much longer superficial interaction like, say, a date. So, this does meaningfully provide you with 20 rapid-fire dates, to the extent that we form as much of an impression in 4 minutes, or 10 seconds, as we do in 4 hours. The thing that’s left out of this neat decomposition of people into attributes, though, is actually learning to love someone. And that’s what I think is kind of missing. Focusing on people as a bundle of attributes almost makes people think about this decision in the wrong frame of mind.

... in the UK at least:

"The most successful lines were ones which were impossible to answer with a simple yes or no, such as 'What's your favourite pizza topping?' and 'Who would you be if you were going on Stars In Their Eyes?'.

Professor Wiseman said at the other end of the spectrum, the worst chat-up lines included 'I have a PhD in computing' and 'My best friend's a helicopter pilot'.

One of the free movies "On Demand" currently, Sleepwalkers (1992), has this description:

Shapechangers – part vampire, part warewolf, totally evil – must nourish themselves by tapping the life force of teenage virgins – it gets messy when the teens don’t cooperate!

My Friday night just got much better.

... OK, so we're not actually going to watch it, but the description is pretty sweet.

At some point I mentioned Nisbett and Ross's book "The Person and The Situation" which challenges much of the research on personality based on evidence that people appear to quite clearly and consistently deviate from their "personalities" and respond to the situation at hand. While this book made me more skeptical of the study of personality, some interesting new research finds that people's scores on personality tests change depending on which language they are using. From a study of a bunch of bilingual Mexicans and Americans:

Participants twice completed a questionnaire gauging the 'Big Five' personality dimensions of extraversion, agreeableness, conscientiousness, openness and neuroticism -- once in English and once in Spanish. Across three separate samples, the researchers observed the same pattern -- when the participants completed an English version of the questionnaire, they tended to score higher on extraversion, agreeableness and conscientiousness, and slightly lower on neuroticism, compared with when they completed a Spanish version.

I really don't know what to make of this, but it is kinda interesting.

(Econometrics note -- This study uses individual fixed effects to identify the effect of language on personality. Just comparing individuals who speak Spanish to individuals who speak English does not produce credible estimates because the effect of language on personality cannot be separated from the effects of culture, etc. that typically are tied to language. To eliminate these concerns and more precisely isolate the effect of language only, the researchers compare the same individuals to themselves when they are speaking a different language. Econometrically, this is the same as including a dummy variable for each person in the sample. This dummy variable picks up all of the unobserved attributes of the individual that are fixed over time (e.g., their cultural background).)

Until relatively recently, people believed that you could tell alot about a person based on their physical features. Ok, we still probably believe that to some degree, but not in the same way. The formal study of character through features is known as physiognomy, and, back in the day, its practicioners believed that they could point out criminals based on their facial features. Bram Stoker was very into physiognomy, and, as such, Dracula possesses most of the features of the physiognomist's "criminal man" -- unibrow, high forehead, ... (I know all this because I wrote a paper about if for my Gothic Lit class as an undergrad.)

Anyhow, some economists are out to quasi-revive physiognomy. Naci Mocan and Erdal Tekin show that unattractive people are more likely to commit crimes. Here's their abstract:

Using data from three waves of Add Health we find that being very attractive reduces a young adult's (ages 18-26) propensity for criminal activity and being unattractive increases it for a number of crimes, ranging from burglary to selling drugs. A variety of tests demonstrate that this result is not because beauty is acting as a proxy for socio-economic status. Being very attractive is also positively associated adult vocabulary test scores, which suggests the possibility that beauty may have an impact on human capital formation. We demonstrate that, especially for females, holding constant current beauty, high school beauty (pre-labor market beauty) has a separate impact on crime, and that high school beauty is correlated with variables that gauge various aspects of high school experience, such as GPA, suspension or having being expelled from school, and problems with teachers. These results suggest two handicaps faced by unattractive individuals. First, a labor market penalty provides a direct incentive for unattractive individuals toward criminal activity. Second, the level of beauty in high school has an effect on criminal propensity 7-8 years later, which seems to be due to the impact of the level of beauty in high school on human capital formation, although this second avenue seems to be effective for females only.

Notice how their last two sentences describe clear economic hypotheses for what might explain this relationship. While I don't find it hard to believe that there is a relationship between appearance and crime, I wonder about a couple of things. First, these authors use the same study that you all used for the empirical exercise, so it is possible that interviewers perceived their "criminality" in their demeanor or dress and rated down their appearance. Second, even assuming that interviewers are able to "objectively" rate people's appearance, beauty is only partially exogenous. Individuals can, and do, invest time, money, and effort moving themselves around the beauty distribution. It may be that the traits which explain investments in appearance also explain low test scores, low wages, and/or higher crime.

I may be the only one interested in this, but two of my TFs from first year, Liran Einav and Leeat Yariv, have written a paper documenting that economists whose last names are towards the end of the alphabet are less likely to be tenured at top departments, are less likely to be members of the econometrics society, and are less likely to write multi-authored papers. I have long wondered about this stuff (and I figured that someone must have written about it). In economics, the convention is to list authors alphabetically. This means that those of us at the end are frequently not tied to our work (lost into the "et al." abyss or, at the least, an after thought). Interestingly, in psychology, where alphabetical listing is not the convention, they do not find a last name effect.

In his new piece in Foreign Affairs, "Fear of offshoring," Princeton economist Alan Blinder says really smart things. In this piece, Blinder outlines, for a general audience, the basic economics of off-shoring. He tries to address people's fears of offshoring by discussing how economies (and individuals) adjusted to the First Industrial Revolution (the movement from the farm to factory) and the Second Industrial Revolution (the movement from factory to services) before discussing the coming Third Industrial Revolution (the information age). This is an excellent piece, and there was something "excerpt worthy" on nearly every page, but I will try and focus on the stuff which is likely most relevant to this class.

The big change Blinder discusses that is likely to accompany the Third Industrial Revolution is that more workers in the service sector are likely to face competition from abroad. Specifically, we should probably start to think of service jobs in terms of those which require face-to-face interactions (personal services) and those which don't (impersonal services). Anything which can be efficiently transmitted electronically will now face more competition. These jobs, however, don't break down neatly into educated and non-educated (they way we now tend to think of "good" and "bad" jobs). As Blinder states:

It seems to me unlikely that the services of either taxi drivers or airline pilots will ever be delivered electronically over long distance. The first is a “bad job” with negligible educational requirements; the second is quite the reverse. On the other hand, typing services (a low-skill job) and security analysis (a high-skill job) are already being delivered electronically from India--albeit on a small scale so far. I could go on and on. Most physicians need not fear that their jobs will be moved offshore, but radiologists are beginning to see this happening already. The work of policemen will not be replaced by electronic delivery, but the work of some security guards will be. (My home burglar alarm is monitored from somewhere in Indiana. Why not from somewhere in India?) Janitors and crane operators are probably immune to foreign competition; accountants and computer programmers are not.

(And to further show how far this can go, some McDonald's now operate their drive thru windows from call centers). For those worrying about your career path, Blinder goes on speculates on how much technology is likely to affect different types of service industries.

Ultimately, after discussing how big these changes are likely to be, Blinder discusses how to adapt to the changes. Education obviously plays a role, but not in the way we typically think -- people skills become more important when the jobs protected from foreign competition (an thus likely to earn higher wages) are concentrated in personal services. He states:

As I indicated earlier, simply providing more education is probably a good thing on balance, especially if a more educated labor force is a more flexible labor force that can cope more readily with non-routine tasks and occupational change. Frank Levy and Richard Murnane have noted that, in the computer age, “the ability to apply well understood routines to solve problems is not as valued as it used to be.” But they believe that “rapid job change raises the value of verbal and quantitative literacy,” and so they prescribe more education. But education is far from a panacea, and the examples given earlier show that the rich countries will be able to retain many jobs that require little education. In the future, how we educate our children may prove to be more important than how much we educate them, as Simon’s words suggest. But educational specialists have not even begun to think about this problem. They ought to start--right now.

Perhaps, contrary to what we have come to believe in recent years, people skills will become more valuable than computer skills. The geeks may not inherit the earth after all, at least not the highly-paid geeks in the rich countries. (Geeks in poor countries should be in great demand.) Certainly, creativity will be prized. Thomas Friedman has rightly emphasized the desirability of steering our youth away from tasks that are routine or routinizable into work that requires real magination. In his colorful words (Friedman (2005), p. 239):

"You want constantly to acquire new skills, knowledge, and expertise that enable you constantly to be able to create value—something more than vanilla ice cream. You want to learn how to make the latest chocolate sauce, the whipped cream, or the cherries on top, or deliver it as a belly dancer—in whatever your field of endeavor."

I agree, but there are two big problems. First, creativity and imagination are notoriously difficult to teach in schools—although, in this respect, the United States seems to do have a leg up on, say, Germany or Japan. Second, it is hard (for me at least) to imagine that truly creative jobs will ever constitute anything close to the majority of employment. Never in the history of humankind have belly dancers outnumbered drones. What will everyone else do?

Blinder goes on to discuss other ways to cope with changes. It is all very good. If you are at all interested in this stuff, I recommend checking it out.

Starting on Thursday, I want to begin discussing the long term trends in social capital and the role of social capital in aggregate outcomes. In order to prepare you for this discussion (and in generally increase your familiarity with social capital research), I want you to read some summary pieces from the Saguaro Seminar website.

First, go here and read the Social Capital FAQ. This will provide you with the highlights of the debate.

Second, go here and read the summary of social capital research -- which seems to be laregly about the impact of diversity, so it is pretty relevant given the current debate over immigration.

Finally, go here and see a variety of papers related to social capital -- and start thinking about your term papers. You don't need to write about social capital, but you should start thinking about writing about something. Then, you should set up a meeting with me to discuss you ideas.