+1000 for Signaling

Aaron Carol, at The Incidental Economist, has a post showing that disease prevalence (including obesity) in the United States is a very, very small portion of what is driving health care costs:

Before you start in on me about how obesity is linked to other things and such, you should know that the overall McKinsey & Company analysis showed that the prevalences of disease in the US could account for perhaps an extra $25 billion in health care spending. Let me make a new chart for you:

Yes, obesity is more prevalent in the US, and yes, caring for it costs real money. But even if we get obesity down to the levels in other countries, it’s not going to magically erase the problem. We are spending two to three times per person what they are. There is no simple fix here. There is no one, and no thing, we can easily blame.

Everyone, always will look for a scapegoat. It is in our genes. The good vs. evil story is the oldest trope in existence. Look at the current outcry against “evil speculators” in oil markets (I wonder why Krugman doesn’t make a post about that?). Humans live through stories, humans respond strongly to in-group loyalty, humans have value preferences that lead them to view the world radically differently. I’ve often stated in debate that those who think that a single-payer health care system would somehow reduce our expenditures to a level consistent with other OECD countries are dreaming, at best, or delusional, at worst. And every single data point that passes by in the health care debate does nothing but strengthen the position that Robin Hanson articulated: health care altruism is a permutation of our evolutionary drive to “show we care”; or rather, make infrequent, and very large expenditures to show our loyalty to an alliance. The frequency has gotten greater as our society has gotten richer, but the underlying motive is still linked to our evolutionary roots.

Against this strain of thinking is the hypothesis that Matthew Yglesias articulated in his Bloggingheads diavlog with Karl, that people are stingy in the voting booth, but acquiescent in the doctor’s office. So separating payment and service would act as a brake on health care expenditure. I’m very skeptical of this argument. After all, health care expenditures have risen at a higher rate than GDP/per capita in many countries around the world.

A more interesting question, though, is why is the US different? My crude outline of a hypothesis is that people in the US have only recently come to “share a heritage” that is the United States. It’s only been around 100-ish years that people have really come to view themselves as “Americans”. In the absence of a shared heritage (which provides a built-in in-group), it has been especially important to engage in acts that show inter-tribal loyalty. The US spent a greater amount of money/life/time ending slavery, securing women’s right to vote, and ending segregation than a lot of other countries. We’ve also spent more money/ink/time securing a the minimal welfare state that we have, that is exceedingly expensive (relatively speaking). Not surprisingly, we also spend a ton of money/ink/time on health care that is of extremely dubious effectiveness. A cynic might say that this represents the greater wealth of the United States…but that doesn’t really provide an satisfactory explanation. We have low taxes, so we get away with a lot of inefficiency, but I don’t think that is the underlying driver of our proclivity to expend a lot of resources doing different things.

I think that history will show that Robin Hanson is right, and that whatever health care arrangement we devise, it will continue to be significantly more expensive than the world norm. That it has relatively little to do with the structure of the market (though I stand firmly behind a completely free market in primary care/pharmaceuticals [except antibiotics/microbials]), and a lot to do with our evolutionary drive as a “multi-tribal” society.

The Elasticity of the “Love Hypothesis”

Karl has a post earlier today where he makes the case that the “love hypothesis” broadly explains trends we see in k-12 education. Specifically, that we school children in ways that show we care, rather than ways that maximally benefit children. However, he then brings up that this wouldn’t explain rising student debt:

What the love hypothesis doesn’t explain is rising student debt. Why are the students themselves taking on ever larger burdens. Is it so they can prove that they love themselves? That’s not totally implausible, but out the gate it doesn’t seem very compelling.

Fortunately, we don’t have to shoehorn the love hypothesis to fit. This is a kind of a form of the principal-agent problem…although not so much a “problem” per se. When children are the agents, and parents are the principals, then parents spend money in the ways that they see fit, which explains how the love hypothesis would provide a transmission mechanism from what parents spend into the type and amount of schooling that children receive, even if children (agents) aren’t really getting much out of it at the margin.

However, student loans are an example of the principal and the agent being the same person. Students are largely mortgaging their own futures in order to increase their marginal productivity. Thus they don’t need to love themselves, that explains why people spend money on other people’s education (indeed, it explains the skyrocketing tuition at ivy league schools, where parents do pay the bills many times).

I like to explain rising student debt (and thus, greater consumption of higher education) using education as a network good. Network goods are characterized by two concepts that would illuminate this: knock-on and tipping points. Put simply, if no one had a bachelor’s degree, no one would need a bachelor’s degree. On the other side of the coin, if everyone has a bachelor’s degree, then you are locked out unless you get one. The more people that have bachelor’s degrees, the more useful they are to those who possess them, until the network reaches a tipping point where employers begin preferring bachelor’s degrees, on to a point where employers require a bachelor’s degree. That pushes people into the market for master’s degrees, rinse and repeat. This could likely go on forever in an with infinitely-lived agents, and infinite degree successions.

Or maybe I’m just too dead tired to reason well today.

