Friday, August 30, 2013

Tax attractiveness

Tax competition is good and bad. It is good because it forces governments to run lean operations and minimize distortions, but it is also bad because small governments can undercut larger ones and increase their own revenue while reducing aggregate well-being. Some firms and individuals are quite responsive as to where they should set up shop, so it could be useful for them to figure out where the best tax conditions are offered.

The latest attempt at figuring this out comes from Sara Keller and Deborah Schanz. They build a composite index from 16 indicators relevant for company taxation. While one can quibble about the weights but on each indicator, their normalization, and some of the redundancy, there is still useful information in the results. While it should be no surprise that the Caribbean is the most attractive area for business taxes, it is more surprising to see that Europe is significantly more attractive than North America. In fact, the United States are among the least attractive of the 100 countries Keller and Schanz considered (only Argentina, Indonesia, Peru, the Philippines, South Korea and Venezuela are worse, and Zimbabwe is in the sample...). I wonder, though, how much this would change by considering South Dakota and Delaware separately.

Thursday, August 29, 2013

More financial education could lead to more market instability

At least part of the blame for the last crisis was put on the lack of financial literacy of some borrowers who accepted mortgages they could not reasonably pay back. More generally, the lack of financial literacy is blamed for the widespread less than optimal funding for retirement and for excessive fluctuations in asset prices, a prime example being gold.

Mario Padula and Yuri Pettinicchi use some theory to understand how financial literacy can have an impact of markets, in particular market fluctuations and instability. In their model, one suffers some dis-utility cost from becoming more financially literate, but this lowers the cost of buying a more precise signal. The policy variable is the effectiveness of becoming financially literate. Beyond the utility/signal trade-off, the model highlights important general equilibrium effects. Indeed, once it becomes too easy to be literate, people stop buying signals because the market advantage of the signal has vanished: too many people share the information, and the benefit from asymmetric information is too small. The proportion of well-informed people is U-shaped as financial literacy is more easily accessible, as so is market volatility. Cheap information may not be worth acquiring.

PS: it looks like the paper has been removed from IDEAS. It is, however, still available here (pdf)

Wednesday, August 28, 2013

Are there biases from monetary rewards in experimental economics?

It has become very fashionable to run economic experiments in the laboratory. Typically, undergraduate students are lured into the lab with some monetary rewards. A longstanding question has been whether this leads to a selection bias that renders the experiment results impossible to generalize. One paper I previously discussed (here) finds no bias within students, a second (here) worries that students in developed economies are not representative at all.

Johannes Abeler and Daniele Nosenzo add another bit of evidence. They invited students to an experiment, either by offering money or not, and either by appealing to the usefulness of research or not. First, the authors observe that appealing to money is much more successful than appealing to research, it triples the number of respondents. Thus, given that participants care more about money, we may think that they would have different characteristics compared to the others. That turns out not to be the case. Thus, no bias from monetary rewards, at least within the student population.

Tuesday, August 27, 2013

Five universal laws for economics

Physicists believe that social sciences can only be described as true sciences if on can figure out some laws that always apply, without exceptions, and if there some invariant constants that would be good, too. Social scientists do not believe this is the right approach, foremost as one has to deal with individuals and societies that make choices.

James Wayne realizes that Physics lacks one ingredient that is essential in social sciences: choice. Fundamentally, I am not convinced that we actually choose, but that it only looks like it at the level of abstraction that we can master at this point and for the foreseeable future. Indeed, our decisions are the results of complex chemical reactions in our brains under the influence of complex environments and likely some randomness. But we have found a simpler abstraction with the framework of choices under constraints, and that is certainly missing in Physics.

Now Wayne adds the concept of choice to Physics, and then determines five new Physics laws: 1) the outcome of any future event is indeterministic; 2) there is a joint probability of future events that helps predicting them; 3) actions can be taken at any time to change this distribution; 4) we cannot retain complete information about past histories; 5) eventually, some equilibrium is reached. He can then use these new laws to re-understand natural and social sciences under a unified framework. All this in only 8 pages of text and not a single equation. True science at work.

