Feldexperimente

Wir brauchen mehr Feldexperimente.

Early tracking--not good.

Austria's conservative party has a new spokesperson for education, Christine Marek. In one of her first interviews, she reiterates the Conservative's view that early tracking (the separation of 10 year olds into an "academic" and "non-academic" school track) is good and that comprehensive education until a later age ("late tracking") is bad.

Here's some evidence: Hanushek and Woessmann compare countries and identify tracking effects by comparing differences in outcomes:

The results suggest that early tracking increases educational inequality. While less clear, there is also a tendency for early tracking to reduce mean performance.

This is, by far, not one single study that shows that early tracking reinforces the impact of family background. Just one further example, Giorgio Brunello and Daniele Checchi :

early tracking reinforces the family background effects on the years of completed education, on the probability of dropping out and of enrolling or graduating in college. Therefore, in countries with less pronounced tracking, the difference in the dropout rate and college enrolment or completion between the children of poorly and better educated parents is smaller than in countries with stronger tracking.

Joint custody, marriages, divorces and the rest

Together with Marco Francesconi and Martin Halla, we examine joint custody and consequences for marriage, divorce, fertility, and female employment. We use registry data for the entire Austrian population and also divorce records from courts to analyze post-divorce outcomes. The results suggest that the availability of joint custody for divorcing parents---introduced in 2001---lowered divorce rates, increased marriage and marital birth rates, and lead to higher money transfers after divorce.

Joint custody provided men with more incentives to invest in marriage specific capital, which lead to more, and more stable, marriages. It also seems to have resulted in more specialization as female employment rates declined after the introduction of joint custody.

Academic research and teaching

(HT Economic Logic)


The main aims of a University are to teach students and to generate new knowledge. These two aims could conflict with each other. For example, when I devote much time and effort for teaching, I might miss these resources for research. An alternative view is that good researchers also make good teachers, because they have a deeper understanding of the problems and this will improve their teaching. The empirical evidence is not conclusive.

It has, of course, implications for the organization of universities, in particular for the incentives related to teaching (typically: none) or to research (typically: somewhat more than for teaching). Some argue that spezialisation is a good way of addressing such conflicting goals, for example, by separating the faculty into "teachers" and "scientists". Others require mandatory teaching training (I have not yet heard of mandatory research training).

Recent research by Aurora García-Gallego, Nikolaso Georgantzís, Joan Martín-Montaner, and Teodosio Pérez-Amaral provide evidence from a Spanish University that

professors with a typical research output are somewhat better teachers than professors with less research. Moreover, non researchers are five times more likely than researchers to be poor teachers. In general, the quality of university-level teaching is positively affected by published research across most levels of research output. (p3)

This does not conflict with my suspicion, but I would like to see more evidence on this. As universities regularly collect students' evaluations of teaching and require staff to provide details on their research, this evidence could be readily provided, for example, as students' term projects.

Wage costs

I encountered these rental rates for boats in Pokhara, Nepal.

What I find interesting are the costs for a "driver", the Nepalese equivalent of a gondoliere. The difference between hiring a boat for one hour with and with no driver is 50 rupees.

However, if you rent the boat for two hours, the difference is 250 rupees, or 125/hour. If you'd take a boat for a whole day, the difference is 800 rupees, or 100/hour in an eight hour day.

The one hour rental rate has the cost of the driver at 14% of the rental price. The wage costs of the daily rental is 800 rupees, or impressive 53% of the rental price.

Is it worth it? You'll have to find out for yourself.

The Economics of Retractions

How much harm is created by "false science"? False science leads others scientists into spending effort, time, and resources on replicating fabricated results, cutting promising research short, and may "stifle the field".

New research by Azoulay, Furman, Krieger, and Murray, "Retractions", aims to quantify the impact of false science on other ("honest") research. Theirs is a tremendous data sifting job:

1. They identify retractions to be caused by standing on the "strong" or "shaky" shoulders (of giants), or the absence thereof. In other words, retractions that are caused by e.g., editorial mishaps, such as printing the same article twice ("strong"); retractions where part of the claims are unsubstantiated ("shaky"), or the almost 600 articles where all claims are false ("absent shoulders").

2. Using the medical PubMed database, they identify related articles through keywords and define a scientific field as the set of articles whose keywords overlap with the retracted articles.

3. By restricting the analysis to articles that were published before the retracted article, excluding articles by the authors involved in the retraction, they construct a treatment group of articles--a field--that were shocked by the retractions. Comparing their citations to articles of a "control group", consisting of articles in the same issues of the journals, they can estimate the causal impact of false articles on the field.

