Contextualizing the Root Carvana Partnership
Dependence on a bigger entity comes with risk. But this is mostly opportunity. It provides access to one of the largest distribution platforms and it serves as a blueprint for more.
Disclaimer: The information contained in this article is not and should not be construed as investment advice. This is my investing journey and I simply share what I do and why I do that for educational and entertainment purposes.
This article is entirely free to read.
For those interested, I posted some high level comments on LMND 0.00%↑ investor day on X. I thought I’d spare your inbox with another piece on that company.
TLDR Summary
Too much entanglement with another entity can be a risk for any company. It creates dependence which might put them into a weaker position down the road. I do however see almost only opportunity rather than risk in the partnership between Root and Carvana.
Of all economic activities surrounding the purchase and use of vehicles, providing insurance the most valuable one. Market value pertaining to car insurance matches combined market value of companies manufacturing cars, selling new/used vehicles and providing financing.
To leverage this opportunity, an insurance carrier needs to be skillful at both pricing risk and finding policies to underwrite. Developing powerful distribution channels is key on the latter. For Root, Carvana is a central pillar for their distribution strategy.
Firstly, even if Carvana, the stock, may not have much upside left, there is likely still a lot of upside in Carvana, the business. This is both cyclical from a rebound in used car sales and structural from a consolidation of the highly fragmented used car industry.
Carvana processes about 400,000 retail vehicle purchases annually, about 1% of every car purchase in the US, whether new or used. It’s one of the largest automotive distribution channels in the country with the ambition to become much bigger in the future. And in many instances they sell a vehicle, Root gets to make a quote to Carvana’s customer with very little additional cost. Policy acquisition costs are central to any insurance business. For established players, they account for about 10% of revenues and up to 50% of their operating expenses.
To illustrate the size of the Carvana opportunity in simple terms: If Root’s policies in force (PIF) ramped to an equivalent of Carvana’s 1% distribution share over time, they’d have 3 million policies, four times the size of the business today.
Secondly, Carvana is highly incentivized to make Root a successful company. Could they eventually decide to break the partnership and use other carriers? Nothing is impossible. But their preferred stock and their warrants could eventually give them a third of the company. These instruments might cause dilution for Root shareholders. But that would happen at much higher levels.
Lastly, this highly successful partnership shines light on Root’s amazing technology platform which provides the opportunity to repeat the same with other distributors. The more of the insurance distribution landscape they can capture, the bigger they will eventually be as a company.
The nature of this partnership
On August 11, 2021, both companies announced their partnership ”to develop industry-first integrated auto insurance solutions for Carvana customers”. They formally launched ‘Carvana Insurance Built with Root’ about a year later on July 27, 2022.
Since then, Root has underwritten $460m in premium through their embedded channel, much of which is likely Carvana. This embedded channel has grown in significance this year. It has accounted for almost all underwriting growth in the last three quarters.
For more details on Carvana and its relationship with Root, check out my original coverage from March here:
The total automotive market opportunity
Selling cars in the US is a huge business that goes far beyond just moving metal. Various stakeholders are involved, incl. manufacturers, dealers, financing providers and insurance carriers. As I will illustrate below, the insurance function trumps all other functions. Perhaps not from a revenue perspective, but certainly when it comes to the value creation opportunity from a shareholder perspective. Please note these are highly simplified high-level estimates to illustrate the significance of car insurance.
New car sales and financing
About 16m new vehicles are sold in the US every year. The average price of a new vehicle is about $48,000 which makes this an $800bn industry from a revenue perspective. The largest brands are Toyota and Ford with about 11-12% market share each.
Over time, their scale has enabled these OEMs to build large financing operations adjacent to their original hardware business. I touched on this when I covered the Tesla Finance opportunity in the article below.
As illustrated in the table below, leading manufacturers generate about a fifth of their operating income from their finance divisions.
The hardware and financing business trades at about 0.4x revenues combined. This implies the $800bn new vehicles market in the US likely provides for an aggregate market cap of about $320bn. 80% or $260bn of that likely pertains to selling the hardware.
You might want to object and respond that Tesla’s market cap alone is four times that estimate and half of their revenues are from the US. I’d counter that the vast majority of Tesla’s market cap pertains to the autonomy optionality and Elon’s brand. In terms of pure hardware sales, Tesla’s North American automotive hardware business is likely worth around $20bn to perhaps $40bn depending on how much valuation premium is assigned to the electrification opportunity.
