I live in a condominium which some time ago changed the rules that permitted dogs to one that forbade them unless grandfathered. I researched the issue a bit and discovered that 39% of households owned one or more dogs. I don't recall the source; don't know the veracity but I wonder how to calculate what impact that would have on the resale value of my condominium if every dog owner refused to purchase any unit prohibiting dogs. (I know that that's not 100% true. Just looking for a starting point.)
It seems there are two markets for condominiums. Units that allow dogs (Market Dog) and units that do not (Market Sad). My black lab mentioned that the latter group are probably sad folks. Ok – my econ jokes in class aren’t any better.
On the demand side, you have two different consumers. Consumers with dogs and consumers without dogs. Consumers without dogs can buy from either Market Sad or Market Dog. Assuming everyone follows the rules, consumers with dogs can only buy from Market Dog. If someone with a dog wanted to move into Market Sad, they would have to give up their dog or find a long-term home for their dog, effectively making that consumer a non-dog owner.
Unless the dog owners are willing to give up their dog, they are already refusing (or prohibited) to buy a unit that does not allow dogs. Therefore, there would not be a change in the market price, since this is already worked into the two prices.
Since the owner(s) of the condominium changed the rules to prohibit dogs, one can ask if there was a financial incentive to doing that? Are there enough non-dog owner consumers that only want to live in Market Sad?
One idea to test this would involve a lot of data collection. Ideally, you’d gather data on the selling price of condos along with other characteristics (sq feet, level of finish (e.g. Quartz counter tops), amenities offered (pool, garage, etc), area characteristics (number of restaurants, crime rate, …) and, most importantly, does the unit allow dogs. Additional information to understand the availability (price) of substitutes would be helpful, such as the number of condo units listed at the time of sale or the average selling price of other condos the during the month of sale, etc. If you had enough observations (1 obs = 1 sale – here think 1,000’s of condo sales, not 10-20), you could run a regression to see if the market price for units that allow dogs have higher or lower prices, when controlling for the other characteristics.
A study similar to the above outline was conducted. https://digital.sandiego.edu/cgi/viewcontent.cgi?article=1131&context=osp-researchweek . However, the results would be specific to the market and time period.
Levitt and Syverson (Jan 2005 - https://www.nber.org/papers/w11053.pdf) examine housing prices in Chicago, studying if using a real estate agent results in a higher price. If you read Freakonomics, you may recall this research. Bernheim & Meer (Jan 2012 https://www.nber.org/papers/w13796) also studied real estate price differences.