A Project of The Annenberg Public Policy Center

Health Care and the “One Way Hash”


Here at FactCheck.org, we like to complicate things.

The statement isn’t meant to be (entirely) a flippant one. It really is true that a lot of what we do here is to take what appear to be pretty simple claims and show that the reality is far more complicated than it might appear at first glance. Quite often we find ourselves saying things like, "That’s true, but it’s misleading…"

Julian Sanchez, now a research fellow at the Cato Institute, has dubbed arguments of this sort "one way hash arguments." The term is a nod to one way hash functions, which are used in cryptography. The basic idea is that certain kinds of mathematical operations are really easy to perform in one direction, but much harder to do in reverse. A simple example: Finding a number’s square is relatively easy; given a pencil and paper, most people could multiply a number, say, 125 times itself. Determining the square root of 15,625 is another story entirely. The same idea applies to a one way hash, only with functions that are complex enough that even the fastest computers need unrealistic amounts of time to reverse the functions.

Sanchez points out that certain arguments have the same basic structure:

Sanchez: The talking point on one side is just complex enough that it’s both intelligible—even somewhat intuitive—to the layman and sounds as though it might qualify as some kind of insight. (If it seems too obvious, perhaps paradoxically, we’ll tend to assume everyone on the other side thought of it themselves and had some good reason to reject it.) The rebuttal, by contrast, may require explaining a whole series of preliminary concepts before it’s really possible to explain why the talking point is wrong. So the setup is “snappy, intuitively appealing argument without obvious problems” vs. “rebuttal I probably don’t have time to read, let alone analyze closely.”

Jess Henig’s Aug. 31 post "Cancer Rates and Unjustified Conclusions" here on the FactCheck Wire provides a couple of sterling examples of one way hash arguments. Critics of health care changes argue that since the U.S. five-year survival rate for many types of cancer is higher than the average European rate, despite the nationalized health care systems common in Europe, U.S. health care is clearly superior to European versions. Supporters counter that these same studies show that insured Americans have higher five-year survival rates than uninsured Americans, which simply goes to show that we ought to have universal health coverage.

Both arguments are a one way hash.

The reality is that an enormous number of different factors affect five-year cancer survival rates. Early detection is an important factor, for a number of reasons. For one, most cancers respond more readily to treatment when they are detected early. But even if two cancers received no treatment at all and developed in exactly the same way, the patient whose tumor was detected earlier would obviously have a better chance of living for five years after detection. The U.S. has a far more aggressive early detection program than the U.K. or most European nations. But other countries, like Australia, Canada, Japan and Cuba, all have five-year survival rates that are comparable to the U.S. and all possess nationalized health care systems. The one exception is prostate cancer, but even there the story is complicated. Since prostate cancer doesn’t always require treatment, aggressive detection means higher survival rates, since many cases that never require treatment will be detected. In other words, aggressive screening of prostate cancer dramatically increases the diagnosis rate by detecting cancer that doesn’t need treatment at all, something that then skews the survival rate.

The argument that health insurance coverage positively affects five-year survival rates is equally oversimplified. After all, the same studies that show different survival rates for the insured versus the uninsured also show that Americans covered by Medicaid have five-year survival rates comparable to uninsured Americans. What’s more, the uninsured often have other complicating factors; because the uninsured are generally poorer than the insured, they might also have different diets or different exposures to tobacco or other drugs, all of which could well affect cancer survival rates. The uninsured are also more likely to have more than one illnesses at a time, which affects survival rates. So simply comparing numbers across diverse populations without controlling for other factors isn’t particularly useful.

But notice that two arguments that took only a sentence each to explain required a dense paragraph each to rebut. And even then, the explanations are abbreviated (go read Jess’ post for the full details).

Keep this in mind the next time you see what looks like a knock-down, one-sentence argument for your favorite public policy option. If it looks like a pretty obvious (but not too obvious) argument, there’s a decent chance that you’ve just found yourself a one way hash.