Tuesday, October 20, 2009

Positive Versus Negative Impingements on Liberty

By Mike Dorf

In my latest FindLaw column I consider the libertarian objection to the proposed individual mandate in most of the health care reform proposals now before Congress.  Some libertarians say that it is both unprecedented and categorically worse for the government to require people to do something affirmative (such as buying health insurance) than it is for the government either to forbid some action or to require some other action as a condition of engaging in an activity (such as practicing medicine) that the govt could prohibit.  I end up concluding that the objection is not sound.

Here I nonetheless want to try to unpack the intuition that there is something worse about affirmative impositions on liberty than prohibitions.  Let's begin with a pair of examples that show how an affirmative imposition feels worse.  Suppose the govt says that I must be at jury duty from 9 to 5 for a week.  That is clearly a greater restriction than a negative imposition that forbids me from, say, going to the movies during that same time.  In the first case, I can't do anything other than go to jury duty; I can't even go to the movies.  By contrast, in the second case,  I can do anything I want--other than go to the movies.  So the affirmative imposition seems MUCH more of an imposition than the negative one.  And we might think this is typical: Prohibitions take one option off the table (the prohibited conduct) but leave us free to do anything else, whereas affirmative obligations tell us exactly what to do.  They leave no freedom of movement.

But the foregoing juxtaposition does not appear to be an inevitable feature of affirmative versus negative impositions. Consider an example suggested to me by Neil Buchanan: Suppose that a parolee is under a form of house arrest.  The parole condition could state that the parolee must be in his house at all times--an affirmative obligation to be somewhere--or it could state that the parolee is forbidden from going anywhere outside his house at any time--a negative imposition.  Yet obviously these conditions are identical.  Indeed, we could frame a much more restrictive negative prohibition--e.g., don't leave your house--than a positive one--e.g., you must be in North America.  The key is the scope of the restriction, whether positive or negative, not whether it is positive or negative.

Can every affirmative obligation be turned into an equivalent negative one and vice versa?  I don't think so.  For example, if I am told that I am forbidden from doing anything other than being at jury duty, that still doesn't get me actually serving on the jury without some sort of affirmative obligation--except in the trivial semantic sense of a prohibition on not serving on the jury.  But even if we acknowledge that there are some such cases where an affirmative obligation cannot be turned into a negative one except by double negatives, it still does not follow--for reasons I explore in the column--that affirmative obligations are necessarily or even generally more restrictive than prohibitions.


Paul Scott said...

I think, not that dissimilar to car insurance (though more compelling, imo), that the mandatory health insurance is simply a recognition that health is generally a matter of luck and that we all must share in the combined burden. To be sure, single payer is a preferable way of accomplishing the same thing, but to me that is the core fact being recognized by mandatory coverage.

As a libertarian myself, this strikes me as both obvious and proper. Once the decision has been made to classify health care as a need, rather than a luxury, forcing all of society to share in the cost is the only rational choice.

egarber said...

This might be more a question for Sherry, but isn't it also true that the civil nature of such a fine would put it more firmly in the zone of legitimate Congressional authority? In other words, if for some reason jail time was the punishment, it seems we'd suddenly be talking about 5th amendment substantive due process. I mean, I suppose civil arrangements can violate the 5th as well (say, if abortion doctors were fined for each procedure), but not so presumably as something deemed criminal. Or maybe I'm creating a meaningless distinction.

heathu said...

Would the same libertarian or federalism objections apply to Social Security? Social Security is universal, and people are required to pay for it - in the case of SS, through the SS tax. Unless I'm missing something about SS or the individual mandate, it seems the libertarian or federalism objections to an individual mandate were answered some 70 years ago.

egarber said...

Heathu, I'll try this one:

1. Even though it's virtually unavoidable, since everybody has to work, you could still argue that SS taxes are conditioned on employment (vs. universal). Of course, everybody has to get a card (I think), so there is a universal component.

2. I think libertarians have largely given up on the battle as to whether the constitution limits the ability to tax (which grants SS taxing power). With healthcare mandates though, they're hesitant to believe the commerce clause extends that far (to "coercement").

Of course, as Mike wrote a few weeks back, there really is no chance that mandates would be thrown out on commerce clause grounds.

Michael C. Dorf said...

Eric has correctly answered heathu: Presumably the libertarians treat a tax on income (which is what social security really is) as triggered by the voluntary activity of working, rather than the mere fact of existence. I share the view that this is not an important distinction.

egarber said...

Except that "coercement" isn't a word. Coercion, I mean. They don't teach spelling at UGA :)

Unknown said...

How is mandatory health insurance any different than mandatory auto insurance, as required by California's Vehicle Code Sec. 16000 et seq.?

Is this argument being used to rebut the libertarian position?

egarber said...

>>How is mandatory health insurance any different than mandatory auto insurance,

From a policy standpoint, it's not different at all -- the idea is to spread costs. But for the libertarian point Mike is responding to, this would be another example of a "voluntary" mandate, since it only kicks in if you want to drive. So it's not quite the same thing.

Blogger said...

