The profession has long puzzled over the question of why commissions in the United States are so much higher than elsewhere in the developed world. Economists George Akerloff and Robert Shiller hypothesize in their upcoming book, Phishing for Phools, that the market hasn’t driven down the price of commissions—which average 6% or more than $14,000 for an existing home—because home buyers tend to be very inexperienced.

They point out that by the time the typical American turns 60, there’s an 80% chance that he owns his own home. The statistics also show that Americans stay in their homes for a long time—24 years on average. Akerloff and Shiller write, “These two numbers mean that a significant majority of Americans buy at least one home in their lifetimes; they also mean that the purchase of a home, for most of us is quite infrequent.”

That means that most people aren’t hiring real estate agents and brokers more than once or twice in their lives, making them uniquely incapable to gauge the quality of what they’re buying. But there may be another reason why commissions in America are so much higher than elsewhere in the world: good old-fashioned collusion.

That’s the claim made in a new working paper published Monday by the National Bureau of Economic Research. Economists Panle Jia Barwick, Parag A. Pathak, and Maisy Wong studied real estate agent commissions of 653,475 residential listings in eastern Massachusetts from 1998-2011 to determine why commissions in the residential brokerage industry are, “relatively uniform despite low entry barriers and advances in technology that have reduced search costs for buying and selling properties.”

They hypothesized that this was because of the fairly unique nature of the residential real estate market, in which sellers of real estate pay both the commissions for their listing agent and that of the buyer’s agent. Given this arrangement, it’s necessary for buyer’s agents and lister’s agents to cooperate to make a sale. Given this arrangement, it’s possible that dominant brokers in the industry could refuse to work with upstart competitors who are setting their commissions too low.

And when they looked at the data, they found that brokers charging commissions of less than 2.5% were in fact discriminated against. They write:

Consistent with real estate agents steering buyers to properties with high commissions, we find that if a property has a buying commission rate less than 2.5%, it is 5% less likely to be sold and takes 12% longer to sell.

This isn’t the first time economists (or regulators) have scrutinized the highly cooperative nature of the real estate brokerage industry. But a 2008 settlement between the Justice Department and the National Association of Realtors did nothing to ban this practice. Perhaps it’s time for regulators to take another look.