The Utility and Limitations of Uber Ratings in Reducing Asymmetric Information
When you enter a vehicle of an unknown driver, you know little about them: from important knowledge such as how safe they will drive, or if they will get lost, to more trivial aspects, like what type of music they will play or if they will want to have a conversation with you. Similarly, if you are a driver, you do not know what the etiquette of the passenger will truly be, beyond any assumptions you may have about them tied to where you are picking them up, what time it is, or what they look like. Star ratings on rideshare apps like Uber seek to minimize this asymmetry, by providing a way for drivers and riders alike to gain some insight on what to expect based on previous experience.
While this rating system may work to incentivize better behavior for users and drivers, giving them power to hold each other accountable, how much of an impact does it truly have? How accurate are the ratings at indicating the quality of the passenger or driver? In a Business Insider article, Uber and Lyft driver Clarke Bowman has experienced that the ratings typically hold true, warning passengers with a rating between 4.5 and 4.59, to “not be surprised if you have to wait a very long time for a ride, as drivers are probably consistently rejecting your requests.”
It could be argued that the process of becoming a certified driver, and the higher standards regarding ratings and maintaining the ability to drive for the app make the exhange more asymmetrical, as passengers have a better expectation for having a quality driver than drivers do for any passenger they pickup. As drivers are working as a source of income, there is less variability in actions they take that would make them less desirable, whereas passengers have more room to exhibit undesirable behavior. The more strict standards for drivers based on ratings do come with some caveats, however, as there have been lawsuits that claim this system is discriminatory.
Using uber does not have the same textbook market asymmetry as, say, selling an old used car you know the true value of to an unknowing buyer, but the rating system does offer some peace of mind in acting as a signal of trust. In fact, Bowman justified the validity of passenger ratings by comparing the inevitable one star ratings of lower range passengers to reviews of purchasing a product: “that $12 Bluetooth speaker on Amazon didn’t earn a one-star review because it worked amazingly well. It earned a one-star review because it quit working, wouldn’t connect, and had poor sound quality.” Another important note to consider is that, while Uber ratings do not impact the actual price of the ride, it may indicate that a passenger will leave a higher tip. As of last year, it was announced that riders with low ratings may be barred from using the app–a consequence that has already been in place for drivers with low ratings. The Uber rating system may not be perfect, and ratings will never be able to always signal quality, but they do offer an additional layer of knowledge going into a ride, aggregated by continued use.