Friday, August 23, 2013

Ethnic ghettos and unemployment

Both in Europe and the United States, minorities face significantly higher unemployment rates. In addition, they live in places that are farther from work than others, or at least their commuting options make it more difficult to get to work. Are the two linked? Obviously, if you do not live where the jobs are, unemployment gets more prevalent. But one could also move, and this may be more difficult for minorities, for various reasons. But before going there, one needs to determine how much of the unemployment rate is due to this spatial mismatch.

Laurent Gobillon, Peter Rupert and Etienne Wasmer pick up on a previous paper of the latter two, which I discussed here. In this spatial search-and-matching model, commuting time acts as a friction, but can only explain a fraction of the unemployment rate gap between "majorities" and "minorities". So other factors are clearly at play. The fact that minorities are de facto confined to particular areas certainly plays a role here.

Thursday, August 22, 2013

A look at faculty workload

A common complaint about teachers is that they have too much vacation time. Such complaints are even louder for university faculty, as the academic calendar specifies even shorter teaching times, and on top of this the weekly class room hours are ridiculously low. These complaints emerge because teaching is the only face time university faculty have with the paying public. We do a lot of other things that the tax payer does not see and in particular does not realize how much time it takes. But how much do university faculty actually work?

Manuel Crespo and Denis Bertrand have analyzed surveys distributed to faculty of a "Quebec research-intensive university." Using results from 130 tenured faculty who agreed to spend significant time thinking about there use of time, the average workweek is 57 hours. That takes into account that there are parts of the year where workload is lighter (summers) and other times where there more to do. Only about a third of the time is dedicated to research, which I find surprising as this is supposed to be a research university. 44% of the time, or 25 hours, are dedicated to teaching, a surprisingly low 3 hours a week to administration and 9 hours a week to "public service" (would my blogging count?). The report goes through more details, some of which I want to highlight: only 10% of time related to teaching is actually in the classroom. The rest is mostly preparing for classes, face time with individual students, and grading. Time spend on teaching has increased over a decade, attributed foremost to increasing class size (I do not think there is much value to this result, as faculty also got older and in some cases tenured). And there are very few gender differences in time allocation.

Wednesday, August 21, 2013

Why is the agricultural sector still protected in the US?

While most industrialized economies are strong advocates for free trade, they somehow manage to make an exception for the agricultural sector. I cannot think of one country that would not subsidize its farms in some major way and this despite the facts that we are far from witnessing a food shortage and that the agricultural sector by now constitutes a voting block that can safely be called small. So why do elected official keep pandering to farmers?

Marc Bellemare and Nicholas Carnes look at this question in the case of the United States. Using roll call votes from the US Congress and congress member ratings by the Farm Bureau, they find that electoral incentives in fact still do matter, while personal preferences and lobbying are less important. Could it be that the median voter is a farmer? I do not think this is what the paper is saying. Indeed, the variable "electoral incentive" is based on the proportion of the electorate that works in the agricultural sector. But it is above half, or approaching this, in few districts. In fact, Bellemare and Carnes restrict the variable to farm owners and managers, who are the ones really benefiting from the subsidies, and they comprise a tiny portion of the electorate in every district. (Of course, this measure is correlated with the total farm population.) For this influence of such a minority to still carry the vote, it must be that there is still lobbying going on, and of the sort that is not captured by the agricultural political actions committees that the authors use to control for lobbying. Maybe individual donations? Somehow, it always boils down to lobbying in the United States.

Tuesday, August 20, 2013

Height and labor market outcomes among twins

Yesterday, I mentioned some research on the link between height and labor market outcomes. Let me continue on the theme. It is not entirely clearly why height would matter. There are many confounding variables that could come into play. One way to sort some of them out is to use twins. Unlike many other twin studies, in this case you want to use twins who have not been separated, thus they have the same genetic material and life environment. They may thus only differ by height, social skills and cognitive abilities.

Petri Böckerman and Jari Vainiomäki use such data from Finland and find that there is no significant height premium for employment. However, looking at total income over a 15-year period, they find a premium for women, but not men. For the latter, it must thus be social or cognitive skills only that differentiates them. For women, the authors conclude that it must be discrimination that leads to the height premium, as health variable do not seems to explain it. I wonder whether this has to do with the fact that taller women are better negotiators, but somehow this does not seem to work for men.

