Oscar Health Flexible Business Model Is Paying Off (NYSE:OSCR)

Oscar Health (NYSE:OSCR) offers an innovative healthcare insurance platform that plays into industry trends like consumerization, digitization, and value-based care.

It uses AI and big data to enable members to navigate through the Byzantine world of US healthcare, matching plans to individual situations. It offers a host of services like, from Investopedia:

These plans offer access to free doctor-on-call services, free care team services, 24/7 access to a provider, and no referral requirement to see specialists, as long as they are in-network… Oscar’s members even receive cash incentives (up to $100 per year) for keeping up with their daily step count, as the app can link up to many smart pedometers and watches.

The platform makes use of AI and Care Teams to guide members to the most appropriate caregivers based on personal situations and is driven by sophisticated cost and quality algorithms.

The data generated further enhances the algorithms reinforcing the cycle of trust by increasing the quality of care and reducing cost, as well as member trust.

These are all nice words and intentions, but does it work? There are indications:

Until recently, this model was successful in rapidly increasing its membership but that growth has stalled as a result of ‘portfolio sculpting’ efforts (focusing more on those activities that generate high returns) which made the company even leave entire states (like Florida, albeit temporary, and California).

So rapidly gaining membership isn’t a sign of a customer-friendly business model anymore, but an industry-leading NPS (Net Promotor Score) score increased to 60 where competitors scored well below that, at least a year ago.

Still, they can grow even without adding new members. The growth rate might have come down quite a bit to ‘just’ 41.3%, but it’s still very robust:

However, as we will argue below, the main drivers of that growth (reducing ceding premiums to reinsurance and risk adjustments) as well as the efficiency improvements are running into diminishing returns.

That is, in order to keep on growing the company has to revive membership growth to keep the growth going. This is a little hard to predict given the shifting state-level economics which led the company to even suspend new members in states like Florida and California.

But management believes they can expand membership again and one has to admire the flexibility in the company’s business model that enabled strong momentum even in times of stagnating membership.

Growth drivers

• ACA growth
• Member growth
• Expanding in existing markets and/or entering new markets
• New programs
• Direct and assumed premiums

The company is present in 18 states, two fewer than a year ago so it isn’t expanding to new states at the moment. It presented its FY24 expansion plan in a PR in October 2023. It does plan to increase its footprint in existing states (PR):

Starting in 2024, Oscar will have a footprint in 18 states and 512 counties and expand its presence in Arizona, Florida, Georgia, Illinois, Iowa, Kansas, Missouri, North Carolina, Ohio, Oklahoma, Tennessee, Texas, and Virginia. The expansion will improve access to care in underserved and rural markets through Oscar’s consumer-focused technology platform.

There are other new initiatives like

• Its Spanish-speaking program HolaOscar, representing a growing part of the ACA.
• They will also introduce new individual and family plans in 2024 like the recently introduced Breathe Easy program for members suffering from COPD and asthma and an enhanced Diabetes plan.

Premium growth is driven by retention, new members, and premium increases, and these produced a 5% decline in Q3 to $1.6B. The decline was caused by a reduction in membership that was only partially offset by higher premiums (earnings PR):

Premiums earned in the quarter were up 46% YoY, driven primarily by the impact of deposit accounting for quota share reinsurance agreements, and lower risk transfer per member as a percent of premiums.

Quota share is an arrangement between an insurer and a reinsurer. The accounting can be a little complex but it doesn’t alter the basic fundamentals.

However, the company traditionally has a healthier and younger membership (as these are more internet savvy) than the market average which obliged them to make substantial payments into the risk adjustment program.

However, this risk adjustment is declining as their members move towards the average ACA population profile (Q3CC):

our premiums before ceded reinsurance, which includes the impact of our lower risk adjustment transfer grew 6% year-over-year to $1.4 billion.

Oscar also recognized a $27M in risk adjustment benefit related to 2023 as their risk moves towards the average ACA and they expect lower risk transfer going forward.

However, almost all revenue growth simply came from keeping more policies on the books, rather than ceding them to reinsurance companies:

Almost all revenue growth comes from ceding almost no policies to reinsurance, just $2.9M worth, while this was $364M in Q3/22 with the nine-month decline even more dramatic. In fact, rather than ceding policies to reinsurance companies, the company shifted to net-assuming policies from reinsurance companies:

+Oscar

+Oscar is the business line where they monetize their technology platform to healthcare providers and payers, or at least some modules, most notably the Campaign Builder (10-K):

Campaign Builder is an engagement and automation platform that enables scalable, personalized interventions and automates workflows, designed to drive growth, deliver best-in-class clinical outcomes, and optimize administrative operations.

It already has two customers (one is Cigna) serving 500K people and added a third in Q3 in the form of Stanford Health Plan. The company is also building more OpenAI solutions into +Oscar (as well as their insurance business).