Their

"findings show that scientific misconduct and mistakes, as signaled to the scientific community through retractions, cause a relative decline in the vitality of neighboring intellectual fields. These spillovers in intellectual space are significant in magnitude and persistent over time. In other words, there is clear evidence of negative spillovers in instances of "false science" to broader swaths of the intellectual field in which they take place."

What keeps unemployment durations short?

Experimental evidence shows that early and intensive support (a the combination of meetings, job search courses, and early activation) keeps unemployment durations short. However, it was not clear from these experiments what "did the trick", was it the meetings, the courses or the combination of interventions?

Pederson, Rosholm, and Svarer expand on these experiments and show that the effects are due to the early meetings. The 6 to 7 meetings during the first two months of an unemployment spell have dramatic consequences:

individual meetings between newly unemployed workers and caseworkers can increase employment rates over the next two years by 10%, corresponding to 5 weeks, and our cost-bene…t analysis shows that the surplus per new unemployment spell is around EUR 4725.

Group meetings also have positive consequences, but these are more moderate than those from individual meetings.

Fame! Fortune! Markets!

Hilmer, Hilmer, and Ransom have a new working paper where they analyze the pay of 1,009 US academic economists and estimated the effect of their journal articles on their salaries.

While I seriously doubt their estimation results---it appears to be a textbook example for endogeneity bias as unobserved ability will influence both the capacity to write good papers and to negotiate good salaries---the compiled data are interesting and they provide some interesting facts.

For example, of the almost 15,000 articles that were published before 2000, about one quarter were never cited (so far). Similarly interesting is the distribution of the Hirsch-Index presented in their Table 5: the median is at 5 and cumulative citations per author hover around 50.

"... but it is only articles in the elite journals that have an impact on pay."

On rent controls

Vienna is a pleasant city and it regularly tops the quality of living surveys. A such, it attracts plenty of new citizens, Statistics Austria estimates "forcasts" that Vienna is "expected to have the most marked population growth till 2060 [in Austria]". It is perhaps not surprising that rent prices increased over the last years. (This is what we expect if demand increases and supply responds sluggishly, e.g., because it takes time to obtain planning permissions et cet.)

What are likely consequences of (binding) rent controls? We may learn from other situations, e.g., Manhattan: Glaeser, Gyourko, and Saks argue "that the limited supply response is not the result of technological constraints, or imperfect competition in the construction industry, but rather the consequence of an increasingly restrictive regulatory environment."

Rent control, i.e., an imposed maximum rent price, is perhaps the most frequent regulation and has been studied widely. In principle, rent controls need not be harmful:

"whether such controls are harmful or helpful depends on the particular package of regulations adopted, which is the outcome of a political process. Thus, second-generation controls should be judged on the empirical evidence and, since the programs are so varied, on a case-by-case basis."

And this is why calls for tighter regulation should be met with scepticism: For Vienna, too few facts are available that permit the judgment of recent calls for tighter rent controls. Beware Greens bearing gifts.

Do immigrants cause crime?

According to a newly published study in the Journal of the European Economic Association by Bianchi, Buonanno, and Pinotti (an earlier working paper is here):

According to these estimates, immigration increases only the incidence of robberies, while leaving unaffected all other types of crime. Since robberies represent a very minor fraction of all criminal offenses, the effect on the overall crime rate is not significantly different from zero.

Contingent Valuation

Jerry Hausman (MIT) launches a broadside against contingent valuation in the latest Journal of Economic Perspectives: "From Dubious to Hopeless".

I do not expect that proponents and opponents of contingent valuation will ever agree. Some bad ideas in economics and econometrics maintain a surprising viability. Numerous branches of the federal government continue to fund contingent valuation research in the hope that it will support their favored policies subject to cost–benefifi t analyses. (p54)

There are two more articles in this issue--somewhat more positive-- and all are worth reading.

Cartels

A signpost I encountered in Nepal:

"all prices are same price": despite my expectations--high prices and low quality--the food was very good at reasonable prices!

marketing by flattery

"Econ Tidbits is an eclectic but insightful blog on a variety of (mainly) economics topics. The author is a professor in Austria, and so naturally has more Austrian, and more international focus than many other economics websites." (100 of the best sites in the field)

It works.

Prussian Historic Data

The ifo Prussian Economic History Database (iPEHD) is a county-level database for all Prussian counties, 1816-1901. These data are an excellent source for empirical research and, above all:

"The data are provided free of charge. We have tried to document the data as good as - we think - we possibly could, including references to the original publications from which the data are drawn."

