Arnold Kling  

From Musgrave to Shaviro

PRINT
Giving Up, Failing Out, and th... Kidney Bleg...

I was sent a copy of Daniel N. Shaviro's Taxes, Spending, and the U.S. Government's March Toward Bankruptcy.


Our march toward government insolvency is a complex historical event with multiple causes. The central causes involve health care technology and demographics, and are being faced by countries around the world...Recent tax cuts and spending increases, however, have made the problem much worse.

Some other quotes from the book.


corporate executives would not go to fail for failing to publish fiscal gap-style measures...But they most assuredly would go to jail if they published financial statements that ignored accruing liabilities...the consequence of this omission for Social Security in 2002...was that the system reported a $165.4 billion increase in assets, whereas it should have reported a $467.5 billion loss (p.7)

...One peculiar consequence of this use of Social Security and Medicare for hidden redistribution has been a frequent tendency for Democrats to favor making the programs less progressive, while Republicans favor making them more so (p. 154)


The theme of the book is that we are in fiscal trouble and that the political environment surrounding fiscal policy suffers from a number of ailments.

I found myself longing for an overall analytical framework. Back when I studied Public Finance, we learned Richard Musgrave's approach (Shaviro briefly mentions it on page 30). Musgrave said that you could think of government as having an allocation function, a stabilization function, and a distribution function.

Allocation means changing the mix of goods and services produced in the economy through production, regulation, taxation, and subsidies. Stabilization is Keynesian changes in fiscal policy. Distribution is redistributing income.

What has happened since 1959 (when Musgrave wrote), is enormous growth in what might be called the intertemporal allocation/redistribution function of government, namely Social Security and Medicare. Musgrave did not have to pay attention to this function. Now, it is the dominant factor in fiscal policy.

Unfortunately, Shaviro does not offer any sort of framework for optimal intertemporal allocation/redistribution. We are left to assume implicitly that the challenge is to try to come up with a politically balanced way to make our existing Social Security and Medicare systems sustainable.

In other words, we don't know what it is about Social Security and Medicare that we ought to want to save, but we need to find a politically centrist way to save them. Kind of unsatisfying in the end. I wish we had a Musgrave-like explanation of what the economic functions are for our entitlement programs.


Comments and Sharing





COMMENTS (11 to date)
Tim Lundeen writes:

The interesting thing about these concerns is that they reflect todays conditions. They totally disregard the world as it will be in 20, 30, or 40 years, which will be a very different place.

It is entirely possible that in 20-30 years that medical costs will be dramatically lower because we will have cured all common illnesses, and that healthy, productive lifespans will be dramatically longer. If this transpires (and if not in 20-30 then it certainly will in 30-50 years), we will not have either of these problems.

Shaviro should read Kurzweil's book The Singularity is Near and stop worrying :-)

Edgardo writes:

I'm surprised by your last sentence: "I wish we had a Musgrave-like explanation of what the economic functions are for our entitlement programs." Musgrave's classification was good for teaching before Public Choice, that is, when neoclassical economists believed that the allocation and redistribution effects of government spending could be separated. Public Choice (it was already in the first edition of Mueller's classical textbook) made clear that this separation was at least misleading. As shown by your own analysis of many policy issues, I don't think we need new tools to analyze them. We need to apply available tools wisely.
I agree with Tim Lundeen and I add another important reason to reach the same conclusion. Having worked on fiscal reform in several countries (including my home country, Argentina), I think that Shaviro's view on a march to government insolvency is wrong. It is based on a simplistic extension of the idea of insolvency as applied to private companies, whereas the relevant concept for public finance is the "soft budget constraint" and how this constraint evolves as expenditures threat to reach unsustainable levels (that is, a flow concept rather the stock concept underlying the idea of insolvency). The evolution of this constraint depends largely on the political system and while the system of most countries may delay a reaction long enough to impose the high social costs associated with large redistributions (the Argentinian experience in the past 70 years), I still believe the US system will cut social welfare benefits before these costs are high.

Barkley Rosser writes:

I would say that poking at Musgrave's approach in comparison with public choice because one cannot fully separate the three functions is not really all that useful. Yes, it is a fact: anything that a government does at any level (including local) will have implications/effects for all three functions, even if the motive for the government action is focused on only one or two, and the other effects are strictly secondary. Thus, one can identify most government actions as being primarily directed at one or two functions, even if it affects other one(s) as well, and it is still useful to think in such terms, even if Musgrave is a) passe, and b) dead.

