Matt is right that “nightlife”, like a lot of other industries, often clusters. People like to have options when they go out, and they like going where there are other people around, so watering holes that cluster together often find that they do better than they might outside of a nightlife cluster, despite the impact of increased competition within the cluster. And Matt is right to say that when you limit liquor licenses in an area, you cut off the potential gains of clustering to the consumer, you cut off the potential gains of clustering to the businesses, and you cut off the potential gains of competition to the consumer, since you effectively hand existing businesses a great deal of market power.

But I’m constantly reminded of another side of this equation whenever I’m in London. London, like cities and towns across the British Isles, is filled with pubs. They vary in type, quality, and clientele. I was very lucky this time around to find a near-perfect gastropub just a five minute walk from my flat. It was quiet and well-maintained with a great menu, and while there were always people there, there was also always a free seat. Kids were welcome during the day, as were dogs. Every time I went I thought to myself how great it would be to have such a place close by back in Washington. And every time I thought that, I immediately reminded myself that such a place, back in Washington, would be perpetually packed and fairly unpleasant. In the Washington area, you can’t have a place that’s both really good and quiet in a neighborhood-y sort of way.

That’s largely because it’s very difficult to open new bars. And the result is a pernicious feedback loop. With too few bars around, most good bars are typically crowded. This crowdedness alienates neighbors, and it also has a selecting effect on the types of people who choose to go to bars — those interested in a loud, rowdy environment, who will often tend to be loud and rowdy. This alienates neighbors even more, leading to tighter restrictions still and exacerbating the problem.

Sadly, this is the kind of dynamic that’s very difficult to change. No city council will pass the let-one-thousand-bars-bloom act, and neighbors can legitimately complain of any individual liquor license approval that it may lead to some crowded, noisy nights. It’s interesting how often these multiple equilibrium situations turn up in urban economics. In general, they seem to cry out for institutional innovation. It’s a little surprising, for instance, that we don’t see more “private club” type bars, that restrict entry by price or membership, in order to preserve the quiet along with the quality. Or maybe we do, and I’ve just not been invited to join them.