Is There Anything Wrong With Surge Pricing?

There has been a lot of chatter about how Uber’s surge pricing is really just disguised price gouging and that they should be ashamed of the mechanic. The backlash appears to have first begun after NYE 2011, and it lead to Uber posting this detailed explanation of surge pricing and forced CEO Travis K to publicly defend his company. The chief complaint, besides the high prices, was that Uber was not being transparent enough with surge pricing… it was not until after the rides were complete that customers realized what they had been charged. To address those complaints, Uber has since implemented a two-step authorization of sorts for surge pricing, whereby you have to enter the surge level into the screen and accept the higher rates. Additionally, on days where surge pricing is expected the app will display a pop-up upon loading warning users. Despite these UI change and pre-NYE surge pricing warnings (2010, 20112012, 2013) New Years Day anger and regret on social media from regretful riders has become an annual tradition.

Recently, a freak snowstorm in NYC led to yet another outcry against Uber, with many claiming that their surge prices in a weather emergency situation is an example of the worst kind of capitalism. Uber responded in the way it always does, defending surge pricing as a way to guarantee a ride is always available and as means to get drivers on the road; that it’s merely a supply and demand mechanic and by raising prices they can quickly boost supply side. Uber has released internal data which supports that claim, but there are still questions whether or not increasing prices 9x is really fair. I think the criticism of surge pricing is more fair in instances like the freak snow storm than during something like NYE, because at least with the former you should know better than to expect normal rates and you have the ability to plan other means of transportation in advance. Either way, I think consumers are being overly entitled to think they shouldn’t expect to pay a premium for a service that has supply constraints in a situation where demand will obviously be high.

I first tried Uber when it became available in Boston, but I didn’t start using it regularly until UberX. Since then, I’ve seen surge pricing go as high as 4x but never higher. Personally I do my best to avoid ever paying surge pricing, but if I really need a ride and it’s below 1.5x I’ll typically bite the bullet. That being said, I’ll check Lyft first to make sure it’s not cheaper. If surge pricing and primetime (Lyft’s version) is higher than 1.5x then I’ll typically try to grab a cab, but that’s nearly impossible in Boston, so I’ll likely end up just taking the T (pro tip: if you really need a ride and UberX has 2.5x-3x surge pricing, non-surging Uber black cars will now be about the same price but typically have shorter waits, nicer cars and more knowledgeable drivers). Even though it sucks when you can’t get a regular priced Uber, I’ve never felt that surge pricing was anything dirty. I simply chalked it up to a company raising prices based on demand in the same way that the travel industry and many others do. However, at least when Uber raises their rates it will (or at least should) actually drive the prices back down by increasing supply.

I would never even think to try taking an Uber on NYE, but this year I was curious to observe the surge pricing dynamics in Boston. My plan was to check in every 30 minutes or so, starting around 8pm, to see where prices were at. I expected things to start picking up at around 8:30ish and steadily rise until 11:30ish, at which point I figured things would die down for a bit due to people being at their NYE destination. Then around 1:00 I expected things to pick back up and peak at around 2:00ish when bars would be clearing out. My apartment is right by Back Bay Station, which for those of you unfamiliar with Boston is a fairly central and popular area, so I hoped it would be a decent barometer for Uber demand.

Around 8:00 I checked for the first time and prices were still normal, which I wasn’t entirely surprised by – I figured things were just getting started later than I expected. As I checked in every ~30 minutes or so over the next hour and a half I was surprised to see that the prices remained at normal, non-surging rates. I had dinner plans with my girlfriend for around 10:00, and I purposefully made us reservations at a restaurant that was walkable, as I was nervous about Uber prices, and I feared finding a cab would be impossible. Much to my surprise, when I opened the app at 9:45 prices were still normal so I figured this would be a good chance to talk to a driver and avoid walking several blocks in the cold :).