Regarding social security, I know I am sounding like a broken record, but I really would like to see one or both of you respond to the point that has been made here before (mostly by me) that there really is no fiscal problem with social security, please. I am not going to repeat the argument at length, but simply remind that reality has performed better than the supposedly optimistic "low cost" projection of the SS Trustees, and that the scenario has the fund running a surplus every year forever, and the projections for the future have not been noticeably altered, while virtually all the MSM and most of the politicians in both parties continue to trot out the thoroughly discredited but scary sounding, intermediate projection as "what is going to happen." Are you guys going to fess up that you are feeding one of these "conspiracies of silence" that Bryan's brilliant lunch partner, Robin Hanson, likes to complain about?

Matt writes:

If the central bank keeps money stable, things cannot go too bad. We have the underground economy after all, and in the worse case, the official economy will be hollow and we operate in the underground.

Chris writes:

I'll paraphrase Stein's Law here. If something can't be sustained forever, it won't. Government can, at any time, adjust taxes or entitlement spending to bring things back into balance. Private companies can't do that and the government can't do that with its publicly-held bonds, because these liabilities are contractually determined. Entitlement spending and taxes are not.

This doesn't mean that the results are going to be pretty. But the choice is among some mix of higher taxes on younger people and reduced transfers for older people. No fiscal measure currently available really takes that into account. The public debt/GDP ratio tells you whether the government has historically tried to violate a transversality condition. It hasn't, and the debt/GDP ratio is actually behaving quite nicely these days. The accrual measure that some people propose says that the value of liabilities/GDP is infinity. That isn't right either because of Stein's law ("error correction" for you time series geeks) but it does capture the need for some adjustments in the future.

The public plus "trust fund" debt / GDP ratio is a weak signal of the government's willingness to increase the future public debt to fund these things, but that has the perverse consequence that an increase in SS/Medicare taxes or a decrease in benefits now INCREASES that implicit debt regardless of future policy actions. It also represents no constraint on what the government will or will not actually do in the future since the government has these "bonds" both as an asset and as a liability on its books.

As far as the economic functions of these things go, the Stiglerite in me says that they buy votes and reflect what some people actually want. No coincidence in my mind that the big expansions in entitlement programs coincided with big expansions in the right to vote and a big increase in the number of older voters. A prediction coming from this is that Baby Boomers in 10-15 years will vote to keep benefits high and tax the hell out of younger people. This is what has already happened in much of Europe. In other words, if you want to get worried, look at the political economy and not the accounting.

Barkley Rosser writes:

I am going to put my earlier remark/request slightly differently, given that it has been greeted with silence so far.

It is perfectly reasonable to oppose social security because it is a big government program if one is a libertarian who opposes big government programs. That is sort of implicitly lying behind the argument about how it is distorting the economy in the post. I may disagree with this position, but I can respect it as a straightforward ideological position.

What bothers me is scarfing along on this baloney that it is financially troubled in the long run. This involves assuming things like that the the economy is going to start growing at 1.8% per year very soon, when even with a housing slowdown going on last year we had 3.4% growth, way over what is needed to have the social security trust fund always run a surplus with no changes.

So, which is it? Do you guys oppose social security on ideological grounds or do you wish to buy into the half-baked intermediate projection baloney beloved by so many commentators, or just give it a boost by repeating it without any commentary, which is what you all have been doing up to now?

Robert writes:

It will be fascinating if the NIH study on EDTA Chelation therapy confirms its usefulness as a low cost replacement for much of the heart-surgery-industrial-complex.

http://nccam.nih.gov/news/2002/chelation/q-and-a.htm

Since there is no profit in it for the private sector, this type of research needs to be pushed by the government.

Robert writes:

Barkley,
I'm just the software guy here. But I'll give it a try.

1. This year's 3.5% growth was fueled by a world wide liquidity bubble that is unprecedented. See http://www.yieldsz.com/ for an amusing reflection of this.

2. Our most productive age group, the boomers, are approaching retirement. Unless most of us choose to work beyond retirement age, it will be difficult to make up the lost economic impact. One way to force that is reduce the value of our Social Security annuity.

3. I personally am in favor of Social Security as a social welfare program. Not as an entitlement program.

Arnold Kling writes:

Barkley,
Social Security eats up a huge amount of the U.S. tax base. You can say that under reasonably optimistic assumptions, that tax base will grow as fast or faster than the promised benefits. However, it is still not a good idea to make those promises in an environment where the government's promises all together add up to way more than its expected revenues at reasonable rates of taxation.

In the context of massive over-promising, it is simply whistling into the wind to claim that Social Security by itself is solvent if there is enough economic growth. It's like saying that part of the Titanic won't be hit by the iceberg. Not very comforting.