As Professor Dorf points out in his FindLaw article, this "affirmative" requirement has also been characterized as a tax, which generally means that the federal government determines what to do with portions your income instead of you doing so. So, in that sense, it's a negative requirement to not do anything with that portion of income other than pay taxes.

But as taxes go, this is really a great one. Not only do you get to determine how much you pay, but you also determine what kind of services you will receive from your taxes.

There's also good news in this tax for the rich and the poor. For the rich, the tax is flat, unlike the progressive scheme in federal income tax. (And it probably won't be any new tax obligation anyway because presumably they already have insurance.) For the poor, they won't have to pay it. (The cohort of middle income libertarians currently without coverage might still have some room to grumble, though.)

egarber said...

How about this angle? Suppose Congress arranged for the states to enforce the mandates, via a threat to withhold Medicaid funds or something.

Does that make it a general welfare clause (tax and spend) question, rather than one based on the commerce clause?

Prup (aka Jim Benton) said...

First, I want to thank you for referencing the Jacobsen case which I have already passed on to one of the prominent fighters against today's anti-vaxxers.

I do think the auto insurance argument is the besy comparison to use, because it is the one most people are familiar with -- and because it can be made clear to someone like me, the rare exception who has never driven or been in a household that owned a car.

The parallel is stunning:

All drivers are required to have auto insurance to make sure that the cost of accidents is not borne by the society at large or by an innocent driver who should not be required to bear the cost of repairs caused by another driver.

All patients are required to have health insurance to make sure that the cost of illnesses is not borne by the society at large or by an innocent doctor who should not be required to bear the cost of treatment required by a patient.

Angus said...

Prof Dorf:

First I'll characterize both the argument to which you respond and your response. Then I'll argue that your response fails.


The libertarian, as you interpret him, thinks that any given requirement to act infringes liberty more than any given requirement to refrain from acting because a requirement to refrain from acting only forecloses one option, whereas a requirement to act forecloses all but one option. That requirements to act infringe liberty more than requirements to refrain from acting serves as a lemma in this objection to the individual mandate:

1. Requirements to act infringe liberty more than requirements to refrain from acting.

2. If A infringes liberty more than B then A is worse than B.

3. The individual mandate is a requirement to act; prohibitions on acting and conditional requirements to act are both requirements to refrain from acting.


4. The individual mandate is worse than either of those alternatives.

You argue that (1)-(4) fails because (1) is false. You invite us to consider this pair of requirements: (a) the requirement to remain in your house, (b) the requirement to refrain from leaving your house. (a) is a requirement to act, (b) is a requirement to refrain from acting, but they both foreclose the same set of options. So they both infringe liberty equally. So (1) is false.


There are two problems with your counterexample. The first, which superficially appears to help your case, is that remaining in the house is not an action, it is a refraining (at least, you can "do" it without making any bodily movements, etc.).

The second - which shows why the first only superficially helps your case - is that, strictly speaking, remaining in the house is not a refraining but an (enormous) set of refrainings (refrain from going out the door, refrain from going out the window, refrain from going up the chimney, refrain from going to the grocery, etc.)

In order to confute (1) you have to select a single requirement to refrain from acting and a single requirement to act and then show that the former forecloses (at least) as many options as the latter.

(Acknowledged this is hard to do - trying to precisely delineate when something counts as a (single) action is a vexing, thankless task. Good luck.)

The Mudslinger said...

Michael, your Findlaw column on the "libertarian objection" not only misses the point, but gets the objection wrong completely!

Maybe you should have asked a libertarian in the first place.

The libertarian objection is not to the mandated coverage requirement--that's one part of the mess. We object to the entire mess! The federal government not only has NO explicit or implicit delegated power from the People under the Constitution to have any role in requiring the People to be responsible for their health care, but it also has no authority to legislate ANYTHING regarding public health. It never has! At worst, that is left to the several States under the Tenth Amendment, and at best it really falls under personal responsibility absent any force by government at any level.

All one has to do is look at the boiling debate over the over-hystericized swine flu and the untested (and illegal!) poison they call a vaccine for it to see that the People--outside of DC or those who have a financial interest in that mass poisoning-- get it about government mandates, force and intervention, and that they *want* to be in charge of their own health--not greedy insurance companies and certainly not the inept and corrupt government, who is well-documented to be in bed with the insurance and pharmaceutical companies and their profit motives instead of serving the People.

Frankly, the libertarian objection to the health insurance debate (calling it "health care" is both inaccurate and intellectually dishonest) can be summed up thus:

"My health is none of the government's damned business!"

It's a simple extension of the libertarian fact of existence that we own our own bodies and have ultimate say in what we do with them, so long as we don't harm or defraud another.

Dissident said...

Hey, Professor. How would you deal with the reverse slippery slope argument? If you can find one or two examples of compelled payments that exist to justify compelled life insurance, what is to stop the government from compelling the purchase of other less compelling (pun intended) insurance, e.g., veteranary insurance? After all, why should pet lovers not be able to spread their costs to non-pet owners?

