Authors: Unknown

Tags: #Psychology/Human Sexuality, #Social Science/Gay Studies, #SOC012000, #PSY016000

Male Sex Work and Society (26 page)

BOOK: Male Sex Work and Society

6.26Mb size Format: txt, pdf, ePub

ads

The reverse is true for Asians, whose passivity and docility are celebrated. Robinson (2007, 2008), Phua and Kaufman (2003), and Han (2006) describe the persistent stereotype that Asian men should be passive, docile bottoms. As with black men, this ethnic sexual stereotype allows the larger gay community to limit Asian men’s socially acceptable sexual expressions. In this case, the counter-hegemonic activity is for Asian men to appear sexually dominant or aggressive. These authors also note that Hispanics are celebrated as passionate, virile lovers who are usually sexually dominant, although not exclusively so, making it difficult to derive quantitative predictions for this group.

Given these racial sexual stereotypes, it is necessary to consider how the values of particular sexual behaviors differ by ethnicity. Phua and Kaufman find that dating “preferences for minorities often are tinted with stereotypical images: Asians as exotic, docile, loyal partners; Hispanics as passionate, fiery lovers; and Blacks as ‘well-endowed,’ forbidden partners” (2003, p. 992). If the market for male sex work mirrors the gay community at large, we expect the analysis will show that black men who advertise themselves as tops and Asian men who advertise as bottoms command higher prices, thus reflecting the value of conforming to ethnic sexual stereotypes.

Exploring the Gay Male Sex Market

In order to overcome a number of problems of sample selection, this study uses data collected from the universe of men advertising online on the largest, most comprehensive, and most geographically diverse website for male escort work in the U.S. (identified herein as site X).
¹⁷This online source has several advantages. First, it allows me to collect information on escort attributes, prices, and information unhampered by the selection problems that would be encountered in a field survey of escorts.
¹⁸Second, I can identify escorts’ home locations, which allows for accurate geographic counts. Third, the escort characteristics I use are entered by the escorts themselves from drop-down menus. This is of particular advantage for exploring the features escorts use to control their pricing models (e.g., body type or hair color). Fourth, website X is free for viewing by anyone connected to the Internet. This ensures that the information provided is for a large general client base and not manipulated to please paying members of a website for a more specific market.

In order to demonstrate that the data source provides sufficient coverage of the online escort market,
table 5.1
compares escorts on site X with the two most prominent competitors from a random sample of smaller cities. The appeal of site X is that escort locality is identifiable, whereas this information is not consistently available on any of the other two comparison sites. Data would be compromised if the comparison sites were included in the study, as the same escorts could be counted individually across the sites inflating the figures.
Table 5.1
shows that site X has a considerably larger escort base identifiable by locality than its competitors. The last column of
table 5.1
shows the number of escorts on site X that could also be identified on Rentboy. The comparison data reveal the many of the escorts who advertise on the comparison websites also advertise on site X, whereas not all escorts advertising on site X were found on the comparison sites.

Escorts list their age, height, weight, race, hair color, eye color, body type, and body-hair type. Advertisements give clients contact information, the preferred mode of contact (i.e., phone or email), escorts’ availability to travel nationally and internationally, and prices for incall and outcall services. Escorts can write about their services and quality in an advertisement’s text.
¹⁹One additional advantage of these data is that the escorts’ claims about their physical characteristics can be confirmed by viewing the pictures posted in the advertisements.
²⁰The analysis is based on outcall prices, but results do not change when using incall prices.

TABLE 5.1

Comparison of Male Escort Websites

Table 5.2
shows summary statistics for the escorts in the data. On average, escorts charge more than US$200 an hour for an outcall, consistent with media estimates of escort services (Pompeo, 2009; Steele & Kennedy, 2006). As one would expect, escorts are relatively young and fit; on average, they are 28 years old, 5 feet 10 inches tall, and 165 pounds (approximately 155cm and 75kg). According to the National Center for Health Statistics, the average man aged 20 to 74 years in the U.S. is 5 feet 9.5 inches and 190 pounds (approximately 154cm and 86kg). Escorts are ethnically diverse; 54 percent are white, 22 percent are black, 14 percent are Hispanic, 8 percent are multiracial, and 1 percent are Asian.

Looking at physical traits, escorts are likely to have black (37 percent) or brown (46 percent) hair; less than 15 percent are blond). More than half of all escorts have brown eyes (55 percent), although significant fractions have blue (18 percent) and hazel (14 percent) eyes. Nearly half of all escorts are smooth (49 percent), 17 percent shave their body hair, but more than a third are hairy or moderately hairy (34 percent). Very few escorts are overweight (1 percent), and relatively few are thin (8 percent); the majority of escorts claim to have athletic (48 percent) or muscular (30 percent) builds. Looking at sexual behaviors, 16 percent of escorts are tops, 6 percent are bottoms, and 21 percent list themselves as versatile.
²¹In addition, 19 percent of escorts advertise that they exclusively practice safer sex. Overall, summary statistics for men in the data are similar to descriptive statistics noted by Cameron and colleagues (1999) for male escorts in British newspapers in the 1990s and Pruitt’s (2005) more recent sample of male escorts who advertise on the Internet.

TABLE 5.2

Summary Statistics for the Escort Sample

Note:
Price is the outcall price posted by an escort in his advertisement.

Analysis

Empirical Strategy

Previous quantitative work analyzing male escorts has not examined prices of male escort services (Cameron et al., 1999; Pruitt, 2005). The data in this study have been analyzed on male escort services’ prices in a hedonic regression; a technique developed by Court (1939), Griliches (1961), and Rosen (1974). The basic technique regresses the price of a particular good or service on its characteristics (such as ethnicity, hair color, body build, etc). This type of regression is widely used in economics, and it is particularly useful for goods that are inherently unique or bundled. This regression analysis is able to tell us how much prices increase or decrease, on average, for services based on the personal characteristics and advertised behaviors of the escorts. Estimated coefficients from hedonic regressions are commonly interpreted as implicit prices because they reflect the change in price one could expect, on average, for a change in that particular characteristic. It is common to refer to positive coefficients as a premium and negative coefficients as a penalty.
²³

Empirical Results

Geographic distribution of male sex workers

Table 5.3
shows the geographic distribution of male escorts advertising on site X. Figures are based on the actual number of escorts by the home locality provided in their advertisements. To my knowledge, this is the first large-scale quantitative evidence on the geographic location of male escorts in the U.S. Size of the escort population varies considerably; there are more than 300 escorts in only one city, New York, which has long been depicted in the media as the largest male escort market (Pompeo, 2009). Atlanta, Los Angeles, Miami, and San Francisco each have more than 100 escorts, but most cities have considerably fewer.

There is a striking trend in terms of location patterns: the number of gay escorts closely follows the size of a metropolitan statistical area (MSA), not gay locality patterns. For example, Detroit is the 11
^thlargest MSA in the U.S. and its gay concentration is 42
^nd, but there are 51 percent more escorts in Detroit than in Seattle, a city with the fifth highest gay concentration index (GCI). Chicago and St. Louis display a similar pattern. Indeed, correlation of the number of escorts with MSA population is quite strong (r = .92), but it is much weaker with the GCI (r = .39). Furthermore, the correlation of per capita escorts with the GCI (r = .69) is weaker than the correlation of escorts with MSA.

TABLE 5.3

Geographical Distribution of Escorts in Random Selection of Cities