Arnold Kling  

Libertarian Basics

Pleased as Punch... Housing Rents and Bubbles...

is the title of my latest essay.

Consider the following classification system for government regulations and programs.

(a) interventions that work so much better than private alternatives that we feel grateful for them
(b) interventions that are better than private alternatives in some ways and worse in others
(c) interventions that are mostly worse than private alternatives
(d) interventions that are evil

Libertarians look at government and see interventions that are mostly in categories (b), (c), and (d). I would put municipal fire departments in category (a), government water treatment in category (b), public education and Social Security in category (c), and protectionist trade measures such as the Byrd Amendment in category (d). Where the United States is really lucky compared with countries like Zimbabwe is that those other countries' government interventions are predominantly in category (d).

My sense is that non-libertarians view interventions as fitting mostly into categories (a) and (b), and they believe that the programs that they favor are all category (a). I believe that their attachment to government interventions owes more to wishful thinking than to a realistic assessment of results. My reading of history is that progressives tend to exaggerate both the need for government interventions and the likely results of such interventions.

For Discussion. How would you classify Social Security, and why?

Comments and Sharing

CATEGORIES: Economic Philosophy

COMMENTS (17 to date)
Eli writes:

My take is that Social Security is (c) for its purposes and (d) for its execution. The goal of everyone having at least some stream of income in their old age is nice, even if it does not justify the coercive intervention of the state to accomplish it. The truly evil part of Social Security is the deceitfulness that goes into it. The employee part of the payroll tax is taken in a way that most workers don't even notice. Those workers who do notice it may not know that they should really be doubling that part of the tax in their mind, because the employer portion also comes out of their paycheck indirectly. On their Social Security "statements", there is no indication of what they have "paid in" so far, nor is there an indication of the actuarial balance. There is only a section that reads along these lines: "If you were to continue working for XX years at your current salary, under current law, you would be eligible for a benefit of $X,XXX per month." That's not informative at all.

If there were any effort to honestly inform "participants" about their participation, I might be able to concede that the program as a whole merits a (c). Unfortunately, I lean towards a (d).

Randy writes:

I go with c, interventions that are mostly worse than private alternatives, and focus on the word "mostly". For most reasonably intelligent people with a reasonably solid work ethic, social security is a worse deal than they could get through private investment of the same funds. But for a few, mostly lacking in the above mentioned attributes, social security is a great deal. The program is focused on the needs of these few, at the expense of the many - which is why it should be allowed to transform into a needs based program.

spencer writes:

I would rate education and social security as b.

In education we really have three systems.

The well off live in affluent communities and have great schools.

The middle class live in average communities and have ok schools.

The poor live in poor communities and have poor schools.

The private sector could provide superior results in the wealthly community because the studuents are prepared to be educated and the parents would pay for superior teachers.

In the middle class community the results would be mixed.

In the poor community the results would be worse
because they kids are not prepared and they could not afford the better teachers..

Social security is similar. The affluent would be better under a private system and the below average would be worse off because SS is designed to be an income transfer from the well off to the poor.

KipEsquire writes:

This is very helpful, but I think there should be an additional category between (b) and (c): "interventions that are positive for some members of society and negative for others, because some members are deemed more in need, or more deserving, than others"

Social Security, and progressive income taxation, and just about everything else falls into this "(b+)" category, and this is where most of the battles between libertarians and non-libertarians occurs.

Timothy writes:

I'm going to go ahead and call Social Security (d), evil.

It forces one to pay for the retirement of others, it puts the government in charge of at least part of your now/future expenditure trade off, and people under 30 (such as myself) won't likely see a dime. It's got great intention, sure, but that doesn't matter. The results are crap, the pay-outs are paltry, and it makes people reliant on government for their income.

ed writes:

This is an interesting classification system.

The main problem I see is that there are multiple possible "interventions," (and perhaps multiple possible "private alternatives" as well, depending on how we define property rights.)

Even if we assume we can somehow rank these possibilities in terms of social welfare, the system we get in practice will usually not be the first best among all possible systems. Furthermore, we don't really even know what systems are realistically "possible" given our limited understanding of political, legal, and bureaucratic processes. And some systems may be possible to operate but politically impossible to enact.

In the end, it's not clear what we are comparing to what when we decide if an "intervention" is great or evil.

Is public education good or bad...compared to what?

Jeff Bergman writes:

I'm not quite sure where I'd place Social Security in that categorization scheme (either B or C), but clearly the relevant factor is the likely level of individual saving without SocSec. I'd like to think that the orthodox libertarian position -- that people will tend to save and invest a rational portion of their income -- is correct, but I don't have complete confidence in it. There is such a thing as short-sightedness among economic actors, and many people can probably be expected to misjudge their own long-term interest and "under-save" (or over-consume). If that's the case, then to the extent that SocSec forces a certain conservation of financial resources it is doing a better job than private behavior alone.

Note that this doesn't address another side of the issue: even as a system of "forced savings" SocSec is pretty badly designed, since we all know that real investment under the program is nonexistent. Private accounts would have rectified this to some degree, of course, but those seem dead in the water at this point.

Robert writes:

The problem with everyone providing for their own old age is that vast uncertainties exist in when any given individual will become unproductive, how much longer after that they will die, and what their financial needs will be along the way. Since the consequences of undersaving (destitution in one's old age) seem more dire than those of oversaving (unnecessary frugality in one's productive years), if the decision of how much to save is left to individuals, most of them, being risk-averse, will oversave.

