The Unintended Consequences of Demand Reduction Strategies for Sex Work

Charles Hill

Saturday, November 1, 2014
One of my students recently forwarded to me an article from the Seattle Times. [i] The article dealt with attempts by the Seattle Police Department to reduce demand for prostitution by targeting the buyers of sex services. “Do you think it will work?” she asked. “I doubt it”, was my off the cuff reply.

Her question was motivated by discussion in my economics class of how government intervention in markets often results in negative unintended consequences. Prohibition in the 1920s, for example, was designed to combat the “moral ills” of alcoholism. In the event, it turned out to be an enormous boon for organized crime, led to widespread corruption in the law enforcement and justice systems, and criminalized the large number of Americans who continued to drink. Paradoxically, prohibition also increased consumption of hard liquor as drinkers tried to get their buzz in a more concentrated form. Nor was all of that liquor safe. Some 10,000 people died during prohibition from drinking poisonous wood alcohol.

Similarly, today the war on drugs is widely seen as a failure. It has filled Federal prisons with low-level offenders at a huge cost to the taxpayer. Thanks to the war on drugs, America now has the highest incarceration rate in the world. It has created enormous profits for drug traffickers, sparked off a truly awful drug war in Mexico (the source of many illegal drugs for Americans), eviscerated many inner city families who have seen parents locked up for minor drug possession offenses, turned physicians who prescribe synthetic opiates into de facto drug suppliers, and all of this while not reducing demand one jot.

So what exactly is the Seattle PD trying to do with its demand side reduction strategy, and more importantly, what might the unintended consequences be? Historically Seattle, like many other cities, has targeted the suppliers of sex services, particularly lower end “street prostitutes” who reputedly ply their trade along places like North Aurora Avenue. More recently, there has been an attempt to reduce the demand for sex services by arresting and then shaming “Johns”. This shift in tactics is a variant of what has come to be known as the Nordic model, since it was first tried in Sweden. In Sweden prostitution has been decriminalized, but purchasing sex has been criminalized. The shift in strategy is rooted in a view that sex workers are victims who enter their trade out of economic desperation and/or because “pimps” coerce them. Interestingly, this view is somewhat at odds with the position taken by many current and former sex workers, who claim that far from being victims, they are free agents who have made an economic choice to enter their chosen profession, and are not coerced by anyone. [ii]

In Seattle, law enforcement officials have stated that they hope to reduce demand for sex work by 20%. The implicit assumption being that if demand is reduced this will somehow shrink the overall size of the market for sex work in Seattle. The main demand reduction tactic seems to be entrapment. Law enforcement officers post fake ads on, a well known publisher of ads for sex workers. Unsuspecting respondents are then lured to a hotel room, where money changes hands, and an arrest is made. By publicizing their efforts, law enforcement seems to be attempting to put the Fear of God into “Johns”. In economic terms, they are increasing the risk-adjusted price of purchasing sex services.

Economic analysis suggests that this policy may not have the desired effect. In fact, it might actually have the perverse effect of increasing the supply of sex services in Seattle. To understand why, I’m going to build an elementary model of the market for sex services. This model is based on a few simplifying assumptions.

First, for the time being assume that the majority of sex workers are independent free agents. They are not coerced and have freedom of entry into and exit from the market. This is not to deny the existence of pimps or coercion in some cases. However, contrary to popular mythology, there is a growing body of evidence on the market for sex work suggesting that the majority of sex workers are indeed independent free agents, so the assumption seems reasonable. [iii]

Second, assume that sex work is just that, work. [iv] Sex workers enter the market in order to earn certain income level. Unless sex work is different from every other source of labor (which I seriously doubt), sex workers provide their services because the income they earn from such work exceeds their opportunity cost. In other words, to them selling sex is a better economic option than their next best alternative, which might be waiting tables in a restaurant, working in a telephone call center, or subsisting on welfare.

Third, assume that once they have reached their desired income level, sex workers will be increasingly reluctant to work more unless the price is high enough. This is simply an example of the well known backward sloping supply curve for labor. Once workers have achieved a desired income level they increasingly substitute leisure time for work. The supply of sex services, in other words, becomes more inelastic as prices rise.

Fourth, for the time being assume that sex services are a commodity, and that all sex workers sell the same basic service and charge the market price (which they can see from scanning ads on services like This assumption is somewhat unrealistic. The market is by all accounts segmented. I will relax this assumption in due course.