Is there any limit on what the government can legitimately mandate indivuals to do? Or, is P.J. O'Rourke right when he says, "The whole idea of our government is this: If enough people get together and act in concert, they can take something and not pay for it."

egarber said...

Dissident, I was thinking about the same thing.

A couple of things though:

1. As I understand it, current commerce clause precedent requires a regulated activity to have a substantial impact on a market to be valid. Some of the smaller-scale hypotheticals wouldn't likely measure up.

2. With health insurance mandates though, people who don't have insurance are *already* market players, because they'll still receive treatment (by law) if they end up in an ER (passing big costs onto those who do carry insurance). That equivalent doesn't exist in your vet insurance example -- pet owners are completely outside the market.

So I think it's possible to argue that it isn't even as extended as what we saw in Raich (home grown pot affecting the market). Even if one doesn't have insurance, he's directly engaged.

egarber said...

I have another question. Might this be a general welfare (taxation) clause matter as well?

If the presumption is that you have to pay a tax absent acquiring insurance, isn't that the same as saying that filers should assume to pay a tax, unless they make charitable donations?

In fact, I think I read that the mandate is being framed that way in the legislation.

Dissident said...

Here is another thought.
What if the tax incentives for providing employer paid insurance were eliminated, e.g., insurance premiums were not deductible as a business expense or had to be included as income by the employee? I think the market for health care and insurance would become a lot more efficient if patients and health care providers eliminated the middle man for most services. Most people would likely get Insurance that only covers health catastrophes.

I never could understand why it is efficient for my health care insurance to pay for my routine checkups. Imagine what would happen to the price of auto repairs if every oil change was covered by tax deductable insurance provided by employers.

In short, the real problem is perverse tax incentives that keep the cost of health care artificially high. The solution to high health care costs, Obama says, is to have even more insurance rather than to let the cost be set by actual users and providers. This is an example of what Milton Friedman called the tyranny of the status quo.

egarber said...

>> In short, the real problem is perverse tax incentives that keep the cost of health care artificially high.

For this to be true, somebody would have to prove that doctors charge more for visits and tests than they would otherwise – those are the costs that are negotiated with insurance companies. But since margins aren’t extravagant for doctors (from what I’ve seen), I don’t see how this can be the case. If anything, it’s possible to argue that as disastrous as it is today, it would be worse without those demand-side incentives, because there would be no buying leverage through pooling. Or put in another context, public sector investments in education make it universally affordable for families to educate their children. Remove those supports, and millions would be priced out.

I guess what I'm saying is that at some point increased demand creates critical mass, where a large number players then compete to drive prices down. Electronics is a clear example of this dynamic.

In an environment where that demand -- subsidized or otherwise -- doesn't create those efficiencies, it's often helpful to look at the market structure itself, to see if de-facto or actual monopolies are in place. And indeed, with health insurance, we know that in the top 40 markets more than half of the business is controlled by 2 or fewer insurers -- hence the recent effort to remove their anti-trust exemptions.

Dissident said...

Imagine if employers could provide tax deductable food insurance that is not taxable as income to the employee. When I go to the supermarket, do you think I will buy the store brand of ice cream or a quart of every flavor of Haagen-Dazs? While the analogy may not be precise, it is clear to me that the further the consumer is removed from the actual cost of the services he gets, the less likely efficient economic discipline will be felt by the health care provider. I wonder if the price of cosmetic surgery, which is generally paid for by the patient, has risen less sharply than other medical costs over the years. Does anyone know the answer?

egarber said...

>>When I go to the supermarket, do you think I will buy the store brand of ice cream or a quart of every flavor of Haagen-Dazs?

Gotcha. Two points:

1. On the economic dynamic I brought up, if enough people start buying Haagen-Dazs, other players will enter the market, which will put downward pressure on price. Remember, I think you were contending that subsidies drive up price. I think you just proved that instead, they might open access to higher-quality products for more people.

2. In the bigger healthcare context, the question would be whether the store-bought brand represents quality / comprehensive healthcare. This is really what the debate is all about -- i.e., is healthcare core infrastructure, or a mere commodity?

egarber said...


One other quick point. There's a massive move toward co-insurance (vs. co-pays) underway right now, which in part is designed to do what you say: force consumers to have more skin in the game.

But is it a truism that such "consumerism" (that's what our open enrollment guy called it) will automatically drive down costs? To me, it just means employees will have to pay more -- and that they might forego necessary treatment as a result. So what suffers in the end is quality.

Crispian said...

Prof. Dorf,

The two examples of federal mandates in your article both have at least an implicit grounding in the Constitution (draft registration and jury duty). Requiring citizens to buy health insurance has no such grounding.

Your other examples are state initiatives, which easily fall under states' police powers.

I know you only intended to address a libertarian argument in that first article, but it is misleading to reject the libertarian argument against a federal mandate by looking to state actions.

Based on my above points, I think there is a great libertarian argument against a federal mandate - just not against a state mandate.

I would be curious to hear your perspective on the tax implications - direct (must be apportioned) v. indirect. It is clearly a capitation which would fail for non-apportionment in my view. Wouldn't it be novel as an "excise" tax, since one is taxed for a non-action?

Anonymous said...