Edgardo writes:

First, my previous comment on Musgrave's classification of government functions was to make the point that any positive analysis of a government program cannot be based on a separation of allocative and distributive effects. Barkley: in the last chapter of his textbook (3rd. ed.) Mueller attempts to save the distinction between these two effects for normative analysis because "... I am reluctant to write off the achievements of the first 50 years of public choice as a catalogue of the deficiencies of democratic decision making." So the relevant distinction is between positive and normative analysis.
Second, regarding social security, my previous comment did not mention it explicitly. It was part of the larger problem of dealing with the "soft" budget constraint problem. Again, the relevant distinction is between "hard" and "soft" budget constraints, usually the first one relevant for private finances and the second for public finances. One of the implications of the idea of "soft" budget constratint is that whenever a government faces the prospect of a deficit it should consider all options to raise money and/or to default in its payments (as I said, I'm Argentinian and old enough to know about the many ways governments can use to do it).
Finally, any serious discussion of social security should be based on alternative programs, paying special attention to how they are implemented. I have now two pensions: one from a benefit-defined program (as a retiree from an international organization) and one from a contribution-defined program (as a retired professor from a Chilean university). The first program is not longer available to new staff because it promised too much, and the second is revised often because politicians want to increase the expected pensions of low-income employees. I have come to the conclusion that (a) I'd be much better off if I had been given the opportunity to save tax-free the same amount of contributions in a private account, and (b) the government would be much better off I had paid taxes when taking money out of this account.

Barkley Rosser writes:

Robert,

Assuming the intermediate demographics projections, the magic number is 2.2%. As long as the economy grows more rapidly than that, social security will run surpluses every year forever. I would agree that there are a variety of reasons why we might expect the US growth rate to decline from its historic average, with a near term issue being the US's rising global indebtedness, which must be dealt with somehow in some way, and is the real problem with our fiscal deficits. But we can go a lot lower than we have been and still be OK.

Regarding the second point, most of the proposals being cooked up do not involve cuts for the baby boomers, especially the older ones like me. We'll get grandfathered in, like the current recipients. It is the benefits to younger workers that are threatened, but most of them are totally ignorant and think they won't get any social security benefits anyway, so they can be suckered with "we are cutting your future benefits so that somewhere down the road in the future we might not have to cut your future benefits."

As for Point 3, and also to Edgardo, I would say that social security is an insurance program. It is a guarantee of income even if everything else goes bad and one ends up with nothing else. We do not expect even positive rates of return from buying most insurance (certainly none that is privately provided), so that there is even a positive rate of return at all expected is pretty impressive. We already have tax-deferred IRAs and so forth, so one can get a break from the government for savings. Those can be expanded without doing anything to social security.

Arnold,

Thanks for replying. I suspect this means that you might agree with something like how I think Ben Bernanke views things. Putting the whole budget together we have the SS system lending the rest of the government to the tune of $186 billion per year, which has been rising, but will probably start declining once the baby boomers do start retiring, thus putting some pressure on the rest of the budget, which is totally in a major deficit. The tax changes of the Reagan years basically amounted to cutting back income taxes (progressive) and increasing FICA (regressive). There will be some pressure, especially as the balance shifts between now and 2030, even if, as I think is likely, the fund never actually runs a deficit.

So, over on maxspeak awhile ago I posted that Bernanke signaled his support for something to be done to/about social security, even though I think he is smart enough and well informed enough to know that the intermediate projections are a bogus joke. I think the issue is a bit like yours: he is concerned about the overall deficit and is looking for something, anything, to reduce it, and also is looking at the politics of what is possible/doable.

So, on the spending side we have five big items: defense, social security, interest on the national debt, medicare, medicaid. As long as the war in Iraq is on, little can be done about the first. Cannot touch interest on the national debt, except by lowering the deficit or reducing interest rates, and those are minor anyway. Obviously there are huge problems and already existing deficits in medicare and medicaid, but these ultimately involve broader problems in the medical care system, as you well know.

On the tax side, we have the general resistance of the Republicans to any tax increase. So, given all this what can be attacked?

Ah hah, we have all these young people suckered into thinking they are not going to get any social security benefits anyway,with the constant drumbeat in the MSM about "social security crisis" simply reinforcing this idea. So, we get Paulson to go to Conrad (as he has) and propose a bipartisan deal: a social security tax increase for a cut in future benefits for young people. If this done there are no immediate savings on the expenditure front (they come later), but there is a nice FICA increase that will allow the social security system to lend even more money to the rest of the government, with some reduction of the near term deficit (I do not wish to imply that you are a supporter of tax increases).

Of course the latest is that Norquist and Cheney are set in concrete against any deal on the tax side, and while some leading Dems seem to favor such a deal, there are others who will fight against a benefits cut pretty hard. Looks like it probably won't happen, but I can see why Bernanke would see social security as perhaps the only large area where some kind of a deal that would lead to a deficit reduction in the near term might be possible.

Comments for this entry have been closed
Return to top