This is the title of a paper by economists Scott Cunningham and Todd Kendal that provides an overview the economics of prostitution and how it has been affected by technology for the forthcoming Handbook on Family Law and Economics. Here are some scattered observations.
Among the many interesting results they report are that 40% of prostitutes have a college degree, and 80% have had some college.
Apparently, the current economic downturn has led to churn in the industry and lower wages, as supply has expanded and demand has contracted. For many women this puts the market wage below their reservation wage, which drives them out of the market.
Uncommon for economists they report ethnographic results as well, for instance:
Ethnographic interviews with various workers revealed that spouses and partners were typically aware of, and even complicit in the sex worker’s labor supply, frequently working as a manager or assistant. These results indicate the necessity for a fuller understanding of the complex relationship between prostitution and marriage.
The authors apply a hedonic analysis to prostitute prices and thus estimate a price premium for various characteristics including race, age, and services offered. Contrary to the economic literature showing either an Asian wage premium or no difference relative to whites, wages for Asian prostitutes are found to be systematically lower than for other minorities:
Another interesting question they address is what happens to the demand for sex ads on Craigslist when a $5 fee is imposed. The graph they provide tells the story:
There are many other fascinating results in the paper, and good information on datasets for researchers interested in these issues. Scott’s website with links to his other research is here, and you can follow him on twitter at @scottcun.
3 comments
Comments feed for this article
Wednesday ~ October 20th, 2010 at 10:11 am
Steve Hamlin
There is a lengthy & interesting section in SuperFreakonomics about this very same topic. Sudhir Venkatesh, the same guy who did the Chicago drug gang economic analysis in Freakonomics, follows it up with street-level prostitution research including surveys, interviews & per-sex-act data gathering, and compares current to historical Chicago prostitution economics & environment.
Fascinating economic analysis & discussion of the world’s oldest profession. Turns out they price-discriminate by race, July 4th weekend brings a price premium and draws out otherwise-non-prostitutes, and that a Chicago street prostitute is more likely to provide a freebie to a police officer than get arrested by one.
See this for excerpts: http://freakonomics.blogs.nytimes.com/2009/11/18/superfreakonomics-book-club-ask-sudhir-venkatesh-about-street-prostitution/
Wednesday ~ October 20th, 2010 at 3:33 pm
Psychohistorian
The craigslist figure is probably misleading, since many of those pre-fee ads are likely fake. If you look at pretty much any personal section on CL, you’ll see recurring ads that are identical except for ages and minor details. So simply looking at the volume of ads does not show how many people honestly advertising such services were affected by the price increase.
Wednesday ~ October 20th, 2010 at 6:37 pm
Scott Cunningham
Thanks Adam for the kind words about the paper. We have several other papers (Todd and I) over at my website, which I need to update. The “Prostitution 2.0” (link to an earlier version) paper is now r&r at the Journal of Urban Economics, and another one of the papers was published at a medical journal entitled Sexually Transmitted Infections. We’re still waiting to hear back from a journal on our paper that is focused exclusively on college education among moonlighting escorts (here).
@Psychohistorian makes a good point. I don’t think the problem was spam prior to implementation of the $5-10 fee at Craigslist’s erotic services (ERS). The problem was one of duplicate advertisements. Craigslist for many years did not charge anyone to post an advertisement to ERS. I suspect, though, that because the format of the ads was such that Craigslist always pushed older ads down when new ads showed up, there was a tendency for individuals to over-post. A kind of stage-game prisoner’s dilemma, in other words. And people we spoke to reported doing that – they’d post the same ad with enough variation that the Craigslist screens didn’t catch it. After the implementation of the fee, there was a reduction, but it’s hard to know the magnitudes of the underlying ads themselves. I wouldn’t be surprised honestly if we learned it made ERS more efficient, though – there was a lot of noise in those ads for customers, and after the fee, it was more likely that each ad was a unique ad as opposed to a duplication.
In another article, which is forthcoming in a criminology book on crime and the internet edited by Thomas Holt, Todd and I examined the impact of the early 2009 experiment where after the “Craigslist killer” episode made national news, Craigslist closed ERS down and replaced it with “adult services”. It was a new portal on the same site with more regulations. I was at the time already collecting daily counts of ads at Craigslist so it was kind of a nice piece of luck. We were also collecting counts at an alternative website operated by the Village Voice called “backpage”. Backpage did not have much traffic to its own “adult services” section, but it did have such a section. After Craigslist closed down ERS (and then immediately opened Adult Services), there was an immediate huge reduction in ads posted at Craigslist’s prostitution portal, and an immediate spike in traffic at backpage (see Figure 2). The effect was temporary, and eventually everyone stayed at Craigslist and over the next several months posts climbed back up. But interestingly, posts started to rise at Backpage too. The combination of backpage and Craigslist after it closed ERS and replaced it with adult services ends at almost exactly where ERS was prior to the ban.
On that note, it’s interesting to see this article today that finds some, but not all, of the traffic that was at Craigslist has gone to backpage. Not clear what happened to the rest of it, or if those exchanges simple are not occurring now.
Anyway, thanks for reading the paper and commenting on it! I used to be a qualitative research analyst in another life, and so doing telephone and email in-depth interviews with prostitutes for this project was fun. I can only run so many regressions with the NLSY79 before I lose my mind, so I enjoy collecting data – particular from hidden populations and participants in the underground economy.