Saturday, March 20, 2010

The Parmenides' Fallacy in Health Care

A couple days ago, my beautiful, and extremely brilliant girlfriend was talking to me about a somewhat rarely considered logical fallacy called the Parmenides' Fallacy...

As I discovered, there is actually a blog (sort of) devoted to this particular fallacy, and the authors of that blog reference Prof. Philip Bobbitt, who coined the term.  Bobbitt's explanation is as follows:
"The Parmenides' Fallacy occurs when one tries to assess a future state of affairs by measuring it against the present, as opposed to comparing it to other possible futures.”
Elegantly said, I think.

The fallacy is probably one of the most widely employed by people in all walks of life, and one of the least discussed or known (perhaps because it was so recently coined).  It is also a bit hard to wrap one's head around, so let me provide an example of how this logical problem plays out in the real world.

...And for that there is no better example than the government's own financial prediction organization, the Congressional Budget Office.

Greek Philosopher, Parmenides
The Congressional Budget Office unfortunately (and somewhat by design) winds up employing this fallacy constantly in its cost estimates of new programs.  Take, for instance their recent cost estimates of the latest iteration of the "Health Care Reform" bill up for a vote this week.  The CBO's estimates were somewhat positive for the supporting politicians because they tentatively said that - if there were additional scheduled cuts that actually happen the way the sponsors claim - the bill will reduce deficits by hundreds of billions of dollars over ten years.  Great news right?

Well... No, actually.

One of the problems with the CBO is that they are extrapolating the future demand on the nation's health care resources based on today's demand.  But as Dan Mitchell of the Cato Institute points out, this is a very bad plan indeed:
"The CBO has a very dismal track record of getting the numbers wrong, in part because there is no attempt to measure how a bigger burden of government has negative macroeconomic effects, but also because the number crunchers do a poor job of measuring the degree to which people (recipients, health care providers, state and local politicians, etc.) will modify their behavior to become eligible for other people’s money. The problem is compounded by similar mistakes for revenue estimates from the Joint Committee on Taxation, which (like CBO) makes no attempt to capture macroeconomic effects and has a less-than-stellar history of predicting behavioral responses."
And thus, we have the Parmenides' Fallacy working right before our eyes.

The Congressional Budget Office is looking at the current state of affairs (the amount of health care resources used per person when individuals still - even to a comparatively small degree from what would exist in an actual market-based health care system - are responsible for paying for their own treatment), and assuming future conditions based on the present.

The trouble is, when health care is believed to be "free", the behavior of the average health care consumer will drastically change.

It's a fundamental principle of economics, and easily observable reality in human nature, that when something is free people consume more of it.  Now, there are often arguments made that "Well, no one plans to get sick, so the idea that they're going to just break their own legs or injure themselves so they can go to the doctor more is absurd".

True enough.  But those arguments miss the point.

What will go up, as has done already with increasing government involvement in health care over the last 50 years, is the frequency with which people go to the doctor even for unnecessary problems, the level of drugs (prescription vs. generic, for instance) they purchase, the number of tests run per visit and other activities that propel costs higher & higher each year.  When the "someone else" is paying for their health care, people - both doctors & patients - on average are much less discriminating on what they actually need to spend their money on and rather go for basically everything possible.  Someone else is footing the bill, so why not?

The problem is that this mentality puts an enormous strain on the finite medical resources of the nation, and as a result, prices rise and access becomes more and more limited.  Of course, by setting terms carefully, insurance companies can limit this behavior to a degree currently, though not that much unfortunately, as there are a lot of restrictions already on those companies divorcing price from the end consumer.  Plus, about 50% of all health care is paid by the government already, further compounding the problem...  But the new health care legislation will change that situation entirely, warping the incentive structure even further towards a disconnect between price of goods & available supply - and the CBO doesn't really take that into account.

The CBO is making a crucial error in the way they calculate costs, and to a large extent, we can't even blame them for it! The system itself leads to this very problem... Politicians don't offer a dozen different bills for the CBO to compare & contrast, they offer them one bill at a time, and deliberately ask them to compare the future with that legislation to the present... The system promotes the fallacy.

And that's not the only problem, by the way!

Politicians are doing everything they can to game the system and use accounting tricks to skew the numbers.  It's no surprise that people who have vested interests in getting bills passed will manipulate the numbers a much as possible to make the bills more tenable.  This is especially true when we're talking about legislation that adds $1 Trillion in new government spending (from what source this magical trillion dollars comes from I still haven't the slightest clue!) in the middle of a severe global economic recession.

The Congressional Budget Office has no choice but to calculate future costs based on the data presented to them by politicians writing new legislation.  And like the rest of us, the folks who make up the CBO aren't psychic and can't predict the future.  When a politician says that in the future, other programs will be cut & salaries reduced, etc. to pay for this massive new entitlement expenditure, the CBO has to take that politician at his word and include that in their analysis...  Even when the entirety of human history shows that governments the world over never makes good on such promises.

