My oped today is on the economics of sex work. For the most part, researching the piece involved reading other people’s work. But there did come a point when I realised that while the Australian Bureau of Statistics has an official occupational category of ‘Sex Worker or Escort’ (ANZSCO 451813, skill level 5, in case you’re curious), they don’t actually have any data on the earnings of workers in this category. So one evening, I turned to my wife and asked: “do you mind if I surf the Internet looking at prices at escort websites tonight?”. I’m lucky to be married to a tolerant woman.

Incidentally, there are a bunch of unanswered research questions out there. For example, I’m curious to know whether the Swedish switch in 1998 to only arresting clients had the effect of lowering wages (as theory would predict). So far as I can tell, no-one has answered this properly.

Full text below, with hyperlinks to the research papers (you can find the other stuff yourself).

Economics Rules the Street, Australian Financial Review, 8 September 2009

There are few standard findings in labour economics: wages generally rise over the lifecycle, there is a large pay premium for education, and women generally earn lower wages than men. Yet the world’s oldest occupation satisfies none of these criteria. So it is perhaps not surprising that economists have only recently begun to turn their attention to understanding the economics of prostitution, and asking questions such as: Why are wages so high? What policies best reduce sexually transmitted diseases? And is legalisation a good idea?

First, some basic facts. According to recent work by the Australian Institute of Family Studies, there are around 20,000 prostitutes working in Australia at any given time, suggesting that Australia has around twice as many prostitutes as dentists. One US survey found that over their lifetime, about 2 percent of women sold sex, and about 18 percent of men paid for sex.

Generally speaking, prostitution falls into three main categories: brothel workers, escorts (who typically make arrangements online), and street workers. About one-tenth of sex workers in Australia are street workers.

On the street, wages tend to be lower, and the risk of violence higher. A survey of street workers in Chicago by Steven Levitt (University of Chicago) and Sudhir Venkatesh (Columbia University) found that these women earn just US$25-30 per hour. These women are frequently subjected to violence, and mistreatment by the police is common (in one widely reported result, Levitt and Venkatesh noted that a prostitute is more likely to have sex with a Chicago police officer than to get officially arrested by one).

Yet escorts earn substantially more. Analysing around 40,000 Internet escorts in the US and Canada, Lena Edlund (Columbia University) and co-authors observed that escorts earned on average US$280 per hour, which places them in the top 0.5 percent of the earnings distribution. These wages are similar to those found on escorts’ websites in major Australian cities.

Economists have proposed two theories for why prostitutes (at least, those not working the street) earn such high wages. Clearly, part of the answer is that these earnings are a form of ‘hazard pay’, akin to that received by coal miners, window cleaners and motorcycle couriers. Another more controversial explanation is that prostitutes’ high pay is effectively a form of compensation for the fact that current or former prostitutes have more difficulty finding a marriage partner. Intriguing as it is, the marriage hypothesis is not yet compelling, since the evidence shows that many women combine prostitution and marriage.

Another question is how best to regulate the health of sex workers. In this regard, a key contribution that economists have made is to prove that different policy regimes can affect prostitutes’ choice of sector. Drawing on survey data from Ecuador, Paul Gertler (University of California, Berkeley) and Manisha Shah (University of Melbourne) show that mandatory health checks affect whether prostitutes work in brothels or on the street. Their research clearly shows that if authorities want to reduce the incidence of sexually transmitted diseases, they should focus their attention on carrying out on-street health checks.

In regard to its overall enforcement of prostitution, Australia seems to have struck a reasonable balance. For the most part, states and territories ban kerbside solicitation, while permitting escorts and brothels. Since the rates of violence are much higher in street-level prostitution, this effectively discourages the most problematic part of the market (which also happens to be the lowest-paying).

Internationally, the two alternatives that are most commonly proposed are the Swedish model of banning prostitution but only arresting clients, and the United States model (excepting a few states), of arresting both clients and sex workers. Yet both these approaches would be likely to have only a minimal impact on the demand for prostitution, which is driven by the combination of testosterone, cultural attitudes towards women, and high male earnings.

As Harvard’s Jeffrey Miron points out, there is little evidence that banning prostitution reduces violence towards women. More likely, re-criminalisation would merely shift more workers towards street prostitution. Indeed, the Australian experience with prostitution regulation suggests that harsher legal regimes have led to more police corruption, without making prostitutes any safer. While it would be nice to live in a society where prostitution did not exist, this is one area where a ban would be counter-productive.

Andrew Leigh is an economist in the Research School of Social Sciences at the Australian National University.