On the negative correlation between price and restaurant quality
From my email, from an anonymous reader:
On your question of whether there’s a model for the apparent negative correlation between food quality and price:
It is often observed that food quality and the “atmosphere” of a restaurant appear to be negatively correlated. (As I’m sure Taleb points out somewhere, they are not actually negatively correlated, but only appear so because the restaurants that are low quality in both food and atmosphere do not survive.)
I think the apparent negative correlation between quality and atmosphere among surviving restaurants presents itself as a negative correlation between quality and price for two reasons:
1) There are far more people who go to a restaurant because it is familiar and/or convenient (“comfort market”) or fashionable/trendy (“mimetic market”) than people who actively seek out restaurants with good food (“gourmet market”). The demand curve is higher in the market for comfort/mimetic restaurant services (in which food quality does not matter above a certain baseline level of palatability) than in the market for gourmet restaurant services. While restaurants serving the gourmet market and restaurants serving the comfort/mimetic markets are substitutes, the cross-price elasticity of demand between them is probably quite low. When PF Chang’s raise their prices the customers go to Applebee’s, not to the high quality, family-owned Chinese place in the strip mall.
People who actively seek out good food are more likely to know how to cook good food themselves, providing a dimension of competition for restaurants in the gourmet market that restaurants in the comfort and mimetic markets do not have to deal with.
Within the gourmet market I would imagine that price and quality are positively correlated. If it really is the case that price and quality are negatively correlated even within the market of restaurants serving people who care about food quality, then I don’t know how to account for that or why anyone would ever go to the higher priced restaurant, unless purely for variety or for mimetic reasons, in which case those higher priced restaurants are not in the gourmet market but are in the mimetic market (since they would be paying extra for something other than pure food quality, even if the food is quite good – the restaurant is not entirely about the food).
The essential point is that since food quality and restaurant “atmosphere” appear to be negatively correlated, and since most will pay more for atmosphere (comfort, familiarity, fashion, etc) than for quality food, it also appears that food quality and price are negatively correlated.