Freakonomics is the partnership between University of Chicago economist Steven Leavitt and New York times Magazine writer Stephen Dubner. In books, blog posts, and a host of other media, they use economic principles to expose “the hidden side of everything.” On June 3, they posted a podcast that revealed some of the hidden sides of tipping, but in doing so they gave credence to researchers with hidden noneconomic agendas that are unsupported by the economic evidence they muster.
In this podcast and transcript, Leavitt and Dubner discuss why tipping makes some people (such as Leavitt) uncomfortable. They interview several people including Cornell Hotel School professor Mike Lynn and Harvard Business School professor Magnus Torfason. Lynn studies tipping behavior while Torfason teaches entrepreneurship. Perhaps inadvertently, the podcast reveals that both have strong policy views that are unrelated to their training and research but which appear to dominate what they make of it. Lynn explicitly states that he’d like to ban tipping; Torfason implies as much. Neither offers an economic rationale for this policy recommendation, so it is reasonable to infer that economics, though the nominal basis for the conversation with Leavitt and Dubner, is irrelevant.
Lynn believes that tipping is discriminatory and hence probably illegal under the Civil Rights Act of 1964:
[T]he Supreme Court has ruled that even neutral business practices that are not intended to discriminate, if they have the effect of adversely impacting a protected class are illegal. And so it’s not inconceivable to me that there will be a class-action lawsuit on the part of ethnic minority waiters and waitresses claiming discrimination in terms of employment. And it’s conceivable that tipping might be declared illegal on that basis.
Lynn implicitly expresses support for this interpretation of labor law, even though it is an employer’s customers, not the employer, whose actions would be implicated. However valid this opinion might be, it has no economic content.
The podcast includes a reproduction of a graph from an unnamed paper by Torfason showing a cou8ntry-level association between the prevalence of tipping and an undefined index of corruption. Association does not prove causation, but Torfason appears to believe that this association is in fact causal:
The more tipping you see in a given country, the more corruption you generally see in that country as well.
[W]e started by collecting information about tipping in some 33 different countries and in 33 different service professions. These are waiters, hairdressers, taxi drivers and so on. And we also looked at the levels of corruption in these countries looking at how they scored on the corruption perception index, which is publicized by Transparency International. Running a simple correlation we found that there’s a positive relationship between those two, which could have been for many different reasons. But we then control for a number of other factors such as GDP, the culture in a country in general, homicide rates and other measures of legal enforcement—and this relationship stayed significant. In other words, we found that even after controlling for these other things we still find this positive relationship between tipping and corruption.
Why these factors were judged sufficient isn’t clear, nor is there evidence provided showing why they would have been sufficient. To easily see why, notice that Finland and the Scandinavian countries are clustered on the vertical axis (corruption) but vary from 0% to 15% in the prevalence of tipping.
Indeed, there is no evidence provided showing that the 33 countries chosen are representative. The association may be driven by the 10 countries located in the upper right corner. If they were removed, would the association vanish?
In the interview, Torfason acknowledges that even if his causal inference is true, it’s not true for the United States:
In the U.S. we have a strong norm for tipping, and so if you’re an upstanding and law-abiding citizen you’re very likely to be also a good tipper, because that’s what the norms tell you. And you’re also likely to avoid corruption.
Normally, economists require persuasive evidence of persistent market failure before they even begin searching for regulatory remedies. Because banning certain products or actions is the most restrictive form of regulation imaginable, bans usually require the most stringent of evidentiary burdens.
Neither Lynn nor Torfason offer any evidence at all of market failure, so it can be inferred that they oppose tipping for noneconomic reasons. Lynn is transparent that he has relatively extreme policy views on labor law. As for Torfason, the interview suggests that he’d like the United States to be more like his native Iceland. Notice that Iceland and the United States are at opposite ends of the distribution of tipping prevalence but they are not far apart on Torfason’s index of corruption, which as noted above is undefined.