The True Economics of Uber’s Surge Pricing
Is anyone really benefiting during peak surge pricing periods?
Photo: JMV http://bit.ly/1dg8LKY
Views expressed below are based on my own opinions and conclusions, which are not backed by actual data. It is just an anlysis of what I perceive may be happening. My opinions only. Rates in Philadelphia peaked at 3.0x Saturday night, not 6.0x as previously stated. 6.0x rates existed in NYC, not Philadelphia.
The supply and demand challenges Uber faces are fascinating. As more people are learning of Uber, the overall demand is increasing and in many cases triggering “Surge Pricing.” According to Uber, this increase in pricing is “to get more Ubers on the road.”
I love Uber. I’ve used it ~10 times and have had consistently great experiences. Out of those 10 rides, I paid less than $20 total out of pocket due to promotions and referring many people to the service. The Philadelphia rates are not outlandish, but are priced at a premium that makes you think twice before choosing it over a traditional taxi.
As Uber grows in popularity, the commentary on surge pricing is starting to pick up as well. Last night in Philadelphia the surge pricing was at 3.0x standard rates and fares reached levels over 7x standard rates in other snow covered cities. A Twitter search during these time periods show the community’s sentiment towards the practice. Despite this making economic sense, the result is nevertheless a lot of unsatisfied and disappointed customers during these periods.
Uber really has a unique problem on its hands around this issue. I understand the purpose of the surge pricing, but its messaging and impact on customers is quite significant. Uber communicates that raising rates is important because it will get more drivers on the road. In theory, this makes sense. If Uber drivers are faced with a situation where they are receiving 3x or 6x increases in revenue for the same trip, it should attract more drivers on the road and that is how it is displayed to users within the app during surge pricing.
What is troubling for Uber is they have no direct control over the amount of rides they can supply. Part of the surge pricing is increasing revenue potential to attract as many drivers as possible, but in reality I think it is more of a panic signal to preserve the integrity of the product. I would argue that during times of surge pricing, most all available Uber drivers are already on the road. These are periods where if you are making a living driving for Uber, you cannot afford to not be on the road. These Friday and Saturday nights are the most lucrative for Uber drivers as they are likely to maximize their billable driving time.
Supply and Demand
For the sake of discussion, we will consider non surge pricing times a period where supply and demand are at equilibrium. During these time periods, all drivers and riders are paying the predetermined standard rates, which include minimum fares.
If demand increases, it makes sense for Uber to charge more under its dynamic pricing model. The result of a true increase in demand, as pictured above, is an increase in the number of rides (Q2) and an increase in fares (P2) as well. Theoretically, more people are paying higher rates for Uber rides which results in increased revenue for both Uber and drivers.
Uber messages these periods of surge pricing as incentives to get more drivers on the road. The principles of economics would state that more drivers would be incentivized to be on the road if customers were paying higher fares. However, during these periods of peak demand, supply is already close to maxed out based on the earnings potential for these time periods. Because the Uber driver base is finite, during high volume periods, true demand has the potential to move beyond the supply curve entirely, unless drastic price increases are applied.
If a shortage, where demand exceeds supply at a current price point, ever happened during crunch time in a major city, Uber would be over. If I wanted to take an Uber at 11:30 PM in downtown Philadelphia, New York, Boston, San Francisco, you name it, and no cars were available, the damage to the brand would be paramount. Yes, people become frustrated when 6x+ rate multiples are happening, but cars are technically still available. I’d argue virtually next to zero rides are taking place during these high multiple periods, but the service still exists. Once drivers become available again, the prices will return to normal, but it has nothing to do with wanting to get more drivers on the road. This instead is Uber’s version of brand and product protection.