Monday, August 19, 2013

Surnames, height and labor market outcomes

It is well-known that taller people do better on the job market, and people get taller thanks to better circumstances in early childhood. But the causation link is not necessarily that simple, as fortunate circumstances in childhood can affect other variables that themselves influence job market outcomes. But this is difficult to establish without the proper data.

Wolter Hassink and Bas van Leeuwen got interesting data from Indonesia. The dataset contains information about army pensioners, including their height, place of birth, ethnicity, religion, education and occupation. What the authors want to emphasize here is that social networks matter in the analysis and they use the surnames to determine which network, and thus social class, people belong to. How well one does both in childhood and on the labor market is determined by networks, they hypothesize. And indeed their empirical analysis shows that while height and labor market outcomes are linked, childhood circumstances and labor market outcomes are not. The story is thus not as simple as previously thought.

Friday, August 16, 2013

The impact of bullying

If you do not manage to be part of the clique of "popular" people in school, or even worse are bullied, the conventional wisdom reassures you telling you that it is going to be all downhill from here on for the popular people, and that the bullied ones are going to be significantly more successful after school. I think the reason is that after high school, people split into the circles they really belong to, do not have to suffer other people they have nothing in common with and can really deploy their talents. The bulliers and popular people have lost their only edge, particular social interactions, once they get into the labor force and cannot progress.

Nick Drydakis tries to bring some empirical analysis to all this by using the Greek Behavioral Study dataset, which includes information about recollection about bullying, including frequency and intensity. It shows that at least part of the conventional wisdom (or at least how I perceived it) is wrong. Being bullied is associated with lower labor market outcomes, and it is hypothesized this is due to lower self-esteem which has also translated in lower academic achievement while in school (before age 18) and is perpetuated once in the labor force. Thus those whose mental health has been affected by bullying suffer significantly, and they seem to be more common than those who manage to brush it off and go on with life. However, the bullied ones achieve more human capital as measured by computer or English skills and higher degrees, but it looks like they do not managed to turn this into more employment or higher wages. The Greek labor market may be in part responsible for this. For example, the impact of bullying is particularly strong for homosexuals, and all those may not have a fair shot in the Greek labor market either. One should not read too much into a causality from bullying to outcomes here, as the author is careful to highlight. The study has nothing to say, however, about the bulliers and the popular people.

Thursday, August 15, 2013

Top Economics graduate programs are not as good as you think

Along with business schools, Economics is where pedigree matters most in the placement of PhD students to academic positions. Students from top ranked (or considered such) programs have a job almost guaranteed in research universities, and students from lower ranked universities find it very hard to break into such universities no matter what their performance is. In part, this is due to the fact that we tend to hire faculty fresh out of graduate school, while other fields go first through post-docs, and that publication delays imply that graduating students have typically no publication. Thus one has to rely on reputations only (or actually read their papers, but then are you going to read the papers of all students from lower ranked programs?).

John Conley and Ali Sina Onder find that while there is indeed a steep gradient across program rankings, there is an even steeper gradient within programs. They use student rankings within programs and cohorts and their publication output after six years, that is, when they are up for tenure. Looking at AER equivalents, they find that the top Toronto student is equivalent to the number three from Berkeley, for example. And to illustrate how steep the gradient is, the median Harvard student has after six years only 0.04 AER equivalent publications, despite coming from the #2 program. This means that more than half of Harvard students are not tenurable in any research-oriented institution.

I see two major conclusions from this: 1) Stop worrying so much about where PhD students are graduating from. It is OK to hire students from lower ranked programs as long as they excelled in those programs. 2) Even the top places should acknowledge that not all students should take research positions and need to prepare them for other ones, like industry, government or purely teaching jobs. These students are screwed twice: they are sent to tenure-track positions that they will never get tenure in, and they are woefully unprepared for the jobs they should take.

Wednesday, August 14, 2013

Strategic self-ignorance

There are times were we kind of feel we have done something stupid and do not want to know the result. For example, the grade of a test or how a recently bought stock is faring. Such situations are linked to regret aversion, where you consciously try to block available information after a decision has been taken. What about blocking readily available information before you take a decision?