Financials

Two relevant ratios are capturing the efficiency of the company

• MLR or Medical Loss Ratio (the share of total health care premiums spent on medical claims and efforts to improve the quality of care). The MLR has to stay above 80% for the ACA.
• The AER or Administrative Expense Ratio.

Both ratios are improving, the MLR improved 610bp to 83.3% y/y (Q3CC):

due to our disciplined pricing actions and total cost of care initiatives

The AER improved 330bp y/y to 17.4% (Q3CC):

driven by lower risk transfer per member as a percent of premiums and distribution optimization.

That’s a combined improvement of 935bp in Q3 (and 560bp for YTD). There is a further boost from the interest ($44M in Q3) they make on their $2.6B cash reserves, which they kept mostly in short-term instruments to take advantage of the rising rate environment.

The reserves were quite a bit higher a quarter ago as they paid their annual risk adjustment transfer in Q3.

No wonder AEBITDA keeps on improving in a pretty comprehensive fashion:

• The company was able to increase AEBITDA margin by an average of 800bp per year over the last three years, there is considerable operating leverage even in the absence of member growth.
• Q3 AEBITDA improved by $140M to a loss of just $20M for the insurance company, and for the overall company, there was a $340M improvement to an AEBITDA profit of $66M.

However, consider the following:

The MLR for YTD is already at the minimum required 80%, that is, further improvements aren’t possible (and unlikely anyway as a big part of the improvement is provided by fewer risk adjustment payments, with Oscar’s population profile approaching the average for the ACA, further gains are more difficult).

Guidance

From the earnings PR:

• Direct and Assumed Policy premiums at the high end of the $6.4 billion – $6.6 billion range
• Medical Loss Ratio at the low end of the 82% – 84% range
• InsuranceCo Administrative Expense Ratio near the midpoint of the 17% – 18% range
• InsuranceCo Adjusted EBITDA⁽¹⁾ of $155 million – $165 million, above the high-end of the prior range of $20 million – $120 million
• Adjusted Administrative Expense Ratio near the midpoint of the 20.5% – 21.5% range
• Adjusted EBITDA⁽¹⁾ loss of ($60) million – ($50) million, above the high-end of the prior range of ($175) million – ($75) million

Valuation

From the 10-Q:

Concerning the latter one should realize (10-Q):

Our Class B common stock has 20 votes per share, and our Class A common stock has one vote per share. As of September 30, 2023, the holders of our outstanding Class B common stock, which consist of Thrive Capital and our Co-Founders, beneficially own 21.3% of our outstanding capital stock and hold 81.8% of the voting power of our outstanding capital stock (assuming the exercise of all options to acquire shares of Class B common stock and the conversion of the 2031 Notes, in each case that are beneficially owned as of September 30, 2023).

There is more dilution on the way, from the 10-Q:

So fully diluted there are 324.3M shares outstanding, producing a market cap of $3.89B (at $11 per share) and an EV of just $1B. We hesitate to base valuation on the EV as the $2.6M cash and investment balance is largely there to underpin claims.

Based on expected FY23 revenue of $5.9B the shares seem cheap to us with a P/S of 0.66x even if the company still produces losses, and insiders seem to agree as they bought 5M shares from a selling holder.

One thing to keep in mind is that there are many moving parts having an impact on company growth and financials and these can move in unexpected ways.

Conclusion

The results that Oscar Health continues to generate are impressive; however, we think that three of the drivers of revenue growth and operating leverage this year are facing diminishing returns with one of these (policies ceded) already at zero:

• Almost all revenue growth comes from a dramatic decline in the ceding to reinsurance companies, since these declined to almost zero this can’t boost growth going forward.
• While the improvement in AEBITDA continues to impress, we’re unsure about the mileage left, especially in improving the MLR, given the fact that by law it cannot fall under 80%. Economies of scale and the increasing use of AI might very well continue to produce operating leverage, but perhaps not to the extent investors have become accustomed to.
• Risk adjustment is another factor where declining marginal returns have set in.

So basically, to continue to grow the company needs to expand its members, which has been stagnating for a year. Management expects member growth in the high teens and premium growth in the low 20s.

If they can deliver on that (and recent insider buys suggest management has considerable confidence in these forecasts) this will still produce attractive growth (albeit not the 50% growth the company has been producing this year).

But they gave very few details about how to achieve that, apart from their expansion to rural areas in states where they already have a presence. We assume that shouldn’t be too hard and given the attractive value proposition to customers and the all-digital and consumer-centric advantages of their business model in a rural setting we assume they can achieve this.

There is also likely to be further operating leverage left, mostly from platform scale economics and the use of AI, which they’re adopting with considerable gusto, but diminishing returns are likely to have set in concerning the AER and especially the MLR improvements.

So we think there are reasons for optimism and given that the shares aren’t expensive we think they can still be bought under $11 as long as investors keep in mind that things can change quickly for the company although they have an impressive track record of producing improvements irrespective of circumstances.