Issue No. 21: The Carvana Saga Continues, Passing the Buck & Clipping the Ticket (AirBnB & SmartPay)
Business lessons from the Carvana saga, and the opportunities still to be found in payments.
The Carvana Saga
Controversial used car dealer Carvana has once again caught the imagination of the wider investing community this week. The name has decidedly split public opinion over the past 5 years. On one side of the divide are the optimists who, despite their better angels, believe in the growth story and the ability of the energetic CEO Ernie Garcia III to eventually reach a sustainable business model. On the other side, you have people with the ability to think critically.
For those of you less acquainted with the name, it has an alluring growth story. For many years now, technology entrepreneurs have been attempting to rationalise the sale of big ticket products/services - vehicles and residential real estate being the primary examples. Both of these markets seem to defy all attempts at reformation and technology led efficiency.
The reasons for this are not easy to divine. My best explanation would be that these types of markets resist centralisation because they are essentially local markets which can only support local intermediaries (the widely reviled real estate agent and used car salesperson). As of this point in history, most attempts to bring economies of scale to bear on these markets have failed. Quite spectacularly too.
Carvana is a play on significantly improving on the local used car dealership. The concept is simple: take the best eCommerce practices and apply them to the used car market. The core of what makes this idea even conceptually appealing is the promise of scale economies. Online retailers - and Amazon is the prime example here - use purchasing scale, and efficiency in the customer experience to offer a superior range of products, at a superior price point, and without the need of going through the hassle of physically purchasing the goods.
This is Carvana’s promise to customers. By drawing on a national inventory of vehicles they should be able to offer a better selection. They do this by algorithmically buying vehicles from the the public and auction yards, then detailing these vehicles themselves. By removing salespeople from the sales process they should be able to sell cars cheaper than their brick-and-mortar competitors. These cost savings kick off a virtuous cycle: cheaper vehicles beget more sales, and more sales beget greater cost savings. Some of these cost savings get reinvested into another layer of services like at-home delivery. Delivery is backed up with a return policy on par with other successful online retailers. Amazing and incredible.
Naturally this model will only show profitability at significant scale. While Carvana’s operations on a single location basis are deeply unprofitable, operational leverage should come into play once sufficient scale is reached. Ever greater sales, with ever greater unit economics (scale economies should drive per unit profitability up) should result in a solidly, if not wildly, profitable operation.
Beyond these impressive operations are the opportunities for layering services on top of the vehicle sale. Carvana also offers financing, insurance, and even vehicle add-ons to customers. This, however, pales in comparison the promised land of 3rd party sales. The prospect of opening up Carvana’s platform to 3rd parties has all the promise of high margin revenue vis-a-vis Amazon’s experience in selling advertising, and implementing merchant servicing and fulfilment fees.
The Other Side of the Coin
The problem with the story I outline above is that reality supports very few of those promises.
It is true that parts of Carvana’s model support economies of scale. There do seem to be genuine cost savings in removing sales people from the process of purchasing a car. There is certainly superior selection on Carvana’s platform as compared to a traditional dealership. Services like at home delivery are nice.
The fact of the matter is that even Carvana’s first, and most mature, market (Atlanta, Georgia) they barely managed to capture 3% of overall used car sales. It is also still unprofitable on a stand alone basis. This is after nearly 10 years of operation. The unit economics of a vehicle sale (less financing and services revenue) is unprofitable. To the extent that Carvana does make money at the unit level is really related to the profitability of reselling the loans they originate - many of which are at the subprime level. This is somewhat analogous to the adverse selection experienced by the BNPL’s: the customers resorting to these fringe services are those who have few options.
The problem’s faced by Carvana in the wholesale purchasing of used vehicles is analogous to the issues faced by the likes of Redfin and Opendoor in the real estate market. The continental United States is not a single real estate market; it’s thousands of markets. The same is somewhat true of used vehicles. The anecdotal evidence of vehicle owners selling their cars to Carvana at rates way above what the local market would support are numerous. Furthermore, each state has their own inspection and titling regime. Failure to adequately comply with various state regulations has constantly resulted in Carvana being suspended from operating in states like Illinois and North Carolina. Ultimately, the big data approach to rationalising one side of this big ticket market actually results in a less rational approach. This was especially true during the enormous volatility seen in those markets in early 2022.
What gives Amazon, and even operators like UPS, scale economies in last mile delivery is that every additional package on a delivery route is incremental on a relatively known cost basis (fuel and driver costs notwithstanding) ex ante. At a sufficient level of market saturation this can be a decently profitable operation, and demand for delivery services will naturally coalesce around low cost operators. Carvana’s delivery is somewhat less efficient:
Of course there is the matter of trucking all of these vehicles around the country for consumers who are explicitly looking to reduce costs and get access to finance that they haven’t been unable to access anywhere else.
So to parrot a phrase, this is a subprime lender masquerading as a used care vending machine company.
The Anti-Value Investment with the Balance Sheet from Hell
Keep reading with a 7-day free trial
Subscribe to Buyback Capital to keep reading this post and get 7 days of free access to the full post archives.