Linda Thunström, Jonas Nordström, Jason F. Shogren, Mariah Ehmke and Klaas van 't Veld relate to the case of temptation, where you consciously block out information about the consequences of your action. Specifically, think about a delicious but calorie-laden meal. You kind of know it is bad for your health, but you decide not to look at the calorie count, although it is available. And that is what they had people do in a experiment where they invited people for lunch with two option: a low and an high calorie meal. It was, however, not obvious which one was high calorie, and participants could look it up. 58% chose not to and ate significantly more calories. How do the self-ignorant differ from the control group? It should not surprise anyone that they smoke more, have lower incomes, know less about nutrition and are more impatient. More interesting is that they are over-represented among males, educated and older people. Having a higher body-mass index leads to less self-ignorance. I wonder whether some of those results are endogenous to the setup of the experiment, where all those questions about nutrition are asked, which may lead people to become more conscious about their weight, especially less educated ones.

PS: too bad the equations in the paper are unreadable. Never use the Word equation editor...

Tuesday, August 13, 2013

What's up with the labor income share?

The labor income share in national income has been generally decreasing across industrialized countries for about three decades now. The consequences of this can be large, as this means a major reallocation of economic surpluses towards capital income (and the fact that Cobb-Douglas production functions become less appropriate). This trend has been exacerbated with the last recession, much to the dismay of many who see the rich capitalists screwing the labor force.

According to Loukas Karabarbounis and Brent Neiman that is at least not the entire story. Rather they emphasize that this drop in the labor income share is due to an increase in capital accumulation as a consequence of the decline in the price of investment goods. The drop cannot be attributed to changes in industrial composition, as it is also happening within industries. Note that this decline in investment good prices is often taken a symptom of technological progress, which means that for the first time since the industrial revolution, technological progress is leading to a decrease in the share of the production surplus that workers can capture. It just so happens that technology nowadays is labor-decreasing, or at least less labor-augmenting that it is capital-augmenting, and I do not think there is much that we should do about it at the technology level. At the fiscal level, that may be another question, though.

Monday, August 12, 2013

When to borrow from friends

When should I borrow from friends or from banks? With friends, I would typically get a lower interest rate, but I could risk losing a dear friendship. With banks, I pay more in interest and have to bring collateral, but I can walk away with relatively little damage. There are some situations where one should be preferred to the other.

Alexander Karaivanov and Anke Kessler study this question in the context of a world where borrowers can default strategically and there is limited enforcement of loan contracts. First they show that the optimal informal loan (from friends or relatives) features zero interest rate and zero collateral. This is because friendship is an efficient enforcement mechanism. That would typically happen with smaller, less risky loans. Then Karaivanov and Kessler show that if there is sufficient risk in the loan, one should go for a formal loan. The reason is that the collateral is divisible. In an informal loan, you either keep or lose the friendship. With a formal loan, you may still keep part of the collateral if things go bad. Using household data from Thailand, they find that their model lines up nicely with the data.

Friday, August 9, 2013

Textbooks do not matter

I have complained before, and I am far from being the only one, that textbooks are too expensive. But we still use them because we think they are useful, or because we are too lazy to come up with class material ourselves. Beyond the benefit for the lazy teacher, do textbooks actually bring something to the classroom?

Maria Kuecken and Marie-Anne Valfort looks at a case where textbooks are sometimes simply not available, classrooms in 11 Sub-Saharian countries. And it turns out the availability of textbooks does not matter, whether owned by each pupil or shared. It is only in one case, the richer kids, where there is a noticeable improvement in school achievement for shared textbooks. So it looks like teachers manage to adapt well to the absence of textbooks. And I think there is virtue in working without them: students have to listen to the teacher, learn to take notes or absorb material on the spot, and they are more active in the classroom. I wish I could go without textbooks, but unfortunately rules are rules. And publishers also need to make a living, right?

Thursday, August 8, 2013

The option of suicide

Suicide is a trigger strategy and when to pull the trigger is a decision that involves forming expectations over future outcomes. It is a difficult decision, as future outcomes are very uncertain, if not difficult to quantify.