Based on their operating income share, the US financing segments of OEMs are probably worth around $60bn. However, financing is obviously not limited to the OEMs. Commercial banks play an important role, too. They have about $500bn in automotive loans outstanding. Assuming they operate this at a 10% equity ratio and they are trading at about 1.5x book value, this is an additional $75bn in equity market value for the financing function. Therefore, I’d estimate that the automotive financing function is worth around $150bn in the US from a shareholder perspective.
These are obviously very high level approximations without much ambition about precision. For example, I am making the simplified assumption that OEMs do not have material net debt beyond the amount that is collateralized by their financing/leasing portfolio. For the purpose of this overview, this approach should however suffice.
Used car sales
Used retail sales in the US average to about 17m units annually. At an average selling price of $18,000, this is a $300bn industry from a revenue perspective.
Most of this happens through dealers outside of OEMs. Carvana accounts for about 400k units, 2% of the market. Were they an OEM, you could think of them like a RAM, GMC or Mazda, all of which sell similar volumes annually.
At a $50bn valuation, Carvana has presumably a lot of growth and industry consolidation baked in which makes them unsuitable as a yardstick for the valuation of the entire used car industry.
A more established and stabilized player is Carmax. They retail about 800k units per year. This implies a market share of 5% and grants them a $12bn market cap. This extrapolates to about $240bn in market value that is available to used car distributors.
New car sales
Assuming a) that a new car dealer will make roughly the same profit per vehicle as a used car dealer and b) that all new vehicles are sold through new car dealers, the aggregate value of all new car dealers may be similar to the $240bn estimate above because the number of new cars sold each year is approximately equal to the number of used cars sold each year.
In sum, vehicle manufacturing, financing and vehicle distribution has a combined market cap of $890bn based on the estimates above.
Insurance
The total US vehicle fleet is about 289m units strong. I am going out on a limb and assume that this is also the total number of policies in force (PIF).
Given that new and used vehicle sales are 33m units combined, a vehicle gets a new owner once every 8 to 9 years. Every time that happens, insurance carriers have an opportunity to underwrite a new policy.
The two largest publicly listed P&C insurers with a significant car insurance product are Progressive and Allstate. With 55m PIF, their combined market share is 19%. Given that car insurance accounts for 90% and 66% of their PIF, we can extrapolate the total market cap pertaining to US car insurance to about $880bn which equates to about $3,000 per policy.
Think of this $880bn as the hypothetical market cap of a monopolist car insurer in the US that operated like a weighted average of Progressive and Allstate. It’s the size of the pie which insurance carriers are fighting for.
As you can see, it’s a huge pie. It’s as big as the pies above combined. Believe it or not, I did not goal seek this. It’s 100% coincidence how closely this matches.
It begs the question why have OEMs not ventured into this space just like they have done with financing?
I attempted to answer that question in the article below where I discussed the Tesla insurance opportunity.
I believe it’s the result of two factors. Firstly, OEMs usually don’t own the customer relationship directly. This argument alone is however not satisfying because the lack of direct customer ownership didn’t prevent them from scaling large financing operations. The second factor is therefore likely more crucial: lack of expertise. Underwriting an insurance policy is much more complex than underwriting a loan. Pricing risk correctly is much more challenging. In that context, it’s admirable that Carvana and Root have managed to create a profitable and attractive insurance product.
Preferred stock and warrants
As part of the partnership, Carvana was issued preferred stock that is convertible into 0.8m shares at $162. And they have 7.2m warrants they can exercise at $180, $198 and $216. All in, they might control a third of the company’s shares if all of these instruments ended in the money.
This is a fantastic situation for Root shareholders. As long as Carvana’s preferred stock is not in the money, Root’s existing shareholders get the benefit of outside equity capital without corresponding dilution. Once this dilution happens, it does so at much higher valuation levels vs. today. And in case the warrants end up in the money as well next year, up to $1.4bn of equity pours into the company at a valuation more than two times the current valuation.
The value of Root’s technology platform
The Carvana model demonstrates that Root can develop and scale an insurance platform quickly and on very profitable terms.
The Carvana Root platform was launched in July 2022. Already in the 3Q22 earnings report, they determined that the short term tranches of Carvana’s warrants would vest rather than the long term tranches. Only one of the two categories can vest and while the latter has more originations as vesting condition, the former requires a quicker ramp-up.
Vehicle and associated insurance distribution is highly fragmented in the US. It relies on dealership networks and insurance brokerages. Root’s success with Carvana will open the door to repeat the same with its competitors.
Sincerely,
Your Fallacy Alarm
Do you know if there has been any info on how close Carvana is to meeting the Vesting Conditions? (# of policies originations)