Now lets look at the market for sex service. Figure 1 gives a graphical depiction of the model. This graphs the hourly price of sex work against the number of hours supplied to the market. The supply curve gets progressively steeper until it bends back on itself (i.e. the supply curve for labor is backward sloping at higher prices). At high hourly rates, sex workers hit their required income level after working fewer hours and increasingly they substitute leisure for work. The price of sex services is determined by the interaction of supply and demand. The initial demand and supply curves before any law enforcement actions are D1 and S. The equilibrium price is P1 and Q1 hours of labor are supplied to the market. Note that I assume the initial equilibrium is in the backward bedding portion of the supply curve. Note also that workers do not start supplying sex services to the market until the price they can get for their services exceeds P0. This price represents the opportunity cost for those willing to enter the market at the lowest price – P0  is what they earn in their next best option (e.g. waiting tables).


Next imagine that law enforcement comes into the market with aggressive demand side reduction tactics, such as entrapment and “john shaming”. This raises the risks to buyers of entering the market, results in less demand at each and every price, the demand curve for sex work shifts down from D1 to D2,  and prices fall. So far so good for law enforcement. However, we have to consider how the sex workers themselves will respond to declining demand for their services and lower prices. Now they are not earning their desired level of income, but since the income from sex work for many still exceeds their opportunity cost, they do the logical thing and offer more hours to the market. How many additional hours will they be prepared to work? Probably enough to help maintain their standard of living after factoring in the impact of the demand side reduction strategies of law enforcement. As a result, paradoxically the supply of sex work actually increases from Q1 to Q2. Moreover, the reduction in the equilibrium price from P1 to P2 ironically creates an incentive for more “johns” to assume the legal risks associated with entering the market. This is not, I suspect, the intention. The demand side reduction strategy has failed to achieve its goal.

At this juncture it is important to note that you get exactly the same result if you drop the free agent assumption and assume that 100% of sex workers are coerced and managed by “pimps”. Only in this case, it is the pimps who want to maintain a certain living standard, and they force sex workers to work longer hours so that they can maintain that standard. In such a scenario, the situation for the workers themselves may be significantly worse than it was before the law enforcement action. They are now coerced into working longer hours doing a job that they do not want to do. This again is clearly not what law enforcement intends, but it is a probable unintended consequence of aggressive demand side reduction tactics.

Lets push this argument a little further and relax the assumption that sex services are a commodity product. Imagine two types of sex worker; the commodity provider who charges a low price for hourly services, and the differentiated provider who offers extended services in addition to sex, such as companionship, and charges a higher price. Lets call the former streetwalkers and the latter escorts.

Like all suppliers of a higher quality offering, reputation is a key ingredient in justifying and maintaining the brand equity and superior pricing power of the escort. The escort trades off his or her reputation. How is reputation established in this market? How is quality assured? It turns out that there are well-established mechanisms for doing this. This should be of no surprise to anyone schooled in economics, for institutions frequently emerge in markets to assure quality and rate brands (e.g. JD Power ratings in autos, BusinessWeek ratings of Business Schools). Sex work, it transpires, is no different. Internet sites have emerged that not only help to make a market between escorts and their clients, but also serve to assure quality and safety (for more detailed discussion of this, see [vi]). On these sites, clients can post reviews of escorts, and escorts can red flag troublesome clients. A stream of good reviews enhances the brand equity of an escort, enabling them to charge a higher price and/or attract more clients. Because they have reviews, escorts are known entities with a reputation value.

The same cannot be said for streetwalkers. Some literally walk the streets, but more often nowadays they advertise on lower quality Internet sites like where there is no option for reviews. This means that the streetwalkers have a harder time establishing a personal reputation and brand. Moreover, the lack of a mechanism for sharing information about troublesome clients also makes this a more vulnerable population. Streetwalkers compensate for their lack of brand equity by charging a lower price for their services.

The fact that the streetwalker market is relatively anonymous on both sides of the transaction opens the door to entrapment tactics. If the client knows nothing about his or her desired provider other than a picture and brief description posted on, if there is no mechanism for sharing information about prior entrapments, it is far easier for law enforcement officers to lure a prospective “john” into a rendezvous where he can be arrested. On the other hand, the entrapment strategy is less likely to work for escorts with well-established personal brands.

So what happens when law enforcement tries its demand side reduction entrapment tactic against streetwalkers in this segmented market? Demand for streetwalkers may shift down as depicted in Figure 1. However, there will be some demand shifting, with clients that can afford the outlay gravitating towards well-known and safer brands (i.e. escorts). In short, aggressive demand reduction tactics will perversely drive up demand for the higher quality product where quality is assured and safety is more guaranteed. As a result, escorts will be able to charge a higher price for their services. Again, this is not the intended consequence, but economic logic suggests it will happen.

