Coverage, Copay, and Chaos

There was a recent Noahpinion article about insurance in the wake of the recent murder of the United Healthcare CEO, after it spurred a wave of netizens proclaiming the evils of insurers. Some others agree. Smith is not one of them: he blames the providers. Smith argues that if insurers are indeed the big bad guys, then they are terrible at it: iInsurer profit margins and ROEs loiter embarrassingly below the averages of the S&P 500s.

I’m a fan of Noahpinions (and Noah himself, though I know him very peripherally). I agree with some points, but as a frustratingly boring Moralist I think there’s more to the story here.

I’m in the business of looking at risk-adjusted rewards. High risk, high rewards, blah blah. The reverse is true. Insurers have a beta of 0.7-1 (and 0.6-7 if you go into reinsurance), compared to 0.88 for healthcare, or 1.1-1.6 for software. So while insurers may not be the most profitable businesses in the world, they represent a reasonably safe, stable $1.4T or so a year in gross written premiums.

Smith also argues that Americans are also not that terribly situated: he presents a graph showing that US out of pocket expenditure as a % of healthcare expenditure is lower than that of other developed countries. Sure, but we forget that US healthcare expenditure as a % of GDP is higher than said countries. The denominator matters here. So in terms of % of GDP, USA still ranks high at ~2% out of pocket costs, versus ~1.5-1.6% in peer nations.

(2019 numbers, because COVID has made things confusing; from Our World in Data)

And one more thing Smith doesn’t address: if providers are the problem, why are they the problem? It’s counterintuitive that an ecosystem as capitalist as the US’ should see price increases without corresponding changes in consumer surplus if there was competition. I Googled/ChatGPTed/Claude’d and well, capitalism does automatically not a laissez faire market make. Institutions matter (Smith has some beef with the purist institutions hypothesis and makes a convincing argument that human capital is inextricably intertwined with its effects), because it seems that most of the modern American healthcare system unfolded form path dependence of policies created around World War II.

Consider this: US healthcare expenditure as a % of GNP (I know that’s not GDP, but the delta was relatively small) hovered between 3-4% in the 1930s to the 1950s. Many other developed countries circled that gravity well; the UK also spent about 3.5% in 1950. Yes, they have both increased, but the US at a much higher rate.

During WWII, wage controls limited employers’ abilities to offer higher salaries. As a result, health insurance was the benefit to differentiate on; and coupled with the 1943 IRS ruling that exempted employer-provided health benefits from taxable income, effectively gave a 30% discount on health insurance purchased through employers vs bought individually.

This created an effective principle-agent and moral hazard issue. Employers give more coverage than employees would purchase if they had to pay with their own after-tax income, and employees maximize utilisation of healthcare plans because they aren’t the ones paying. The ones who disproportionately lose out are the ~40% of Americans who don’t have employer-provided insurance, which I presume disproportionately represents the disenfranchised and lower income groups.

There’s also the issue of fee-for-service, where healthcare providers get paid separately for each individual service, test, procedure, visit, etc. independent of outcome. This is not unusual in and of itself, but most other countries have some way to regulate prices such as single-payor systems bulk buying (and therefore negotiating) at the national or regional level using monopsony power to bring costs down. This also results in few systems; administrative costs for insurers are ~2x in the US versus other comparable countries (a point Smith concedes but argues is not the primary cost driver, which is fair).

And then there’s the lobbying. In the 1930s-1940s, the American Medical Association spent ~$1.5M (or ~$25M today) fighting Truman’s national health insurance proposal. A third player whom we haven’t discussed yet has joined the game: big pharma. PhRMA prohibited Medicare from negotiating drug prices (this has recently been undone with the IRA, but it won’t start til 2026 and is piecemeal) - effectively creating the bizarre monster of a large buyer with no market power. They also basically enforced protectionism by banning drug imports. Prescription drug prices in the US are almost 3x of other OECD countries.

In the social experiment I hail from, healthcare expenditure % is markedly low for a rich country, but our fewer number of humans makes us far more manageable. We have an interesting mix of government intervention and free market competition:

• Public infrastructure doesn’t suck: Singapore government-owned providers cover the majority of healthcare in Singapore.
• Price transparency: The Ministry of Health maintains fee benchmarks so that patients are better informed of costs and further create competitive pressures to push down costs. While as far as I can tell, there is a ton of asymmetric information in the American healthcare system and providers can set their own rates in their own information silos.
• Price discrimination: tiers upon tiers upon tiers. There are wards with different classes of comfort and corresponding amounts of subsidies in each hospital; the government-subsidised health insurance is based on income;
• Talent Management: Most of the doctors in Singapore are government scholars. The government pays for their education, and the scholars pay off their bonds by working in public institutions for 4-5 years. This creates standardized training and controls costs by ensuring a stable supply of well-trained doctors to public providers, reducing wage inflation and ensuring competitiveness with private providers.
• But also, note the out of pocket %! Significantly higher than other countries’. Singapore was built on the narrative of self-responsibility; things are subsidized but rarely given with no strings attached. Something something meritocracy, or the imitation of it something something.

Sounds great!! Except … 5 years ago, in 2014, healthcare expenditure was 3.9% of GDP. My father was telling me about a conversation he had with a CFO of a large healthcare institution and his concern about rising costs. The wrench in the gears is that Singapore has become the hot spot for medical tourism for the wealthy in surrounding ASEAN. Singapore had historically constrained doctor supply (by limiting enrolment) to ensure there was enough talent for other sectors of the economy (yeah, they plan that far in advance). Things go well when they go according to plan. But when your plan misses something - in this case, infinite demand with high willingness to pay - your careful market set up can go awry. Specialists in Singapore now earn in the millions and millions, and to compete, public institutions have to bid up their offers as well. So what was a ~$300K job is now ~$700K or more. That increase drives up insurance costs, hammers deeper the inequality wedge and focuses more resources onto serving the ultra wealthy, who then keep coming in droves for the quality service they get, and so on.

I don’t begrudge the doctors. My wonderful, wonderful sister is one, and it’s still insane that she has to do 30+ hour shifts and still be expected to give sound medical help. But the market giveth, the market taketh away. And when you play with nature, you have to be prepared to handle the resulting mosaic chimera.

OK, madame venture capitalist, so what? So tech startups that want to take over the world can’t do it without understanding the intricate lattices that underlay the systems they want to disrupt. Insurance began as informal mutual aid in church communities - the concept persists regardless of its specific form and we can’t Don Quixote or SaaS or LLM it away. That we still exist within the constraints of human greed, of broken information loops, and of legacy; and that is the muck we have to wade through to move forward.