In addition, the data are well documented: Becker, Sascha O., Francesco Cinnirella, Erik Hornung and Ludger Woessmann, "iPEHD - The ifo Prussian Economic History Database", CESifo Working Paper No. 3904.

Forensic Economics

A new field is emerging in economics, Forensic Economics. Eric Zitzewitz has an article in this issue of the Journal of Economic Literature providing an exciting overview on this emerging field.

Examples include teachers cheating on exams, road builders skimping on materials, violations of U.N. sanctions, unnecessary heart surgeries, and racial biases in employment decisions, traffic stops, auto retailing, and even sports judging. In each case, part of the contribution of economic analysis is in uncovering evidence of wrongdoing.

Make babies, not war

I am currently reading John Irving's novel "Last night in Twisted River" where I came upon "Kennedy fathers": "in March 1963, President Kennedy had issued an executive order expanding paternity deferment. ... Having a kid could get you out of the war".

This, I assume, would make for a nice empirical project: analyze birth rates for different cohorts (Nixon put an end to this type of avoiding the war in 1970). I expect to see an increase in fertility, driven by young couples (men), attempting to avoid the draft.

However, Andrea Kutinova did already do this in her article "Paternity derferments..." and she found--surprise, surprise!--a huge response of the birth rate: "the calculated conservative increase in the number of first births by 15,532 in June and August 1966 represents more than 7% of the total number of first deliveries in those 2 mo[nths]."

What do we expect about the lifes of those babies, in comparison to those born before/after the paternity deferment? Now, this would make for an interesting research project...

How trade unions increase welfare

Alejandro Donado and Klaus Wälde, in a paper that's just been published by the Economic Journal (a previous version is available here), show that trade unions might substantially improve welfare by contributing to workplace health and safety. (Spoiler: be warned, this is "mainstream" economics!)

They enrich the standard model by assuming that unions do not only care about their members' wages, but also about their health. Because workers cannot precisely judge the effects of working conditions on their health--an information asymmetry--, competitive markets cannot provide an efficient solution, leading to low standards (standards are costly), to more accidents, and to inefficent high sickeness levels. (Without information asymmetry, workers would require higher wage rates, resulting in compensating differentials.)

If unions are better informed than their members, or, alternatively, invest resources into the analysis of workplace hazards, they will bargain for better health and safety standards. This will ultimately result in healthier workers and lower absence rates. (The union "internalises" at least some of the externality arising from information asymmetry.)

So far, exciting stuff! However, as often in this type of research, there is almost no data available to test this empirically. They make an heroic effort to interpret union densities across countries in the light of their theory, but they conclude with an appeal

"It might be a good idea for unions to go and collect data on these issues." (p1008)

To which I'd like to add: I'd love to analyze these data!

RCTs and Policy

We need more evidence-based politics! And randomized control trials can help to provide this evidence.

Some countries have realised the need for such evidence and push ahead: the UK Cabinet Office is hiring four researchers to join their Behavioural Insights Team! ("The Behavioural Insights Team (more commonly referred to as the Nudge Unit) was set up in 2010 to help apply behavioural economics and behavioural psychology to public policy in the UK.")

While I do not want to see you running off to Old Blighty, I'd like to see more of this happening here! And, this goes without saying, I know that there are a great number of colleagues, me included, who like to be involved in such RCTs and (economic) policy.

Handelsblatt Ranking

According to the The Lumpy Economist, who comments on the Boykott of the Handelsblatt ranking by business professors, three structural differences between Germany and the US tend to increase the importance of the Handelsblatt ranking:

1) Smallness of departments, which often leads to the situation that nobody in the field to be hired is there and can competently judge the papers of the candidates – the HB ranking is an easy way out;

2) lack of a general education in a PhD program, which means that faculty with a traditional German PhD eduaction often lack the background to talk about papers in other fields (of course, I would always defer to my theory or econometrics colleagues to judge such papers, but it’s also not true that I can be completely bamboozled in these areas);

3) I think the most important one is the lack of a department structure to begin with, which has two direct consequences: a) I honestly think that many German professors do not care as much as US professors about who is going to be a colleague – as I said, in the US senior hiring is one of the most important services one can provide to the department; and, rightfully so, as they run a chair, which is ultimately an isolated unit for which they are responsible on their own. [my emphasis]

Hear, hear!

(Working at a university, where the business departments are, according to some ranking, not the most prolific when it comes to publications in peer-reviewed journals.)

(Update: hear, hear!: http://www.youtube.com/watch?v=3Nz846dDids&feature=fvwrel .)

