, , ,

Currently Listening
By Jimmy Eat World

see related

Give me a lever long enough and a fulcrum on which to place it, and I shall move the world.” – Archimedes

If I were feeling cheesier, I could turn this into an allegory and go on ad ridiculously about how the fulcrum is public opinion and we must move that through persuasion towards the weightier truth or some such nonsense. Archimedes was a mathematician, but this quote points to the same truth, albeit different context, as part 1, 2, and 3 of On the Margin; In the Aggregate: given the right scenario, small changes can have extremely large effects. And it makes such a great opener.

To recap: In part 1 I examined how things change over time, and how there seems to be an inevitability to history—at least when viewed ex post. Part 2 looked at how Simpson’s paradox can apply to choices: although any given deviation from a path may make sense, the path still makes more sense because of the precedence that deviations set. Part 3, finally, looks not at the intent of a thing, but what that thing can justify in the future.

So now to part 4: if history really is the irrational result of a myriad of unavoidable, rational choices made by billions of people, how—if at all—can we change the world? More generally, what can aggregation tell us about public policy should do? In the previous posts I talked a lot about what policy shouldn’t do, so let’s switch it up and try to be positive for a change. How can we find Archimedes’ lever?

Let’s take Healthcare in the U.S. All national health systems want three things: high quality of care; high accessibility of care, including to the immobile  rural, and indigent; and a low, or at least sustainable cost. In the U.S., we have the highest quality of care in the world (note: quality of care should not be confused with quality of life or quality of health, which in neither of those categories do we rank particularly high among OECD countries). But the problem is that 10-15% of Americans don’t have insurance. Although he may be stupid for saying this, Mitt Romney is right that technically these uninsured folks do have access to care through EMTALA, but generally speaking, the uninsured’s access to treatment for chronic illness treatment certainly, and many types of acute treatment is hindered by their lack of insurance. Why do we need insurance here? (we certainly don’t need renter’s insurance to be able to afford an apartment, so it really is an odd premise.) Answer: because national expenditure on healthcare has risen by twice wage growth and all non-healthcare costs of living such that it is now the second largest section of the economy—second only to defense, which is an ironic juxtaposition. So healthcare has become so prohibitively expensive that we “need” insurance to be able to afford it. That’s the state we’re in, so I’ll unquote need, but instead of treating the problem directly, let’s treat it indirectly, by looking backwards.

Problem: Healthcare is too expensive for the median person to afford (though the average person is still doing all right).

We have two possible solutions:

  1. Make people richer (by giving them access to cheaper/free insurance)

  2. Make healthcare cheaper

I’m going to dismiss option 1 out of hand. I’m not saying there aren’t good arguments to be made on that side and that there are some reasonable proposals that would improve accessibilty of care by making people richer without harming everyone else. I can’t say I’ve heard any reasonable and sustainable proposals in the last year, but I believe they are out there. What I am saying is that empirical evidence shows that this is at best a temporary fix [1].

How do we make healthcare cheaper? Although I would love to go on a long pontification about healthcare and it’s intricacies, I’ll cut to the chase a bit and leave out several thought processes. I think they’re fascinating in their own right, but a detailed discussion of the idiosyncrasies of market for the health and healthcare doesn’t really add to this discussion, so it will have to wait. What does add to my point is the less detailed progression of going deeper into root causes.

Taking a more direct look at insurance, among other things it means that I can go to the doctor and not have to pay for it. Yes, I pay insurance premiums, but by the time I’m at the clinic, that’s a sunk cost. I also know that if I need surgery I won’t have to pay for it, or only pay very little relative to the billed amount. So quite frankly, I’m not all that concerned about the risk associated with indulging in some behaviors that increase risk of whatever problem. It’s also well documented that when we don’t pay for something we don’t appreciate it as much [2]. The result is this simple fact: people are less likely, on the margin, to take ownership of their own health, and rationally so. Medical technology has progressed to the point that we can revive people from myriads of illnesses and complications that were previously terminal. And these interventions won’t really cost me anything more. The so-called epidemic of obesity makes this case in point [3]. So if healthcare’s $0 marginal pricetag creates inducded demand [4], one answer could be to increase deductibles and coinsurance.

Take a situation, and continually ask “why?”; getting deeper and deeper; more and more basic. This won’t always lead to the same place, and it shouldn’t. Any asshat saying they have the silver bullet to fix all the issues we have with [insert complex/convoluted/arcane/intricate/embedded issue here] is dangerous. Increasing reliance on high-deductible health plans is just one of several conclusions I could have drawn from this, but it’s a fulcrum. If we trace forward using the inevitability principle from part 1, we can see a progression that might look something like this:

  • Increased deductibles lower expected expenditure by insurance companies, so premiums fall to maintain competitiveness. These lower premiums offset the increased out of pocket cost of healthcare.

  • Now that a doctor’s visit has an up-front cost, people value it more.

  • The increased value judgement and monetary cost of the visit corroborate to increase propensity to actually follow physician instructions. Particularly in the case of visits to primary care physicians, this means more effort put towards preventative medicine.

  • Prevention and healthier lifestyles decrease demands placed on hospitals; decreased demand facilitates a decrease in price, which decreases the risk to insurance companies who further lower premiums.

Furthermore, and probably more importantly, increased out-of-pocket expenditure increases propensity for patients to shop around for care, increasing competition between hospitals, and competition, through its “creative destruction”, generally leads to better outcomes on both price and quality. That said, it would take more insurance reform than just increased utilization of high-deductible plans for this to happen, but more on that in a future post. For a final point on the healthcare portion idea, this is the direction Singapore has gone with their health system, and the outcome has been extremely positive [5].

But this isn’t about healthcare, this is about aggregation. My goal is to develop a cohesive framework for looking at the world and not just criticizing proposals, but coming up with viable alternatives. I’m not saying this is the end-all-be-all of thinking, but as a framework for looking at large, interrelated systems, I like to think it’s a useful tool.

To review:
When we have an issue, US Politics tend to only look one, maybe two layers deep. But if we ask “why” enough, we can find some proximate causes. Proximate causes can be dangerous, because we tend to stop there and we forget that proximate causes often have their own proximate causes and so on ad infinitum. That said, knowing the ultimate cause—if such a thing really exists—isn’t always necessary. What we need to find is fulcrums. Trigger points. Inflection points. Whatever you call them, if we dig into the issue—thinking critically about what the data say, what theories built on those data say, and how that applies—we can usually find areas where a small change can have large, cascading effects. Rather than throwing lots of money at a problem and treating the symptoms, we should throw smaller amounts of money at these fulcrums, and if we’ve picked our fulcrum well, and have lever built out of data and knowledge, we can turn the world [6].


The Tattooed Economist sends his Love


[1] The Aggregate Effects of Health Insurance: Evidence from the Introduction of Medicare, Amy Finkelstein; http://www.nber.org/papers/w11619

[2] Russ Roberts; http://www.econtalk.org/archives/2010/04/ravitch_on_educ.html. Also, much psychological research has been done on this topic, but I can’t for the life of me find references to the studies I’m thinking of.

[3] Before I get angry comments saying I’m blaming overweight people for all the United States’ healthcare problems, let me stress two points: first, this is one small wedge, and secondly, in my opinion, this is not even one of the more important pieces, but it was the least verbose example I could think of.

[4] http://en.wikipedia.org/wiki/Induced_demand

[5] http://www.gapminder.org/ take a look at life expectancy over spending on healthcare as a percentage of GDP. Then look for Singapore in the upper left corner of the graph. Then spend the next 3 hours playing.

[6] Cheesy? Yes, but you have to admit that sentence ties the post together nicely.. 🙂