As Michael Cannon (also at Cato) discussed recently as well, there are a ton of completely obvious problems with the political math already anyway simply due to politicians using their knowledge of how the CBO process works to game the system itself.  For example:
"As former Congressional Budget Office director Donald Marron has explained over and over, the figure that Democrats consistently cite for the cost of their bills is only the CBO’s estimate of the cost of federal spending related to the expansion of health insurance coverage.  It is not the full cost to the federal government, because each bill also spends taxpayer dollars on other items.
Moreover, the on-budget costs of the legislation probably account for only 40 percent of the total costs.  The other 60 percent come from the private-sector mandates.  But Democrats have systematically suppressed any estimates of those hidden taxes, probably because such an estimate would reveal the full cost of the legislation to be closer to $2.5 trillion over the next 10 years."
So in truth, while it seems like I'm railing against the Congressional Budget Office, I'm really not...

I think they're doing the best they can, and probably the best job any country could hope for in the field of analyzing the future costs of government... But their job is, at root, completely impossible.  Even at their absolute best, their numbers will still always be wrong... And I think this all goes into why there can be no rational economic calculation with government.

It's not the CBO's fault, but with politicians busy playing fast & loose with the numbers and using tricks to hide the truth of how much the American citizens are going to be on the hook for in the coming years, the very nature of how our cost-estimating agency operates pushes it into accepting a pretty serious logical fallacy as the basis for it's predictions.

The thing the CBO really needs to be doing primarily is reviewing multiple courses of action and assessing their relative merits in the future against other possible futures.

What will health care cost if we do nothing based on current estimates?  What will happen to that number if so-called "Health Care Reform" is passed?  What will happen to that number if real reform happened instead?  It's useless to view the future and say it's better or worse than today - of course it will be, one way or another.  What isn't useless is trying to determine which of many possible futures are going to bring about the best results and then using that knowledge to pursue intelligent courses of action.

Of course, this actually requires people at the CBO and within the government in general think more like decent economists and weigh options - but this isn't really a possibility.  They are charged only with reviewing legislation as it comes to them, not with producing alternatives... And the political process would never give them the time or latitude they'd need to accept 100 different plans (which aren't designed simply to game their review process) and then analyze each plan and legitimately compare one future to other possible future alternatives.

The asinine thing about the current health care legislation is that even if it were true that this particular iteration would reduce budget deficits over 10 years (and believe me, it isn't), it's *not* true in any case that the bill as it is is the best possible way of accomplishing that.  Comparing a future where we reduce some deficits to the present or the results of the present course we're on may show improvement, but assessing the future based on what is currently happening is a fallacy precisely because it precludes the possibility that there are other possible futures which would be even better.

This is a common theme with government, and a huge point of contention with free market economists and myself.  Politicians claim, for instance, that bailouts have saved millions of jobs and that we should be thankful because we would have lost more than we have if we had "done nothing".  However, I would contend that had we followed the prescription set out by folks like Ron Paul, and let the artificial bubble collapse quickly and help adjustments in the market with streamlined bankruptcy proceedings, and clear processes and even-handed application of the law to prevent theft & fraud... Not only would we have even more jobs today than we do, the correction would be over by now and the country wouldn't be in the grips of a depression, and businesses wouldn't be constantly leaving the US for more easily navigable climates - such as Singapore.

But that possible future was ignored two years ago, and we are now stuck hearing the President tell us how bad things would be today if they had "done nothing" - which is a conveniently unfalsifiable statement, since we can't just travel to another universe where the government actually embraced liberty in a time of crisis - but he doesn't bother to point out how much better things would be had they done the right things...

(I feel compelled to note here that the same Parmenides' Fallacy applies to proposed climate change legislation too, by the way... But you can ponder that one on your own.)

But back to the topic at hand... One way or another, the cost savings are a lie, but politicians can point to the present state of affairs and say that in 10 years with health care, they're going to "save" $100 Billion (after spending $2.5 Trillion, of course).  Sounds great, until you realize that if they actually took a market based approach they would not only not have to spend $2.5 Trillion in the first place, but would actually wind up reducing the cost of health care around the United States significantly and save many hundreds of billions of dollars, or even trillions, for real.

However, because they employ the Parmenides Fallacy - we don't actually get presented with that option.

The way it is all presented to the public is that there is essentially only one alternative, and our choice is stay with things exactly as they are or agree to a plan that politicians would have us believe will be marginally better than the present condition.  And we're supposed to cheer our great leaders for their bravery.

It's kind of an insidious fallacy in a way simply because it's very subtle and requires a depth of thought that a lot of people don't really have to get around.  It's just not enough to be able to think "If I do X, will that make me better off than I am today in 10 years?". We must also, and more importantly think, "If I do X, Y, or Z - which of these will make me best off compared to each other in the long run?"

The CBO, and more importantly our Congress, never do this... And even if that task were remotely easy to do, the system is designed so that they can't.

No comments: