Category: Economics

After spending a weekend with 11,000 economists at the American Economics Association 2015 meeting, here’s my perception of the Big Picture…

The two most important questions on which economists disagree are the following:

• Can higher education make a person more productive at his or her ultimate job?
• How much of a society’s resources should be put into reducing CO2 emissions?

Why is the economic value of higher education so important to establish? An assumption driving much of the debate regarding inequality is that a worker with a college degree is more productive and therefore will, in a properly functioning market economy, receive a higher wage than a worker without a college degree. Those who propose radical action to address inequality note that an increasing prevalence of college degree holders has not resulted in an increase in inequality. From this they infer that capital is stealing from labor, specifically by not paying workers for the increase in their marginal productivity as predicted by Econ 101. A possibility that the Big Thinkers don’t seem to consider is “Maybe colleges aren’t teaching anything of value to employers?”

This blind spot is curious because there is a fair amount of evidence that many American college graduates learned little during their four-year sojourn. The book Academically Adrift, for example, cites data from a Collegiate Learning Assessment test showing that many students don’t improve much from freshman to senior year. Studies on students who were admitted to elite schools, such as Harvard, but elected not to attend have found that there was little difference in lifetime income attributable to actually attending the elite schools (though being qualified for admission had a lot of value). There were some interesting additions to this literature at the conference.

Despite the evidence that people who weren’t academically inclined prior to college get little benefit from college and simultaneously suffer four years of lost income, Big Thinker-driven public policy has resulted in a trough being filled with federal tax dollars and a large group of non-selective colleges feeding from that trough. For-profit online schools, such as University of Phoenix, have been particularly aggressive feeders, with roughly 75 percent of their revenue coming from this federal source. “An Experimental Study of the Value of Postsecondary Credentials in the Labor Market” (Deming et al) sent out 10,492 fake resumes to employers on an “online job board” (presumably monster.com) and found that, for jobs that did not explicitly require a college degree, there was little increase in the chance of being contacted by virtue of having a college degree. In other words, the public and private investment in college might well be worthless for these entry-level positions. For jobs that did require a degree, an online degree was about 22 percent less attractive to employers than a “non-selective” bricks and mortar school, e.g., Cal State. For jobs that required a license that itself entails an exam, such as practical nurse or pharmacy technician, employers didn’t seem to care about anything other than whether or not the applicant was licensed. (This suggests that if colleges stopped grading their own students, as proposed in my “Universities and Economic Growth” article, and every college graduate took a comprehensive exam, employers might see a lot of value in those graduates who had scored well on a neutrally graded exam.) An author of the paper summarized by noting that the lowest level of employer interest was in resumes of graduates of the most rapidly growing and most expensive sector of higher education.

Academic attitudes toward business were on display in this session. Karl Marx had sympathy for the employer, constantly at risk from competition and incorrect estimates of demand. Today’s economists have mostly rejected Marx and apparently have abandoned this sympathy. Nobody raised any questions about whether it was ethical to waste employers’ time with 10,000+ fake resumes. When the authors presented their conclusion that there was no difference in employer interest correlated with sex or race, the audience was flummoxed. As racism and sexism were assumed to be high priorities for America’s employers, how to account for this result? Nobody was willing to ask “Could it be that these employers just want to make money and don’t care whether they make money with black male workers, white female workers, or any combination?”

Justine Hastings, of Brown University, presented “Earnings, Incentives and Student Loan Design: The Case of Chile.” It seems that Chile did what the U.S. did, i.e., offered a lot of student loans for higher education. Their program was more intelligently designed, however, in that they didn’t allow universities to raise tuition in response to this new source of funds. Schools ended up with more students, but not more money per student as has been prevalent in the U.S. Nonetheless, the default rate has been high, especially for graduates of non-selective schools and especially for those who majored in humanities and arts. Unlike Americans, Chileans don’t like to keep flushing cash down the toilet, so now they are experimenting with adjusting the maximum loan amount according to the expected return to getting a particular degree (in Chile you don’t apply to “University of Santiago” you apply for a specific major). It turns out that when students see that the government won’t lend them the maximum for a particular degree program they get the message and try to switch into a degree that will result in higher post-graduate earnings. This is especially true for “low SES” students. SES? Due to the rejection of Marx, mainstream economists apparently can’t talk about class so they refer to “Socioeconomic status“. Hastings has a separate paper “The Labor Market Returns to Colleges and Majors: Evidence from Chile” with the discouraging result that attending a lower quality college and majoring in poetry will not set the country’s employers on fire and, in fact, many people would have higher lifetime earnings if they refrained from attending college.

Amanda Pallais of Harvard presented “Leveling Up: Early Results from a Randomized Evaluation of Post-Secondary Aid”, a paper on the Susan Thompson Buffett Foundation scholarship for lower income Nebraskans who have a high-school GPA of at least 2.5 and maintain a college GPA of at least 2.0. It turns out that people who are going to attend college and graduate will do so even without this grant and people who were marginally attached to academic will become only slightly more attached. The cost of keeping one student in college for an additional semester is $40,000 of foundation funds.

Gregory Clark, in my opinion the best of the Big Thinker economists out there, pointed out that when looking at the cost and benefit of programs to get more people into college, the correct approach is to focus on that last person who, without the program, would not have attended college but with the program will. If the objective of the program is to spur economic growth or reduce inequality, the question must be “How much more attractive will that person on the margin be to employers once graduated? (And how likely is that person on the margin to graduate.)”

Economists generally seem to divide into two camps on the question of mass higher education. Some economists believe that good college students are born, not made. If you weren’t born with a talent for math and studying college can’t help you. Others believe that you can send almost anyone to college and they’ll come out with more “human capital” that employers will pay for. My personal view is that traditional American K-12 and college are reasonable matches for some people, e.g., those who love to sit still and do what a teacher tells them to do. The remainder, however, should not be regarded as discards. They could probably learn the same material if it were presented in a different way. (This view is partly informed by research finding that lectures are extremely ineffective for student learning and they are the cornerstone of K-12 and universities, partly by my experience teaching lab courses and seeing how quickly all kinds of people can learn to program, and partly by experience as a flight instructor where I have had no difficulty teaching qualitative physics to people from all walks of life.)

To sum up, the question of education is critical to the question of whether we should consider the kinds of drastic changes to tax policy that are proposed by Piketty and followers. If K-12 and college are making ever-better American workers then capitalists are using their class power to steal from the working class (potential exception: globalization has made a lot more workers available, thus reducing the market-clearing wage). If on the other hand we are to believe the test scores, the problem is simply that engineers keep designing better machines (capital) while our education system turns out workers no better than those of 30 years ago (but at a vastly higher cost in dollars and time).

Turning our attention to the question of climate change… Energy expenditures are about 8 percent of U.S. GDP. (Compare to 8.4 percent for finance (WSJ), which means that we pay as much to banks than to run our cars, heat and cool our houses, smelt aluminum, and spin our hard drives!) If we do something big in energy it will mean cutting back in other areas, e.g., shoveling out money to universities.

The talk that I was most drawn to was by Juliet Schor, a Boston-based economist who has addressed the question of “Why do Americans work so many hours?” in popular books, starting with The Overworked American: The Unexpected Decline of Leisure (1993). She noted that the 1990 standard of living was roughly double that of 1950. Why wouldn’t Americans all work half-time instead of enslaving themselves to buy more stuff? (amazon.com hadn’t even launched at the time!) Schor’s work starts from the presumptions that there are declining marginal happiness returns to earning more while extra leisure time makes people happier. She also implicitly relies on my theory that buying expensive organic locally produced food accelerates environmental damage (due to higher spending). Now that the planet is melting, says Schor, what better time to consider a coordinated reduction in hours so that our economy stops growing and therefore our contribution to global warming is reduced. We’ll all be happier. [sidenote: From Schor’s talk, I learned that Maslow’s hierarchy of needs, a required subject for FAA certificated flight instructors, has been discredited by academics.] Schor proposed a $1.2 trillion annual carbon tax on American industry to be distributed per capita to the American people. This would, she noted, have a salutatory effect on income inequality as well.

For each paper there was supposed to be a “discussant” that provided a critique. Schor, however, got a cheerleader in the form of Frank Ackerman, who teaches in the MIT Department of Urban Studies. He said that it is obvious that we need an immediate massive carbon tax and said that there is a 6:1 ratio in per capita carbon emissions among U.S. states (California: good; Texas: bad; this government chart shows that, excluding a couple of barely populated outliers, states actually fall into a pretty narrow range and nearly all of the variation is due to industry, not consumers). Germany overall emits only half the carbon per capita that we do, according to Ackerman. Why haven’t Americans taken drastic action? Ackerman says that it is because a majority of Americans are (1) stupid (anti-science/Tea Party), (2) racist (oppose CO2 action because action is supposed by Barack Obama, whom they identify as black), (3) stupid (because they see an opinion advertisement from Exxon and don’t recognize the likely bias).

I asked “Do these people you posit all have to be stupid and racist? Isn’t it possible that Americans are simply selfish and like their SUVs, McMansions, etc.? Or that they recognize that this is going to be a big government project and they are skeptical because a lot

Full post, including comments

Bostonians:

The American Economic Association is holding its annual meeting in Boston, January 3-5, 2015. The event is very reasonably priced for non-members and is packed with interesting speakers. One of the good things about economics is that many papers are understandable by anyone with a good high school math background.

Saturday, January 3

Saturday at 0800 is going to be an exciting time. Gregory Mankiw, the Harvard professor who sometimes steps out into the popular realm (example New York Times article) leads “A Discussion of Thomas Piketty’s Capital in the 21st Century.” (Note that a 10:15 paper, by two Swedish economists looking at “socio-economic mobility across three generations in Sweden in the period 1813-2010” found that, for earnings, there was “no association at all between grandfathers and grandsons” (so Piketty is wrong about dynastic accumulation), but for social status/class there was a “clear association” (leading to genteel poor?).) Simultaneously in the Hynes Center, Room 201, Maya Rossin-Slater and Miriam Wust present “Parental Responses to Child Support Obligations: Causal Evidence from Administrative Data” (full text). This is important because child support systems are designed under the presumption that parents’ behavior won’t be affected by writing child support checks to the person who sued them and/or that paying $50,000 per year to a plaintiff will improve the quality of parenting delivered every other weekend (embedded in a larger system of custody and divorce law that assumes people won’t follow economic incentives that are held out). Here’s how we summarized the paper in our forthcoming book:

[the authors] found that what a mother might have gained financially from child support, the child lost in terms of reduced contact with and effort from the father: “mothers, who have substantial say in custody decisions [in Denmark], have the opposite incentive to refuse to share custody and instead receive the higher payment [for child support, compared to shared custody]. … fathers may treat financial transfers as substitutes for other forms of non-pecuniary investments and contact with children, which would also lead to a negative relationship between child support obligations and father-child co-residence.” The economists found that “an increase in the father’s obligation may lead to less attachment to his existing children and more time available to invest in new offspring.” (See the “Divorce Litigation” chapter for our interviewees’ perspective on how the main opposed interests in a divorce lawsuit are the plaintiff parent and the children, not the plaintiff and adult defendant.) The researchers also found that fathers who were ordered to pay more child support were more likely to have new children, thus diluting the time and energy available to prior children, and that fathers who were ordered to pay more child support reduced their working hours due to “market distortions generated by the ‘tax-like’ nature of child support mandates.” Mothers who received more child support cash for existing children were motivated to have additional children, either with or without a live-in partner: “mothers receiving higher child support payments for current children may expect higher transfers for future children if they separate again.” Note that this research was done with data from Denmark, where child support is tax-deductible and capped at $8,000 per year. The effects that they observed would presumably be larger in the U.S. where child support payments are not tax-deductible and can be $25-100,000 per year.

The authors also question an additional building-block assumption of the U.S. child support system. The government assumes that a typical woman who has custody of a child and cannot meet all of her own expenses from child support revenue will, rather than work, turn to the various taxpayer-funded welfare programs offered to women with children. Thus a dollar of child support extracted from a custody lawsuit loser is a tax dollar of welfare saved. Rossin-Slater and Wust say that it isn’t so simple: “Our results suggest that although child support mandates may shift some of the cost of single-mother household support from welfare programs to the non-custodial fathers, they also pass part of this cost on to other government programs such as disability insurance and early retirement.” (i.e., fathers in Denmark will retire or declare themselves disabled in order to cut their child support obligation from $8,000 per year back to the “basic/minimum” $2,000 per year)

If you’re a shareholder in a public company that underperforms the S&P 500 and want to know why most of what would have been your dividend money nonetheless goes to an apparently mediocre CEO, there is a “CEO Incentives and Compensation” session at 10:15 am.

Many sessions seem to be aimed at trying to understand the Collapse of 2008 and predict the next one. Also at 10:15 am on Saturday is a session looking into why the ratings agencies turned out to be worthless. “Did Government Regulations Lower Credit Rating Quality?” by Behr, Kisgen, and Taillard, concludes that “defaults and other negative credit events are more likely for firms given the same rating if the rating was assigned after the [1974] SEC action compared to before. … the market power derived from the SEC led to ratings inflation.” Did Jack Nicholson’s sponsorship of a child in About Schmidt (awesome movie, though very different from the novel; disclosure: my cousin Harry produced it) make a difference? At 10:15 session “Developing Hope: The Impact of International Child Sponsorship on Self-Esteem and Aspirations” says “yes” (on average). History buffs will skip all of these 10:15 sessions and head to the Boston Marriott Copley for a session on “The Economy of Ancient Israel.” The “dismal science” does not shy away from looking at war. At 10:15 session on “conflict and development” includes a paper by David Yanagizawa-Drott that villages that suffered the most violence in the Rwandan genocide had the highest living standards six years later: “These results are consistent with the Malthusian hypothesis that mass killings can raise living standards by reducing the population size and redistributing productive assets from the deceased to the remaining population.” At “Health Insurance Reform” session confirms the world’s most stuffed-with-cash health care system is not going to suffer any lean years due to government tweaks. A “Competitive Bidding in Medicare” paper concludes that there is little actual competition among America’s health insurers. A session on “The Minimum Wage, Family Income and Poverty” reveals that economists have no idea whether or not a higher minimum wage reduces poverty.

Are those crazy Super Bowl ad rates justified? A paper in a 12:30 pm Digital Media Economics session concludes “yes” by looking at movie ticket sales (for movies advertised during the Super Bowl) in the cities in which Super Bowl teams are based and therefore in which more fans are watching. Why is the U.S. a more tolerant society today than it was in the 1970s? A paper by Berggren and Nilsson in the 12:30 “Economic Freedom and Minority Groups” session suggests that it might be due to the deregulation started by Gerald Ford and the tax rate reductions started by Ronald Reagan: “We suggest, as one explanation, that a greater scope for voluntary transactions and private usage of incomes and wealth creates more meetings that increase understanding for people different than oneself – or at least for the value of letting people different than oneself have their say.”