Shin Ikeda models the suicide decision as the decision to exercise an American option on future wages. Seen this way, the suicide option is straightforward to quantify once you have wage profiles of suicide candidates (to determine timing) and non-candidates (to determine future wage profiles, their distribution and how they may differ form suicide candidates). From anecdotal evidence, anxiety seems to be an important factor, thus modeling at least risk aversion right is very important, as well as bankruptcy. Unfortunately, this is not at all how the paper proceeds. Individuals are risk neutral, but returns are adjusted for market risk. Individuals hold no assets or debt, except their human capital. The wage process is identical for everybody. It is then no surprise that the results are not realistic, indicating that the strike price corresponds to 90% of the average initial wage in perpetuity, meaning that a majority of workers is at suicide risk at some point during their life. Any study in the value of life literature gives numbers much higher than this value, and this is because people value more than just wages. Instead of only looking at money flows, one needs to consider concepts like utility and preferences...

Wednesday, August 7, 2013

Babysitting and labor market outcomes

It is well established that what happens early in life matters a lot for adult outcomes. But some adult life choices may rather be influenced by event later in childhood, like what to study in college or which profession to take on, controlling for all the cognitive skills that seem to be largely set by then.

Zeynep Erdogan, Joyce Jacobsen and Peter Kooreman look at the impact of having worked as a teenager, and in particular as a babysitter, on fertility and labor market outcomes. Results are a mess, and I suspect that endogeneity and selection bias have a lot to do with it. The authors, though, find one rather strong result: Working during 10th grade has a positive impact on labor market outcomes and delays fertility. It is rather strange that having babysat (at a specific age) would encourage a woman to have children later. But babysitting in general leads to having more children, according to their results. That all looks like spurious correlations to me, especially as instrumental variable results looks very different from OLS ones.

Tuesday, August 6, 2013

Human capital and corruption

It is commonplace to assume that corruption is bad, although the evidence is far from clear about this. Empirical investigations are typically at the macro level and have very little to say about the micro channels of corruption. In fact there is very little structural modelling or estimation in this area.

Spyridon Boikos concentrates on the impact of corruption on the accumulation of human capital using an endogenous growth model. Two channels are investigated: the first is about public resources being misdirected being education and production sectors, and the second is complementarity between human and physical capitals. Putting this to the data, it is found that corruption does not have that much impact in the education sector, and it is conjectured that corruption does not have the same bite in the education sector as in the rest of the economy. I cannot help thinking that the model misses the big elephant in the room in terms of human capital and corruption: a very common form of corruption in this regard is bribing for entry into schools, passing exams and even getting diplomas. This means that the signalling effect of diplomas is getting lost, and hence the incentive to get an education vanishes, in particular for the talented ones. And for those who attend a school, there is little incentive to learn.

Friday, August 2, 2013

Which academic field contributes most to economic growth?

If you have sat on any inter-disciplinary academic committee, you must have witnessed frustrating discussions about how one field is more useful in some respect than another, especially when resources are involved. And irremediably people compare apples and oranges, as academic fields can be very different and their metrics impossible to compare.

Cristiano Antonelli and Claudio Fassio decided to open this Pandora box and concentrate on one impact: economic growth. They perform a cross-country study and take the number of graduates in each field as an indicator of academic output, and see where that leads us in terms of economic achievement. They make the distinction between engineering, hard, social, medical sciences, and humanities in a 11-year panel of 16 OECD countries. The horse race ends with two clear winners, engineering and social sciences, and two big losers, medical sciences and humanities, the latter having a significant negative contribution to growth.

That said, should we believe those results? Beyond the obvious issue with panel cross-country regressions, the problem is that we are still comparing apples to oranges. In some countries, medical studies are at the graduate level only, while it is undergraduate elsewhere. There are also stark difference for Economics as well. In the US, many students graduate in that field and have actually only two years of classes in this major, having to take general education classes first for two years. In Europe, Economics students spend their whole four years on the topic. And the same applies to other fields. Thus counting students, and especially if you want to make the claim they are specialized in a particular field, is rather heroic. I would not yet claim social sciences have won this battle.

Thursday, August 1, 2013

AEA elections are on, you know for whom to vote

As explained in a previous post, write in Gregory Burge for every slot in the AEA executive, and no one else.