The St. Louis Fed published a paper that bears on this subject. The title is "Is the United States Bankrupt?" (http://research.stlouisfed.org/publications/review/06/07/Kotlikoff.pdf). The author (Laurence J. Kotlikoff of Boston University) looks at the relationship between immigration and the future finances of the U.S. I quote

"CAN IMMIGRATION,PRODUCTIVITY GROWTH, OR CAPITAL DEEPENING SAVE THE DAY?

Many members of the public as well as officials
of the government presume that expanding
immigration can cure what they take to be fundamentally
a demographic problem. They are wrong
on two counts. First, at heart, ours is not a demographic
problem. Were there no fiscal policy in
place promising, on average, $21,000 (and growing!)
in Social Security, Medicare, and Medicaid
benefits to each American age 65 and older, our
having a much larger share of oldsters in the
United States would be of little economic concern.
Second, it is mistake to think that immigration
can significantly alleviate the nation’s fiscal problem.
The reality is that immigrants aren’t cheap.
They require public goods and services. And they
become eligible for transfer payments. While most
immigrants pay taxes, these taxes barely cover
the extra costs they engender. This, at least, is the
conclusion reached by Auerbach and Oreopoulos
(2000) in a careful generational accounting analysis
of this issue."

It turns out that the standard Social Security immigration sensitivity analysis contains a large flaw. The analysis assumes that immigrants have skills and earnings equal to natives. Obviously, this is not the case and the gap has been growing for decades. Of course, the analysis also includes the SS taxes paid by the last cohort of immigrants (in the final years of the 75 year projection period), but not the costs they will eventually impose.

The National Academy of Sciences report in the 1990s (and many other studies since) have shown that low skill immigrants are a major net burden on taxpayers. No one should be surprised by this. Native poor people are clearly a net burden on taxpayers. Why should imported poor people be any different?

Actually, anyone who tells you what the skill mix of immigrants will be in 2030, 2040, and 2050 is telling you that they have no clue what they are talking about. The world will have an enormous supply of highly-skilled workers who could potentially immigrate to the United States in these years. How many actually come will be a matter of policy.

In principle, projections are supposed to assume current law stays in place, but it is extremely hard to say what current law is, since "current law" means having laws that are largely unenforced as a matter of policy. So, a projection involves projects the degree of non-enforcement of law in a very changed economy. That ain't easy.

It would be stupid to make immigration policy based on its impact on Social Security since its impact on the economy and environment is far more important. Nonetheless, it is outrageous to have Social Security projections that do not incorporate the best available assumptions on immigration. On this point, I would take the views of the economists and demographers at the Census Bureau over President Bush's Social Security trustees any day.

btw, if we made a point of reducing protection for doctors, lawyers, and other highly paid professionals, it would reverse some of the upward redistribution of the last three decades, which would in turn eliminate a large share of the projected shortfall in Social Security.

Economic projections of almost any kind 30 or more years in advance are nonsense. Politicians' real time horizon is the next election, so when they talk about "reforming" Social Security for 2045 you have to ask qui bono in the short term. Privatizing SS would mean an immediate windfall for Wall Street and present stock owners.

In fairness to the Trustees they are handicapped by a requirement to assume current law and particularly the caps built into the Immigration Law of 1990. Something I didn't understand until I read the recent Social Security Advisory Board's Technical Panel Report on Assumptions and Methods. The Technical Panel suggests changes the numbers in a way consistent with the new Census Dept numbers and also suggests some changes in methodology that would allow the Trustees to escape the handcuffs that tie them to current law. The discussion of this specific point comes on p. 45 of the above report.

It should also be noted that illegal immigrants will never draw a penny against wages with SS withholding...

As to skeptonomist's point absent other changes in tax rates and/or benefit cuts privatization simply would mean diverting the current cash surplus of around $80 billion a year into other asset classes which in turn ceteris paribus would add an equivalent amount of borrowing needs. So you would get some offsetting effects as the bond and stock markets adjusted to a new supply of cash on the one hand and a new demand for cash on the other.

I am not saying that $80 to $100 billion per year for the next decade (at which point Social Security cost is projected to fall behind Social Security tax income) is chump change or that Wall Street wouldn't want to skim off whatever fees they could, but the fundamental drive to privatize Social Security is not really driven by pure greed but instead by ideology. If you examine the actual chronology you can see that the sequence was Cato going to get support from Wall Street rather than the other way around. See Cato Journal Fall 1983: Social Security, Continuing Crisis or Real Reform and particularly Butler and Germanis.

Some people have always hated Social Security on principle and not because they have any real desire to collect the miniscule fees you would get on a below median wage worker's PRA. If the goal was to generate income for Wall Street the appropriate strategy would be to push for lowering the cap for higher income workers, implementing a mandatory IRA carve out plan for them and them alone, and leaving lower income workers completely out of the picture. Instead the ideologues are out to prove that Social Security is a failure root and branch even though in truth they don't have the numbers.

Certainly greed is in the picture, but pissing on FDR's grave and discrediting the New Deal is what established the basic frame.

