The trend was made famous by a 2001 study by Richard Florida and Gary Gates, which found that there was a higher concentration of those identifying as gay in cities that were also vibrant centers of technological progress and innovation. Their theory was that places that were open-minded would also be economically vibrant.

But even as same-sex marriage and gay culture is increasingly accepted across the United States, this trend has continued, according to a new analysis from Ralph McLaughlin, a housing economist at the real estate website Trulia. McLaughlin studied the prices and price appreciation of homes, from 2012 through 2015, in neighborhoods with the highest concentration of married same-sex couples, both among male and female pairs. He found home prices increased on average 23% in zip codes with high concentrations of male same-sex couples and 18% for neighborhoods with high concentration female same-sex couples.

What’s even more interesting, and perplexing, about McLaughlin’s analysis is that the “gayborhood effect” isn’t simply a product of the fact that same-sex couples live in cities that have dynamic economies. That’s because when you compare the neighborhoods with high concentrations of same-sex couples to the broader metro areas where these neighborhoods exist, you still find higher prices in the so-called gayborhoods. Writes McLaughlin:

As it turns out, home prices in almost all of the gayborhoods that we look at in this study are more expensive than their metros as a whole. For example, homes in the Castro neighborhood of San Francisco cost $948 per square foot – which is 34% more expensive than the San Francisco metro area ($705 per square foot), while places like West Hollywood, CA and Provincetown, MA are 123% and 119% more expensive, respectively. The only gayborhood that isn’t more expensive than its respective metro is Guerneville, north of San Francisco, CA, which is only 2% cheaper per square foot than its metro area.

Furthermore, when looking at home price appreciation rather than absolute value, McLaughlin found that neighborhoods with high concentrations of female same-sex couples are appreciating faster than than neighborhoods with high concentrations of male same-sex couples. Here’s the possible explanation he offers:

The top gay men neighborhoods are places where prices were already high relative to their metros, and thus were not hit as hard during the housing crash as other less expensive neighborhoods within their respective metros. Second, gay women couples are 2.4 times as likely to have children than gay men couples, so it could be that gay women seek up-and-coming neighborhoods with good schools to raise their children.

Either way, this analysis would seem to contradict Florida and Gates’ original hypothesis that it was the city-wide culture that made places with high concentration of gays more expensive than other cities, because even neighborhood-by-neighborhood, more gay couples mean higher home prices.