Eine allgemeine Wehrpflicht ist teurer als manche glauben

An verschiedenster Stelle, z.B. im Standard, wird behauptet, bei einer Volksabstimmung über die Wehrpflicht stimme die "großteils davon nicht betroffene Bevölkerung" ab.

Das ist eine seltsame Behauptung. Wer soll diese nicht betroffene Bevölkerung sein?

Es sind vermutlich weder:
1. Die Wehrpflichtigen, denn die müssen den Wehrdienst leisten.
2. Die Steuerzahlenden, denn diese finanzieren die allgemeine Wehrpflicht und alle weiteren sozialen Kosten.
3. Unternehmen, denn diese haben uU Kosten durch die Wehrpflicht, wenn Lehrlinge zum Dienst müssen, bzw profitieren uU von den Fähigkeiten, die die jungen Männer beim Heer (auf Kosten der Allgemeinheit) erwerben, z.B. Führerscheine, o.ä.
4. Die Kranken und Alten, da diese von den billigen Zivildienern profitieren, die die Kosten der Pflege und Versorgung niedrig halten.
5. Politiker und Politikerinnen, die das Thema für Selbstdarstellung verwenden (können).

Wer bleibt da noch über?

1. Aus ökonomischer Sicht ist die Wehrpflicht eine ineffiziente Form der Organisation einer Armee, da die positiven Auswirkungen einer Spezialisierung vergeudet werden, d.h., wer in der Armee produktiver als in anderen Bereichen ist, sollte dort tätig sein. Eine allgemeine Wehrpflicht ignoriert Unterschiede in den Fähigkeiten und setzt Personen daher ineffizient ein, d.h., vergeudet Ressourcen.

2. Es wird argumentiert, dass eine allgemeine Wehrpflicht billiger als eine Berufsarmee sei, da die Wehrmänner keine Bezahlung im üblichen Sinn erhalten, sondern nur den Sold. Das ist klarerweise eine falsche Betrachtung (Opportunitätskosten!), da diese Betrachtung die sozialen Kosten der vergeudeten Ressourcen ignoriert!
Kerstens und Mayermans zeigen, dass die gesamten Kosten der belgischen Wehrpflicht rund doppelt so hoch als die reinen budgetären Kosten waren.

3. Eine allgemeine Wehrpflicht kann auch zu langfristigen sozialen Kosten führen, wenn die Wehrmänner wegen der Wehrpflicht ihr Verhalten ändern, zum Beispiel, wenn sie weniger in Ausbildung investieren: Keller, Poutvaraa und Wagender zeigen, dass eine allgemeine Wehrpflicht zu weniger Studienabschlüssen führt. In manchen Fällen dürfte es allerdings zum gegenteiligen Effekt gekommen sein, Maurin und Xenogiani zeigen, dass Männer in Frankreich wegen der Wehrpflicht vermehrt studierten, vermutlich um die Wehrpflicht aufzuschieben und vermutlich in der Hoffnung, sie gänzlich zu vermeiden. Beides(!) ist ineffizient--weder sollten Männer studieren, die es nicht wollen oder anderswo produktiver sind, denn auch das hat Opportunitätskosten, noch sollten jene nicht studieren, die es wollen und könnten.

4. Es gibt widersprüchliche Evidenz über die Konsequenzen einer Wehrpflicht in Friedenszeiten für die Wehrmänner: Card und Cardoso finden, dass junge, schlecht ausgebildete Portugiesen von der Wehrpflicht in Form von höheren Löhnen profitierten. Bauer, Bender, Paloyo und Schmidt finden hingegen, dass derartige Lohnsteigerungen bei deutschen Männern von unterschiedlichen Merkmalen stammen, d.h., gesunde Männer leisten ihre Wehrpflicht ab und weniger gesunde nicht--und gesunde Männer verdienen besser als weniger gesunde.

Fazit: Wer behauptet, die Wehrpflicht ginge nur Männer etwas an, macht einen Fehler--die effiziente Verwendung von Ressourcen betrifft alle!

Literatur

Bauer, Bender, Paloyo und Schmidt, 2012, "Evaluating the labor-market effects of compulsory military service", European Economic Review.
Card und Cardoso, 2011, "Can Compulsory Military Service Increase Civilian Wages? Evidence from the Peacetime Draft in Portugal", NBER working paper 17694.
Keller, Poutvaraa und Wagener, 2009, "Does Military Draft Discourage Enrollment in Higher Education? Evidence from OECD Countries", IZA discussion papers 4399.
Kerstens und Meyermans, 1993, "The draft versus an all‐volunteer force: Issues of efficiency and equity in the Belgian draft", Defence Economics.
Maurin und Xenogiani, 2007, "Demand for Education and Labor Market Outcomes", Journal of Human Resources.
Rauscher, 2012, "Energiesparlampe: Das Lobby-Licht", Standard vom 31.8.2012.