Are those legal or illegal immigrants and will they be paying SS or not?

Dean,

Excellent post.

As usual, Peter Schaeffer is full of misrepresentations. The Kotlikoff quote is a typical example of "entitlement hysteria," with him lumping social security in with medicare and medicaid. Yes, the latter two have big fiscal problems, but these involve the medical care system in the US more generally and are really distinct from the far-less troubled social security system.

As for Peter S.'s freaking out over low wage immigrants (most of whom are young), jim in austin has the ultimate reply. Indeed, many illegals are paying social security taxes but none will get social security benefits. Even if P.S.'s cited studies on low wage immigrants ultimately costing more for ss than they put in, that will almost certainly be offset by the huge net gain from the illegals.

Mr. Baker,

I would be the first to agree that the skill mix of immigrants in 2050 is hard to predict. Indeed, given peak oil, global warming, Asian ascendance, etc. it is hard to predict whether immigration will greater or less than current levels, 42 years from now. It is worth noting that when the Immigration Act of 1965 was passed, the authors assured the American people that the overall level of immigration would not materially change. Obviously a lot can change over a period of 42 or 43 years.

I agree that the environment and the economy are more important immigration issues than Social Security. Indeed, the Center For Immigration Studies released a report assessing the impact on immigration on greenhouse gases (actually just CO2) a few days ago. See "Immigration to the United States and World-Wide Greenhouse Gas Emissions" (http://cis.org/GreenhouseGasEmissions).

It is not unreasonable to suggest that the SS projections be based on empirically based estimates of future immigration, rather than some fixed formula. However, the same logic of using real world data also indicates that the low skill levels of the actual immigrant population be taken into account. Given the progressive nature of the SS benefits formula, this might well mean that larger immigrants flows will diminish future SS solvency.

I don't think the accusation of political bias with respect to the SS Trustees is legitimate. The 2000 SSTR report assumes 900,000 immigrants as the base case. The comparable number in the 2008 report is 1.07 million immigrants. These numbers aren't very different and perhaps more relevantly, if the Bush appointees were trying to make SS appear less solvent, presumably they would have reduced the number of assumed immigrants in the intermediate case. The other problem is that the Bush administration has strongly supporter higher, not lower immigration levels...

I am not so sure that higher levels of skilled immigration will actually reduce inequality. Indeed, the converse case can be made. While it is quite true that returns on education have increased in recent decades, that is not the entire story of growing inequality. Piketty and Saez have demonstrated the amazing growth in incomes in the top 5%, 1%, 0.1% and even 0.01% (of the population) that can not be explained by increasing returns on education. Providing cheaper labor to corporations via increased white collar immigration could accentuate rather than diminish this trend.

BR,

The only way illegal aliens can provide any long-term actuarial benefit to the SS system is if they stay illegal. In other words, if the U.S. continues to employ large numbers of illegal workers who never gain legal status or return to their own countries. Somehow that I rather doubt you advocate any such future for the United States.

In any case, the SS taxes paid by illegals are rather small, inevitable given their very low incomes. Checking online, most estimates are in the range of $7-9 billion per year. That's around 1% of total SS revenues for 2007. The idea that the U.S. can (or should) attempt to sustain the solvency of the SS via illegal aliens doesn't appear to be factually supported (much less morally justified).

"The only way illegal aliens can provide any long-term actuarial benefit to the SS system is if they stay illegal."

PS,

You're absolutely right. BR and the open borders crowd understand that fact.
Except they never intend for the illegals to stay illegal.

P.S.,

Your argument depends on a combination of age of starting to work and wages they earn of immigrants (the legal ones, that is).

More generally, the longer term outcome depends more on prodcutivity growth than demography.

If you wish to refer to me as part of the "open borders crowd," well, heck, I shall refer to you as part of the "out from under a rock crowd."

"BR and the open borders crowd understand that fact.
Except they never intend for the illegals to stay illegal."

Actually, you're wrong there. They're not really open borders, they're just a pro Hispanic lobby. They will not open up any guest worker programs for anyone else.

And when the illegals become legalized, they'll want their SS benefits and then some.

BR,

Please check the posts. You are replying to the wrong person.

Dean, can you give a source or explain your calcs for the 30% number? I'd like to follow up on it.

Thanks,

Steve

Dean,

What you've basically done is focus on one assumption from the Census projections (immigration) that make the problem look smaller but not another (life expectancies) that makes it look larger.

One way of looking at how Census's total projection nets out is to compare the ratio of working age to retirement age individuals (this is similar to the worker-beneficiary ratio). Under the 2008 Trustees projections, the ratio of age 20-64/65+ in 2050 will be 2.81, while in the Census projections it's 2.68. So on the demographics alone Census projects a slightly worse situation.

Plus, since immigrants tend to earn less and collect more from Social Security than native-born, a larger share of Census's working-age population will receive more in benefits than they pay in taxes, which obviously doesn't help solvency. (SSA's projections that immigration improves system financing is in part based on the assumption that immigrants earn the same as native born, which isn't the case.)