I asked our driver if he regularly drove Ubers and he said semi-regularly, but that he was driving tonight because the money was too good to pass up. I asked if he was surprised that prices weren’t surging yet, to which he emphatically said yes, mentioning that he had driving NYE last year and that it was certainly surging by the same point last year. Now he was nervous that he had traded in his NYE for a less than stellar check and an angry wife. My suspicion was that the prospect of a big payday had brought out a lot of drivers who were only semi-regulars, which had flooded the supply side and many would be passengers had made sure to plan alternate modes of transportation for fear of surge pricing.

After dinner we walked home so we could watch the fireworks and walk off dinner, but we still could’ve grabbed an Uber at normal rates. I kept checking the app throughout the night until going to bed at around 2:30 and never once did I see any surge pricing. To say that I was shocked would be putting it mildly. While this was no doubt a big win for consumers, I couldn’t help feel bad for drivers like the one we had earlier in the night, who had given up their own NYE plans because of the allure of a big payday. Granted, it could just be that my area didn’t have high surge rates for whatever reason, and that other areas of Boston saw crazy rates, but who knows. It would appear that things in New York were more or less as expected, as evident by this article.

In order to stop the surge pricing backlash, people have suggested that Uber change how surge pricing functions, which I don’t necessarily agree with. If they were to mix things up, these are the only plausible alternatives in my mind:

1) No surge pricing – and instead have “no cars available” show up

Let’s be honest – this is absurd and not an option. The guaranteed availability of Ubers is huge part of what makes the service so sticky and fules the habit forming/changing. This would be a terrible user experience, particularly for any new customers, who might initially open the app to see “no cars available.” It would also be the worst solution for drivers, as they would be constantly slammed with requests but reap no rewards.

2) Cap surge pricing and if demand still outstrips supply then go to “no cars available”.

In my mind you set the cap at 4x to 5x, because at that point the Uber black cars are much cheaper. This would then allow them to re-configure the UI/UX so that it automatically syphoned some of the UberX demand into Uber black car when their pricing is equal or less, which would help lower the frequency of the “no cars available” being displayed. There is still the issue of the bad user experience if you are presented with the “no cars available” screen, but I think that would be fairly rare at a 5x cap. But the biggest issue is, why should Uber prevent someone from paying more than 4x – 5x surge pricing if they’re willing to do so?

3) Uber doesn’t take a cut from surge rates as Kevin Rose suggests here, instead they give the excess profit to the drivers.

Hopefully this would have the affect of getting even more drivers on the road during common surge pricing times and during extreme events like the snow storm, due to the increased incentive. It might also stop people from hating on surge pricing so much and help Uber shed some of its evil corporation moniker. People would be less apt to think Uber was surging just to make themselves more money, as I feel people often assume. However, I don’t think this is very fair to Uber and I’m not sure how much that excess 20% would really motivate drivers. Unlike the other solutions, I think this one is worth experimenting with.

4) Uber lets drivers set their own rates

Turn Uber into a true marketplace where providers set their rates rather than the service model with marketplace dynamics that they currently employ. The biggest issue here is that the experience would become a lot less seamless and I don’t think this would benefit drivers or Uber. Drivers would have to deal with undercutting and there would be more work involved with getting each passenger, creating more downtime between rides. All that would contribute to less rides and money flowing through Uber and so this is nearly as bad a solution as eliminating surge pricing entirely.

5) Uber continues with their current model

I think this is probably the correct solution and, in all likelihood, what will happen. The product team can continue to work on improving metrics/apps/notifications for drivers to help them better understand when they should be driving for the most demand, as I’m sure they’re already doing. This will eventually lead to situations like the one I witnessed on NYE where despite high rider demand, driver supply is high enough to keep pace. Eventually I imagine their algorithm and ability to get drivers on the road will be so powerful that only unique situations such as the freak snow storm will cause a big surge pricing spike.

Regardless of how you feel about surge pricing, it’s tough not to be in awe of what Uber has built. It’s funny how in that original article about surge pricing during 2011 NYE Travis K called the night a success because they saw requests in the 5 figures, I’d impinge they’re seeing WAY higher number of requests on an average day in most of their biggest cities. However, because they’re adding so many new drivers daily, demand, which would have previously caused surge pricing, no longer does, and in my mind that means the surge model is working.