Kein akademischer Nachwuchs?

In einem Artikel des
Standards über eine mögliche Studienplatzfinanzierung wird behauptet, "So viele Professoren seien gar nicht am Markt." Diese Aussage stammt vermutlich von einem nicht näher genannten "Experten" des Wissenschaftsministeriums und soll aufzeigen, dass die Änderung des Betreuungsverhältnisses von 1 ProfessorIn zu 200 Studierenden (1:200) auf 1:40 nicht am Geld oder am politischen Willen scheitere, sondern--it's economics, stupid!--schlichtweg am zu geringen Angebot von professorablen Personen.

Als empirischen Arbeitsmarktökonomen interessiert mich daher ganz besonders, wie man zu dieser Diagnose kommt, denn meine Erfahrung aus mehreren Berufungskommission ist schlichtweg eine andere--es gibt mehr ausgezeichnete BewerberInnen als Stellen! (Nebenbei bemerkt, auf eine Professur bewerben sich nicht nur ProfessorInnen, sondern gerade auch solche, die es noch nicht sind, nämlich KollegInnen aus dem Mittelbau mit einer Habilitation oder ähnlichen Qualifikation.) So kenne ich viele KollegInnen, die im Ausland Karriere machen, weil es in Österreich keine bzw keine attraktiven Stellen gegeben hat.

Weiters--it's economics, stupid!--gerade weil es wenige Professuren gibt, entscheiden sich viele gegen eine akademische Laufbahn: wenn die Aufstiegsmöglichkeiten gering sind, und an einer Universität gibt es nur wenige Möglichkeiten zur Beförderung, dann wählen einige eine andere Laufbahn. Zusätzlich gibt es noch diejenigen, die aus--realer oder erwarteter--Chancenlosigkeit (Netzwerke, "falsche" Publikation, "falsches" Geschlecht, zu jung/alt,...) keine Bewerbung versenden.

Die Aussage ist unglaubwürdig und vermutlich schlichtweg falsch--wenn nur die existierenden Professuren gezählt werden, darf es nicht wundern, dass diese Zahl gering ist! Schaffen Sie die Stellen, starten Sie die Berufungsverfahren und Sie werden vermutlich überrascht sein, wie viele gute KollegInnen sich bewerben werden!

google scholar updates

Scholar Updates: Making New Connections - Google Scholar Blog

so many things to read, so little time...

scientific progress

discovering things. (RT @TimHarford).

Love and Markets

There's a market for proposals--and for those not so lucky, there are mail order brides. In case it is all a bit rocky, you may get an insurance. You think that a marriage is for life? You may want to consider this insurance for wedding gifts...

PhDs and Politics

The Guardian on plagiarism in European Politics
http://www.bbc.co.uk/news/18962349

However, Johannes Hahn is not mentioned.

poor grammar

"grammar is relevant for all [companies]" (Harvard Business Review)

Grammar signifies more than just a person's ability to remember high school English. I've found that people who make fewer mistakes on a grammar test also make fewer mistakes when they are doing something completely unrelated to writing — like stocking shelves or labeling parts.

I get the same impression from reading students' seminar papers and disserations. If you cannot use a spell checker, you should not be in college.

strategic management?

German Prof stumbles over plagiarism. Not so strategic? "the whole affair has the potential to turn into a major academic scandal" (http://lichtenthaler.bwl.uni-mannheim.de/34+M5f823134e1c.html ) -- as if it not already...

for Stata users

Econometrics beat alerted me to this rather interesting site Econometrics by Simulation, a treasure chest of useful Stata code.

172 faked articles

Probably the most successful scientific fraudster: retraction watch

missing p-values..

via MR: "rearchers might be tempted to inflate the value of their tests by choosing the specification that provides the highest statistics. Note that Inflation is larger in articles where stars are used in order to highlight statistical significance" Star Wars: The Empirics Strike Back

Ray Bradbury-a great obituary

http://www.economist.com/node/21557294

on austerity

reading up on austerity measures:

"This eCollection provides a synthesis of the current debate, on whether the consolidation programs adopted by several countries in since 2011, cut too much and too soon, thus creating an unnecessary contractionary bias and macroeconomic risk. In the Eurozone, the backlash from such policy mistake could easily undermine social consensus towards the common currency and jeopardise progress towards integration."

vox