I watch and read everything I can when it comes to Uber. The company infinitely fascinates me. I remember Travis saying in one interview something along the lines of it is definitely possible to get rid of surge pricing, but then we cannot guarantee cars are available. My view is simple: even though cars exist on the system during high surge pricing periods, they are not accessible to 99% of the user base. Yes cars are “available” to everyone, but in reality they are “accessible” to few. With a $72 minimum fare, Uber quickly goes from “Everyone’s Private Driver” to “The super rich’s ride home.”
Most criticism of these surge pricing periods are assuming users are actually paying these extreme rates and Uber is banking tremendous amounts of money. Although they would not likely admit it, these are pain periods for Uber where revenue is likely declining. When rates hit 6x standard fare, they know next to no one is actually paying these rates sans carelessness or emergency. Uber has made the conscious decision to increase rates in crazy ways and have virtually no use of the service, rather than have no cars available at the busiest times in the busiest places. They’ve made it a priority to offer a consistent, fast service and price is clearly a secondary concern.
Uber does move to surge pricing based on market economics. However, it is not strictly a market driven economy. As Uber becomes more popular, there is a real risk during peak hours that the demand curve can move beyond the supply curve, which would result in a “zero cars available” scenario. What Uber’s algorithms do to prevent this from ever happening is use extreme price inflation (P3 above) to create an artificial price floor above the price point where equilibrium would be if the supply curve was long enough to intersect the new demand. The price surges remain small during forecasts where the shifts in demand will intersect with available supply. However, when algorithms predict the shift in demand will move beyond the supply curve, extreme measures are necessary. The new price floor barely nips the new demand curve and serves the very few customers willing to pay those high multiple rates. The biggest problem this price floor causes is it leaves many of the available drivers (dotted red portion of supply curve) with empty cars during the busiest times of the week. This explains why even during periods of 6x price increases, there are many cars that show up as available on the map. The availability is there, but nobody is pulling the trigger at these rates.
I get it, Uber cannot afford to have zero cars available in downtown centers, ever. The damage that would cause to the company’s reputation is more severe than the high multiples charged during these peak periods. I’d argue the high multiples are punishing drivers, as there will always be empty cars during these periods that would be easily occupied at lower prices. This really is Uber’s biggest challenge. As the amount of drivers continues to grow, so will the length of the supply curve resulting in less extreme price surges during peak hours that are more market based.
Uber needs to make a decision. Many users of the service are becoming accustomed to these extreme price surges and are thinking twice about even opening the app during peak hours. Uber knows the problem it faces well and is trying to get as many cars on the road as quickly as possible. With minimum fares, Uber would always benefit in situations where the supply far exceeds the demand. The cost to put a new driver on the road is very small in the scheme of their strategy. The more cars they are able to put on the road, the quicker pickup times are and the more revenue they collect while eventually even being able to reduce prices.
Uber’s response to these situations to its customers is sympathetic. The customer service canned response to users is one of this is only to incentivize drivers to come on the road at higher rates. This may be partly true, but its only a small part of the reasoning behind the extreme price surges. The 3x and 6x surges occur to protect the quality of the product, but at the same time they are punishing drivers with empty cars during periods where they could be making a lot of money.
It’s now a race. A race for Uber to increase supply of drivers and increase the length of the supply curve. However, if they are not able to do so swiftly, the result is an increasingly disgusted customer base and equally disgusted cohort of drivers that aren’t truly maximizing revenue from the higher rates. Is the answer to lower surges and run the risk of zero cars being available during peak demand? If history is any indication, according to Travis, probably not. The surge pricing creates social media buzz and the illusion that everyone and their dog is using Uber, whereas zero availability at key times would mean failure. Uber is trying desperately to get more cars on the road through financing deals for new drivers, but I’m afriad that is not enough. With all the technological innovation in the world, the problem that is making Uber less relevant is one of not enough drivers in seats when it matters most.
Right now Uber is walking a fine line between growing as a world changing technology and becoming less relevant by pricing out its users and diminishing goodwill in the brand. I have a lot of ideas about ways Uber can attack the issue, but it will be interesting to watch what will develop in the near term.