Hiring kayaks or canoes to go down the Saale. Stopping for a drink in Döbritschen. Ending up in Camburg.
Taking part in the walk along the Saale Horizontale all around Jena (100km), stopping exhausted in the middle and limping back home painfully. Do some little parts over the following years, not forgetting to walk at night too. (http://www.horizontale-jena.de/)
We were investigating what happens after the nudge, i.e. the impact of nudges on people’s attitudes towards the behavior they are encouraged to adopt. Our question was whether the nudge changes people‘s mind about the “desirable action”, or if they rather react to it mechanically, like one reacts to an obstacle in the way of one’s preferences.
In our experiment, we offered people a total of 4€ to fill a survey. At the end, they could pledge the money to a charity. There were four treatments:
– Nudge: The default was to pledge to one charity.
– No nudge: The default was to keep money to oneself
– Nudge with choice: The default was to pledge to one of three charities.
– No nudge with choice: The default was to keep money to oneself, the alternative was to pledge to one of three charities
Taking the alternative involved a (limited) effort, i.e. writing “I want to do this” rather than clicking on a button to do what the nudge suggested.
In all cases, whether people pledged the money or wanted it for themselves, they had to come collect the money in our offices. They could then put the money into piggybanks for each charities.
We found that the nudge did work in eliciting about twice more pledges to charities. However, those who were nudged to pledge were less likely to actually come to claim the money. In the end, treatments with nudges did not result in more money actually contributed to charities.
We also elicited people’s attitudes to the charities and to us, the nudgers. We found that there was no apparent effect of nudges on people’s perception of charities, or of us, the nudgers. The good thing is that they did not make people more cross against charities or against us, so people do not usually react like me when people try to push me to do something.
In fact, the only robust effect on perceptions of charities was that if you give people the possibility to pledge, then they become less keen on charities than in a control treatment where they did not have the possibility to pledge. This is probably due to reactance, i.e. people trying to find reasons not to pledge by devaluing charities.
Giving more choice whom to pledge to improved matters only in so far as people were then more satisfied with their choices, but they did not pledge or contribute more, and giving choice did not improve their perception of us, the nudgers.
Subjects were not very fond of the possibility to contribute the money to a charity and were not keen on the possibility to donate proceeds from experiments in further experiments. Nudging them to pledge also made them more likely to say they felt forced to contribute.
Our experiment confirms the view that nudges act in a mechanical way, i.e. they lower the threshold at which people adopt a given behavior. We were able to exclude the other possible effect of a nudge, i.e. that nudging people to pledge makes them keener on charities. This could have been the case if people rationalize their nudged pledge ex-post by telling themselves they liked the charities. Instead, we found that nudges had no effect on people’s perceptions of the charities.
All this means that those who are nudged to do something will not keep on doing this in the absence of a nudge. We knew this already, but it also means that nudging people to do something that will require subsequent effort is unlikely to work if that subsequent effort is not also incentivized. Indeed, for our case, the people we nudged to pledge were “at the margin”, i.e. more or less indifferent between keeping the money or contributing it. This means that they were not so keen on contributing in the first place and were therefore unlikely to actually do the effort to collect the money in order to give it to charity.
As in the proverb, “there’s many a slip ‘twixt the cup and the lip”: getting people to “fake it” (pledge) does not always work to get them to “make it” (contribute).
Of course, charity muggers probably knew this already. You have to make sure to be ready to collect the money right then and there and you should not let people get away with just making a pledge.
In a nutshell, clever consumers are not enough to avoid collusion and high prices. Consumer protection may therefore require some encouragement for firms to present prices and products in a common format. For example, firms may have to be required to provide some standard information to help consumers in comparing products, for example an index of energy efficiency for fridges and other appliances.
Unit price information is already generally available or even mandated in supermarkets. Some standardization is also present at the national level for presenting lending rates in terms of annual percentage rate of charge. There is however a lot of progress to be made for example in the automobile market, where fuel economy information is often misleading and wrongly conveyed.
Agreeing on common formats for measuring the performance and prices of different type of relatively undifferentiated products could therefore be a valuable extension to the efforts that have led to the progressive spread of the metric system for physical measurements.
Crosetto, Paolo and Gaudeul, Alexia, (2014), Choosing whether to compete: Price and format competition with consumer confusion, No 2014-026, Jena Economic Research Papers, Friedrich-Schiller-University Jena, Max-Planck-Institute of Economics, http://EconPapers.repec.org/RePEc:jrp:jrpwrp:2014-026.
A very busy summer for me this year, after last year’s ban on conferences to focus on doing experiments.
For me the summer started at the end of May with a research stay in Rennes, where I was invited by Eric Darmon and gave a talk about my research with Paolo Crosetto on how firms can profit from keeping consumers confused about how their product compares with the competition. A shorter version of my presentation is here. I blogged about it here. Paolo and I are putting the last touches to the paper before releasing it.
Rennes was a very enjoyable stay indeed, as Eric Darmon and I had very fruitful discussion about how far consumers would trust firms that advise them on what product to choose based on private information collected for example via an analysis of their behaviour on the Internet.