On a macro level, this can eventually produce deflation. (See Japan & Germany for possible present-day examples.)

So I can see a benefit in ensuring everyone some level of income throughout their old age, one that may well outweigh the moral hazard of some people extracting uncompensated value from the pension scheme. So on that principle, SS could have as high as a (b).

However, the economy's overall return on the investment seems disappointing, so in practice, it gets a (c).

DK writes:

Are you serious that "non-libertarians view interventions as mostly category (a) and (b)"?

Perhaps you are defining libertarian more loosely than most people do, but, there are a lot of small goverment and free market types who are skeptical of government intervention but who can't be described as libertarians. Professor Bainbridge is an obvious blogosphere example.

Jon writes:

Those who advocate privatizing primary and secondary education forget an important part of economics 101--what drives efficiency in the private sector is that people get only what they are willing and able to pay for. This eliminates education approaches or services that people don't value, but it also means if you have no money means your children don't go to school. As soon as government starts paying for some people's education this constraint on costs disappears.

Glen Raphael writes:

I'm curious why you put municipal fire departments in category (a). Have you read _The Enterprise of Law_? Private, competitive subscription-based fire departments can be - and often have been - cheap and effective.

Sure, one could imagine a world in which free-riding would be a problem, but it's not the world we actually live in. Given that essentially everybody has a mortgage, all mortgage companies require fire insurance, and fire insurance is cheaper if the property has a preexisting relationship with a competent nearby fire protection company, all the pieces are there to get a fire subscription bundled with insurance or mortgage payments. Or to have the insurance policy reimburse /actual/ costs to put out the fire if one isn't a subscriber.

Bernard Yomtov writes:

Well, Glen, let's see. You're walking down the street and notice that a building has caught on fire. You decide to call for help. Which fire company do you call? Oh, there's a central dispatcher, so you just call them? OK, but the building is a condominium. Which unit is on fire, because, you see, different units have different fire protection services, just as they have different mortgages. Besides, maybe that's the unit the same people have lived in forever, and it's paid off and they don't have a contract. what then?

Myabe, just maybe, getting rid of all those information problems and free riding issues makes sense.

Nicholas Weininger writes:

Social Security definitely gets a (d). One-eighth of people's income is a *lot*. There's tons of things people might choose to do with that money if they had it back-- and we have no business passing judgment on *any* of them.

Consider, for instance, the fact that, without SS, some people would save relatively little and be poor in their old age as a result. Of that segment of people, part would be composed of the shortsighted or generally dumb, and part of the devil-may-care: the folks who would honestly rather enjoy their money in the prime of their lives, even if it means privation later. How do you systematically distinguish between those two subgroups? And how on earth do you justify taking away the right of the second group to order their lives as they please?

monkyboy writes:

I would rate SS an a.

It's popular and cheap to administer. Imagine the level of corruption that would occur in a private sector version of SS.

No wonder Wall St. is funding the SS privitization campaign. With derivatives about to hit the fan, they need a new source of easy money to loot.

Randy writes:


Re; " you justify taking away the right of the second group to order their lives as they please?"

Exactly. The answer to that question is the dividing line.

Victor writes:

I would rate SS a C- : Interventions that are mostly worse than other possible interventions.

To whit: gov't could require private party investments in individual retirement annuities and simultaneously eliminate specious underwriting criteria (creating so-called "guaranteed issue" products much like what HIPAA did in the small group health market). Couple this with some progressive taxation of the monies accumulated prior to retirement in such funds and/or transfers to the poor, and you have something that is unarguably superior to what we have now, IMO.

On the flip-side, I cannot imagine anyone who would actually argue in favor of mandatory "investment" in a moderately risky "average wage index" portfolio coupled with mandatory purchases of consequence free "retirement options" for an unlimited number of possible spouses. The disincentive to work early in one's career because of the 35 year max payment history is also striking and surely sub-optimal; although this feature does provide "insurance" against temporary loss of job during one's lifetime, this simply doubles the effect of the progressivity of the taxation/benefit scheme while tacking on the obvious work disincentive. Social Security as it currently stands is extremely hard to defend. (for a perspective from the left on this issue, see Shiller's 1998 paper on Social Security and macroeconomic risk)

Regarding your classification, if I could live in a world where I was supreme (and benevolent) dictator, I wouldn't be a libertarian because I could mandate optimal policies including flexibility for gov't and regulators to adapt to future conditions. However, because all policy proposals have to wind their way through the political process (a good thing insofar as this causes us to avoid policies in category (d) ), it seems like we are always left with suboptimal policies. The result is that people like me have to not only argue against the suboptimal policy, but if we want to really prevent the suboptimal policies we have to nip them in the bud by also arguing against the nirvanic policy-mirage of what-we-could-theoretically-do.

In other words, it isn't "private v. public" that drives my views, it's an issue of practically recognizing "the best we can do". I think you did hit this theme effectively, but in terms of your classifications, a "C-" category seems useful to reinforce this point.

Christian writes:

I would classify SS as legal plunder. :)

Same with everything else that I am forced to pay for that I don't use and everything that isn't constitutional.


Comments for this entry have been closed
Return to top