The New Rich

Robin Hanson critiques Chrystia Freeland’s take on our billionaire overlords.

It doesn’t seem to matter to Freeland how deservingly the rich obtain or spend their wealth; they still must be taxed to help average Americans, even if that slows the lifting of Chinese and Indians out of poverty. It isn’t clear why she recommends the rich eagerly submit to such taxation; she suggests taxation will happen whether they like it or not. Why fear “populism” beyond its taxation? The point seems more to scold the rich, in order to reassure the rest of us that we are justified in taxing them.

It may be the case that many calling for higher taxes on the rich are belittling or even demonizing their contributions as a way to justify higher taxes.

I am starting to believe that the belittling and demonization are at least as important to many of the new rich as the taxation itself.

That sets up the possibility for a trade. We praise the rich as we redistribute their money. I am not completely joking when I suggest there might be room to turn taxation from a welfare destroying necessity to a welfare enhancing activity for everyone involved.

Suppose we start a government website which shows the taxes paid by the wealthiest Americans but instead of listing them in terms of dollars we list them in terms of “kids treated under CHIP” or “soldiers sponsored” or some other meaningful measure of government spending.

Right now paying taxes is a sign of low status. You allowed the government to get your money. However, if we advertise broadly what the money pays for then we can push tax paying as a high status activity.

This is crucial because most of the spending among the wealthy whether it is for a giant yachts or malaria vaccines is ultimately a status competition. Its either who has the biggest toys or who has done the most to save the world.

This is not a dig at the rich. Human are programmed to focus almost exclusively on status competition whenever material survival becomes a non-issue. This is why both high schools and senior homes are full of cliques.

What we want is not destroy status competition, that would be impossible even if it was desirable. What we want is to channel it.

Some status competition is obviously destructive. Tax evasion as a status competition – see Leona Helmsley– is destructive. However, if tax paying is promoted as a status competition it becomes productive. It serves the same function for the people in the competition but it serves an improved function for society.

Now, a key question Robin would be sure to ask is, why should we think paying taxes is productive. What about giving money to the poor in India. Isn’t that better than paying taxes.

There is a longer moral case to be hashed out. However, my short answer is that maintaining a pro-market polity is productive and this is becomes less likely when the average person sees the market as primarily benefiting others.

Should you donate to think tanks?

Ezra Klein recently advocated donating to non-profits that can affect policy rather than charities:

If you donate money to a food bank, it can provide only as much food as your money can buy. If you donate it to a nonprofit that specializes in food policy issues, it can persuade legislators to pass a new program – or reform an existing one – that can do much more than any single food bank.

Robin Hanson disagrees with Ezra, arguing that these types of think tanks and organizations will mostly spend effort on partisan issues which he argues are likely wasteful:

Assuming all parties think they seek good, partisan changes can only be good if some parties are right while others are wrong about what is good. In contrast, you can be right about a non-partisan change without others being wrong…..

Donations to change policy within the partisan subspace, however, only achieve good when they happen to be on the right side of partisan disagreements. Averaged over the disagreeing parties, such donations cannot on average achieve good unless there is a correlation between between donations, or donation effectiveness, and which sides are right.  Even if you think you are right at the moment on your particular partisan policy opinions, you can’t think it good on average to encourage partisan donations, unless you think donations tend overall to go to the good or more donation-effective sides.

There are a couple of ways to look at this with respect to Ezra, his donations, and his advocating for such donations. On average, Ezra surely believes he and the organizations he donates to are on the right side of partisan disagreements. Also, being that Ezra’s reading audience probably on average agrees with him, he also probably believes they are on average on the right side. So I think Ezra could agree with Robin that if you’re writing for a random or sufficiently broad audience, advocating for donations to policy organizations would do no good, while also believing that he should make such donations and advocate for his audience to do so.

On the other hand, I’m not convinced that equal donations to, say, the Center on Budget and Policy Priorities and the Cato institute are zero sum.  I see part of the function of these organizations as pushing politicians towards empirically grounded and efficient policies and away from populist, extremist, and obviously inefficient ones. I see raising the influence of these institutions relative to populism as raising the probability for any given outcome the government tries to achieve, whether it’s more liberal or more conservative, the policy used is more efficient.

For instance, I’d venture that obviously inefficient and suboptimal policies like the minimum wage, tariffs, and the gold standard are much less popular among the median think tank, or think tank dollar, than they are among the median voter. Think tank solutions on average tend to utilize more transfers and market oriented allocation and less government mandated quotas and prices than what popular support calls for. In addition, I’d venture that futarchy and policies that seek to cut medical spending –two of Robin’s favorites– would be, on average, much more popular within the walls of a random think tank than within the walls of a random U.S. home.

In short, these places think like economists. Even if they cancel each other out along partisan lines, think tank donations push policies away from the biases of the rationally irrational voter.

One test of this idea is to ask whether people on the right would agree that we’d have better policies if the CBPP was more influential on the left. Likewise, would people on the left agree that we’d have better policies if Cato were more influential on the right? Perhaps it reflects my personal policy biases, but I happen to think the answer is yes.