Right after returning, I participated in the Jena Science Slam, which I won! (blog post here)
Alberobello was the location for the bi-annual industrial organization conference organized by the University of Salento. I presented my paper with Paolo on whether firms can collude to maintain a confusopoly. Particularly interesting presentations for me were in the field of behavioral IO: Anastasia Shchepetova on confusopolies, Giovani Ursino on deceptive advertising, and Sofronis Clerides on consumer inattention. There was also a plenary talk by Mark Armstrong that provided a summary of some efforts in modelling markets with uninformed consumers.
After two weeks back, it was the start of the Jena Summer School, with two parallel programs, one for the GK-EIC focused on innovation and the other for the IMPRS focused on experimental economics.
I was the happy adviser of the IMPRS team that won the group work contest, which involves running an experiment from idea generation to design, data gathering in the lab, analysis and presentation of findings, over 4 weeks. Their theme was investigating the difference between uncertainty and strategic uncertainty. See p. 13 here for a nice pic of the group.
The conference included a number of sessions of interest for me, notably a session I chaired and which focused on measures of subjective well-being (Hans-Jürgen Engelbrecht), whether governments should focus on increasing happiness (Christian Schubert) and how paternalism could be autonomy enhancing (Martin Binder).
The sessions I enjoyed most were however about history, notably talks by Mark Knell, Heinz Kurz and Harald Hageman with insights about the differences between Schumpeter and Keynes.
A very enlightening presentation was about the future of capitalism by Carlota Perez which predicted a new “Belle Epoque”, and a very dynamic talk was by Mariana Mazzucato on how firms, notably banks, socialize risks and privatise rewards. A bit too black and white for some, but I liked it all the same.
Finally, the summer ended with the ESA European conference in Prague, where I met many present and former members of the soon to be disbanded Max Planck Institute for Economics.
The second was about social preferences when faced with random wealth assignments mechanisms – also knowns as “catastrophes” in some circles :). I blogged about that work here but the paper I wrote at that time was very misguided and I completely changed my method of analysis, getting rid of the parametric assumptions about utility functions and focusing on non-parametric measures of risk aversion and their relation with lottery characteristics. Preparing this presentation was the occasion for me to make sense of two experiments I ran on that theme, and to plan for a third that will bind those two.
The most interesting presentation was that by Han Bleichrodt on PRINCE, the PRior INCEntive system. I intend to use his suggestions in my future experiment, with an easy adaptation to incentivizing two-person interactions rather than simply individual choice. I also liked a presentation by Claudia Neri on preferences for decision rights, which has some similarities with work by Nadine Chlass on purely procedural preferences.
Overall, a very busy summer indeed, during which co-authors Marian Panganiban and Ayu Okvitawanli also ran the sessions for our job market experiment on gender discrimination. We vary not only the composition of the hiring pool but also the composition of the hiring team and relate discrimination to a number of individual variables, including an Implicit Association Test between competence and masculinity.
Beyond work, I also spent week-ends in Leipzig and Dresden, went kayaking on the Saale, spent many evenings hanging in my hammock in Paradies Park and kept on improving my skills in chess. I reached what is now a pretty decent club player level and intend to take part in my first chess competition next year!
A really good picture of the Science Slam victory celebration with GSBC and MPI colleagues. The photographer managed to do a really good group picture. I will remember his technique to encourage people to do a wide smile: not the usual “Cheese!”, but stretching his own mouth wide with his fingers. It seems to have worked!
The videos of the 4th Jena Science Slam are also now online:
I am slammer number 4. I am not sure I like my presentation so much but at least it got people laughing. I was told I already won with the first picture in the presentation, which features THREE cats. Cats are for the win apparently!
In other remarks, it is surprising how things that appear to last only a second on the video seemed to go so much slower when I was on stage!
My (winning!) Jena Science Slam presentation, with notes in small character added. The title of my slam was “Why is shopping such a baffling ordeal, and what can you do about it?”. I am very thankful to the audience, they had amazing energy in their cheering and applause!
There should soon be videos of this year’s Slam on the Graduate Academy’s website. It will be funny to see how it looked like :)
I felt like giving a little primer on the results of an experiment that Paolo and I did in December, about the incentives for firms to confuse consumers. I was motivated by seeing the following recent article in JEBO:
As in most (all?) articles on shrouding, this paper considers only the one stage game and finds as usual that firms will want to make their prices transparent… while ignoring the impact of collusion.
Yes, I did the same thing in my paper in Economica with Bob Sugden on spurious complexity and common standards (http://dx.doi.org/10.1111/j.1468-0335.2011.00895.x). But now, from a recent experiment with Paolo Crosetto where we test a similar model in the lab, we found that firms were quite able to collude not to unshroud after experiencing the horrors of an unshrouded equilibrium!
I am excited about writing up the results, as in fact, being able to choose to make prices transparent or not can help collusion compared to a standard setting with no possible shrouding, and this in two ways: by serving as a signal that one wishes to make peace with others, and also by making the punishment phase harder on those who deviate and unshroud. Once said, this looks obvious now, but I guess this has been missed by others than us, so I don’t feel so bad about missing it.
I will be working in the next month or so on writing up the results, but I am now busy finishing writing a paper on the relation between exit costs and incentives to exit in a repeated version of the public good game with imperfect public monitoring and stochastic outcomes. Very interesting, but it has been hell to find the correct perspective in writing about the experimental findings.