At 2:30 there is a session on “Explaining the Energy Paradox.” Why do Americans buy fuel-inefficient cars and houses? Why has no American residential developer carved out a niche in German-style double-wall houses that can be heated or cooled for almost nothing? Why did the Federal Weatherization Assistance Program not work? (“overly optimistic engineering estimates of returns to the energy efficiency investments,” say the economists; translation: Americans are stupid). “Investor Behavior,” at the same time, explores the 2012 8X increase in the value of shares of DI Corp. “because the company’s chairman and CEO is [South Korean rapper] PSY‘s father.” (translation: investors worldwide are stupid; related: MIT Gangnum Style) A 2:30 session on “Redistributive Taxation” includes “Income Inequality Influences Perceptions of Legitimate Income Differences” by Harvard’s Kris-Stella Trump. She finds that people think the system under which they live is a good one: “When income differences are (perceived to be) high, the public thinks of larger income inequality as legitimate.” She calls this “system justification motivation.” [We found this to be true when interviewing divorce litigators in different states. Generally a litigator in State X thought that State X’s system was fair and just and so did a litigator in State Y, despite the fact that State’s X and Y had completely different outcomes for the same fact patterns.] Larry Summers talks about “Secular Stagnation” in a 2:30 pm session as well. “Do Vehicle Crash Tests Save Lives? Impacts on Market Decisions and Accident Mortality” by Damien Sheehan-Connor of Wesleyan says that Americans bought safer cars in response to Insurance Institute for Highway Safety tests and manufacturers designed safer cars in response to the testing (translation: Americans are not stupid). Why do venture capital firms underperform the S&P 500? Three Harvard researchers in “The Cost of Friendship” ” find that venture capitalists who share the same ethnic, educational, or career background are more likely to syndicate with each other. This homophily reduces the probability of investment success, and the detrimental effect is most prominent for early-stage investments..” Why do hedge funds underperform the S&P 500? “Recovering Managerial Risk Taking from Daily Hedge Fund Returns: Incentives at Work?” by Kolokolova and Mattes finds that “During earlier months of a year, poorly performing funds reduce their risk. The risk reduction is stronger for funds with higher management fees, shorter notice period prior to redemption, and recently deteriorating performance, which is consistent with a managerial aversion to early fund liquidation and loss of future management fees. Towards the end of a year, poorly performing funds gamble for resurrections by increasing risk.” Why does money invested in private equity (even with geniuses such as Mitt Romney at the helm!) tend to underperform the S&P 500? Korteweg and Sorensen say that “Based on past performance alone, an investor needs to observe an excessive number of funds to identify the PE firms with top-quartile expected returns, implying low investable persistence.”

The 2:30 slot includes a “Puerto Rico and Cuba” session that would no doubt be enlivened by a panel discussion about the recent political changes. The papers already scheduled show that Puerto Rico’s government policies starting around 1940 have led to the county becoming impoverished from an income point of view: “After seven decades of government sponsored development efforts, a benign macro environment compared to the rest of Latin America, and massive transfers from the mainland, GNI per capita remains at Uruguayan or Argentinian levels.” On the other hand, presumably due to federal welfare programs, consumption per capita is quite high: “Puerto Rico has succeeded in ensuring a standard of living for its citizens largely divorced from the productivity of its workers.” For actual divorce issues, there is a 2:30 pm “Structural Models of Family Interactions” session that looks at “Welfare Effects of Divorce Legalization” in Chile and “Deadbeat Dads” (trying to answer the question of why women would get pregnant with low-wage fathers). “Unemployment Insurance and Disability Insurance in the Great Recession” calls into question the assumption that Americans go on SSDI when their unemployment insurance runs out: “Only 28% of SSDI awardees had any labor force attachment in the prior calendar year, and of those only 4% received UI income.”

Is it all about the Benjamins? A 2:30 session “Well-Being: Measurement and Policies” suggest that maybe we can move “Beyond GDP”. Weina Zhou’s study of Chinese youths “sent down” for hard manual labor actually ended up doing better as 40-55-year-old adults and “these findings are robust against a variety of family backgrounds.” (time to plant some daffodil bulbs!) A “Cycling to School” paper finds that giving girls bicycles was “much more cost effective at increasing girls’ secondary school enrollment [in India] than comparable conditional cash transfer

Full post, including comments