As for streetwalkers, the prior analysis still holds, only more so. The decline in demand is likely to be proportionately greater for a given level of police action given that clients have a safer alternative in escorts and some will shift their buying behavior. This may lead to streetwalkers increasing the supply of hours beyond that illustrated in Figure 1, which as the prior analysis demonstrates, compounds rather than solves the perceived problem. Yet again, demand side reduction strategies have not achieved their objective.

If demand side reduction strategies do not work, you may be wondering what law enforcement should do here to reduce the size of the sex economy? As an economist I think this may be the wrong question to ask. What we perhaps should be asking is what is wrong with a contract to sell sexual services between consenting adults who enter voluntarily into the transaction, neither of whom is harmed by the exchange? Objectively speaking, many argue the answer is nothing at all. No harm is done here, so why should this be regarded as a crime?

One may object to this line of reasoning on moral grounds, which are inherently subjective, but then you are imposing your subjective morality on others who might disagree with you. Alternatively, one might argue that sex work represents the objectification and exploitation of women, and that the seller is always harmed, even if they don’t know it. The problem with this line of reasoning is that it might be at variance with the sex worker’s view of the situation. One might ask who are you to tell another person how to think, feel and act? What gives you the right to reinterpret their experience through your worldview?

On the other hand, it’s clear that forced labor of any form is an abomination, whether the coercion involves sex work, making shoes in a Thai sweet shop, or mining for diamonds in Africa. Forced labor is rightly classified as illegal since it represents a violation of basic human rights and is simply slavery. Slavery should be prosecuted with the full force of the law. But demand side reduction strategies are not well designed to do this. As such, they manifestly fail to deal with what could be a real problem in the market for sex work, although one of an unknown magnitude. Law enforcement, in other words, is going after the wrong target. If you are worried about pimps coercing women into forced sex labor, some of whom are underage, go after those pimps instead of focusing on soft targets that generate publicity but do not solve the human rights problem, and may have perverse effects counter to those intended.

[i] Sara Jean Green, “County’s ‘Buyer Beware’ Program Aims to reduce Demand for Prostitution”, Seattle Times, October 15th, 2014.

[ii] The Economist, Prostitution a Personal Choice, August 9th, 2014.

[iii] The Economist, “More Bang for Your Buck”, August 9th, 2014. Maggie McNeill, “Lies, damned lies and sex work statistics”, The Washington Post, March 27th, 2014.

[iv] For example, see Laura Agustin, Sex at the Margins: Migration, Labor Markets and the Rescue Industry, Zed Books, 2007. And Melissa Gira Grant, Playing the Whore: The Work of Sex Work, Verso, 2014.

[v] Drew Atkins, “Seattle’s Online Sex Industry: Can You Regulate and Invisible Economy”,, October 22, 2014.


Charles WL Hill
Seattle, Washington State, United States
Charles Hill is a Professor at the University of Washington’s Foster School of Business in Seattle where he has taught since 1988. His areas of expertise include business strategy, international business, and micro economics. He is also an avid sailor, skier, and mountaineer.


Un pensamiento en “The Unintended Consequences of Demand Reduction Strategies for Sex Work

  1. Check out our website and go to our “Operation Do the Math” to see the numbers of actual arrests of males from the FBI website, comparing the number of overall arrests made by cops each year (average number of arrests per cop per YEAR 2013: 13.11) men of eligible age in the US: approx. 107,000,000 between age 20 and 79… Number of males arrested in 2013 for prostitution: approx. 13,000. If each sex worker has but 10 ‘unique’ clients a year, and there are only 100,000 sex workers, that’s 1,000,000 men. How many men can cops arrest if cops can only make 13 arrests per year? FBI says there are approx. 765,000 state and local cops, and in 2013 they made approx.
    12 million arrests for all reported crimes and non reported ‘crimes’ combined. Of the arrests made, only about 18% were for reported crimes, the remaining arrests were for non reported activities. End the demand? What a joke! Why not go after the perpetrators of rape and sexual assault where there was a victim who asked for help?


Introduce tus datos o haz clic en un icono para iniciar sesión:

Logo de

Estás comentando usando tu cuenta de Cerrar sesión / Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión / Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión / Cambiar )

Google+ photo

Estás comentando usando tu cuenta de Google+. Cerrar sesión / Cambiar )

Conectando a %s