The arrival of Uber and other reservation platforms has strongly affected the market of taxi licenses. The chart below provides is an eloquent illustration: in Paris, the average price of a taxi license according to the website of CTT was as high as €230K in 2013 and dropped to barely €125K at the end of 2016 i.e. a decrease by nearly half (45%). Would it be possible to convert these figures into profitability rates for Parisian taxis?
There are 17,137 active licenses which means that the industry of Parisian taxis was worth €3.4 billion in 2013, against only €2.1 billion today. The value lost is mainly transferred to consumers, knowing that alternative platforms are all suffering heavy losses at the moment.
In an article from 2015, Nicholas Buchholz, an American economist, made a calculation for the city of New York, showing that before Uber’s arrival, the industry’s annual income reached $1.6 billion. Since the price of a license in New York at that time was of $1.2 million i.e. 5 times higher than in Paris, and the overall “capitalization” of licenses was of $16 billion (New York accounts for 13,587 licenses or “medallions”). Therefore, the taxi industry has a capitalization multiple of 10X (16÷1.6). For the record, today, the price of a New York license has dropped below $500K, a decrease of 58%.
When applying, somewhat carelessly, the same multiple to the Parisian market, it can be noted that the anticipated gross of taxis at the height of 2013 was around €340M per year, and has dropped to €210M today. Based on these figures, the net normative income for drivers is very low (€23K per year at its height, and €12.5K now – New York taxi drivers were at $117K at their height). The figures are low but we are looking at normative and long-term profits. The arrival of platforms has largely expanded the market. Consequently, the decline in instant revenue for taxi drivers is probably much less. In contrast, the “market” anticipates continued market share decline for registered taxi drivers in favor of other services’ drivers. Furthermore, until recently, the investment was attractive for any taxi driver: buying a license meant being able to take a debt (few individuals, including among artisans, are in a position to raise a loan), moreover a cheap one, while the price of the asset would constantly rise. Operating a taxi meant building wealth: something useful for retirement.
The situation is totally different today.
What about Uber’s profitability? Its value in the stock market is of $70 billion, which represents, based on the same multiple, a normative profitability (EBITDA) of $7 billion. Today, it is losing $2.8 billion. How can this valuation be explained, despite such high losses?
Investors seem to have in mind two criteria in their appreciation: first, Uber is following a logic of “winner takes all.” It needs constant investment to capture market shares as fast as possible. When Uber arrives in a city, it pays “partner-drivers” to go out during the first period, while the network effect is not yet in place, so that the first customers can grab their taxi within 5 minutes. The operating account of each city follows a J curve, and their aggregate account is a sum of J curves, shifted in time. If Uber stopped taking over new locations, there is good chance it would quickly recover its profitability.
Secondly, Uber prices are still very low. A recent paper of a group of economists “Using Big Data to Estimate Consumer Surplus: The Case of Uber,” shows that its activity generates large amounts of “consumer surplus.” This technical term, invented by a remarkable French engineer from the 19th Century, Jules Dupuit, the forerunner of economic calculation, asnwers a simple question: when consumer pay their taxi fare at a price of 100, what price would they have been willing to pay it? If the reserve price is 150, it means that the consumers saves 50 on their fare. Part of the marketing game is to retrieve as much as possible from this consumer surplus, forcing consumers to reveal the real bid price. For Uber, economists have the means to calculate this surplus: the booking system provides them with the impact on demand (elasticity) of price increases Uber realizes during peak periods (surge pricing), a mechanism that is supposed to increase taxi supply over these periods.
But the study shows that the price elasticity is very low (between -0.4 and -0.6: an increase of 10% of the price only decreases the price between 4% and 6%) and the surplus is very significant, around 1.57X of the base price. In other words, all things being equal, Uber has an income potential that could double given its current customer base (from 100 to 257).
Obviously, this calculation is a rough overestimation because things never stay equal to themselves. This simulation assumes that once implanted, Uber will keep its leading and monopoly position, combing both G7 and Les Taxis Bleus’ position, for instance, in Paris. This overlooks the dynamics of the market, namely: the reaction of the regulator; the entry of new competitors; the gradual spread of high-performance software for mapping and matching supply/demand i.e. Uber’s real innovation; the growing demands from drivers, a particularly sensitive segment of the active population, as shown by recent strikes in Paris or Santiago; the evolution of their status under labor law, etc.
It would be a mistake to grant Uber a “mythical” status. When teletext was invented, taxi booking centrals appeared with ladies taking reservations at the other end of the phone. That’s how powerful and highly profitable operators, owning large fleets of taxis, emerged and it was an undeniable improvement over searching a cab by walking in the streets.
Ultimately, Internet was the next leap forward: real-time reservation centrals, with no operators in between. Matching offer and demand was much more effective, not to mention customer returns (i.e. ratings), a disciplinary element which previous systems lacked.
That was it – but it was far from being insignificant. These innovations shattered the corporatist balance, which had once been effective, before becoming increasingly expensive and Malthusian. However, there is no justification in Uber’s attempt to become a global exclusive service provider (there are very few operational synergies in working across several cities: the market is always a local one and only a small number of cosmopolitan travelers find it useful to have the same app whether in Milan or in Chicago). Uber doesn’t really fit in the category of high-tech firms.
Uber’s only strong edge is its financial weight to create the market and fight against legal barriers of the old order. The irony is that it overestimates the barriers to entry it created itself (destroying the old ones wasn’t enough: new ones need to be created) and paves the way for competitors. The current price of a Uber stock is a daring wager.
Note: the following article from Richard Beales in the New York Times (Breaking News, December 22, 2016) is worth mentioning as an excellent insight on the same subject.