2004 August 13 Friday
What Would Revenue Positive Immigrants Look Like?

The redoubtable Tyler Cowen of Marginal Revolution has posted a bullet of policy items he's like to see implemented in Bush's second term (assuming Bush gets reelected). One of Tyler's items concerns immigration and, practical economist that he is, Tyler argues for the benefits of immigrants who pay more in taxes than they receive in benefits.

7. Take in more immigrants, but demand higher levels of skills and education. At the very least, take in any revenue-positive immigrant.

A thought occurs to me: We are so far in the hole with tens of bllions of unfunded liabilities that under current entitlements rules most US citizens are currently scheduled to get more in benefits than they paid in taxes in the course of their working lives. Let us leave aside for the moment the fact that this can not work and that a financial train wreck looms in our future.

What I'm wondering is just how high would a native born person's yearly average or total working life income have to be for that person to be truly revenue-positive after retiring and having Social Security and Medicare paid for them until they die? Has anyone ever tried to calculate the answer to that question? Is the average medical doctor revenue-positive? The average lawyer? (leaving aside the externalities some of them cause) The average truck driver? The average welder? You get the idea.

Granted, all sorts of assumptions have to be made. We could assume, for example, that a person was never jailed and never collected welfare. Of course, if someone was jailed for some years they'd have to make much larger sums after release to pay back the cost of jailing plus the lost taxes on wages not collected while they are in jail.

The same question can be asked about immigrants. For immigrants the picture is more complicated. One who arrives at age 18 wasn't educated at taxpayer expense for grade school and high school for example. But the later they arrive in adulthood the less total they will pay in taxes. So there is obviously an ideal age of entry and skill level and type of education.

I can not offer a model and a set of calculations for how to determine who is "Revenue Positive" and who isn't. But think about some basic numbers. One estimate for the size of the unfunded US Social Security and Medicare liabilities is over $70 trillion. Another estimate puts it at $50 trillion. But that estimate probably doesn't take full account of the new drug benefit. That is a huge unfunded liability that is in need of some solutions that will reduce its size. As Alex Tabarrok has pointed out " demographics are the problem not the solution". But could a radical change in immigration policy at least make future immigration part of the solution?

Well, here's a step in the direction of an analysis of who pays more in taxes than they get in benefits. First off, consider the $70 trillion in unfunded liabilities could be divided across about 280 million people in America. The number may he higher than 280 million. But some of those are illegals and are not yet guaranteed to be eligible for Medicare some day. So lets go with 280 is a rough cut on the permanent US population. Well, $70 trillion divided by 280 million is an astounding $250,000 per person. A cool quarter million dollars is needed additional per person to pay current unfunded liabilities.

Any immigrant needs to pay as much as the average American does in taxes over a work career plus an additional $250,000 in taxes in order to be "Revenue Neutral" - and even more in order to be "Revenue Positive". Does that make sense? How much does the average American earn over their lifetime? How much more than that would an immigrant have to earn in order to be "Revenue Positive"?

An immigrant male who brings a foreign non-working spouse with him would have to earn enough and pay enough taxes to pay for both their liabilities. So the working male would need to pay $500,000 more in lifetime taxes than the average American. If that immigrant was taxed an average of 33% on the additional income then the immigrant would have to earn $1.5 million more of lifetime income to pay his and his wife's liabilities. Over a 40 year working career that immigrant would have to have an income that averaged $37,500 more per year than the average American worker. earns. Well, we have millions of immigrants who do not even make that amount total per year, let alone that amount more per year above the average. However a working immigrant couple would have to average only $18,750 each more per year than the average American

Of course tax rates could be raised. But some people never make much money to be taxed in the first place. Even if they worked for 50 years they still couldn't come up with enough extra income to be taxed to yield an additional $5000 per year average in taxes collected. So others will have to pay even more taxes to fund the $250,000 per person in in unfunded liability.

Currently the United States has a per capita GDP of $37,800. If every person could somehow work 7 more years and turn over all the wealth they generated to the government that would be enough to pay the unfunded liabilities. Of course that isn't possible. Aging people become less productive and less able to work. Also, most people are not going to work purely to pay taxes. So delayed retirement. while it would help, would not entirely solve the problem of unfunded liabilities.

General tax rates could be raised. But some people, if taxed at a higher level, would treat the higher taxes as a disincentive to work and would opt to produce less and spend more in leisure time or by performing activities that personally benefit them without generating taxes. For instance, one could work fewer hours at a job, earn less income, and compensate for the lower income by no longer paying for some services. For instance, a person could stop paying a house cleaner and gardener and then do that work one self. This would reduce the amount of taxes collected on activities that used to be performed as a consequence of taxable market transactions. Higher tax rates will reduce the use of services and increase the frequency with which people pursue do-it-yourselfer options.

So how high would the incomes of immigrants have to be in order for those immigrants to pay more in taxes over their working lives than they receive in beneifts while working and after retirement? That is the question we need to answer if we are going to pursue an immigration policy that is a net economic benefit for the American public.

Aside: It occurs to me that it is more expensive for society to jail a high income person (at least if the person has a productive job earning an honest income in addition to whatever got them thrown in jail) than a low income person because the lost of wages and therefore taxes while the person is jailed. For example, what taxes are never collected as a result of putting Martha Stewart in jail? Will we reach a point due to the old age funding crisis that jail sentences will be changed in ways designed to reduce losses from income lost due to incarceration? Will higher income people be given sentences that require them to get out of jail part of the day to go to work and earn taxable income? In the 2010s and 2020s the United States government (like a number of European governments) is going to become desperate for revenue. I would not rule out jail work release programs for high income criminals as one response to the country's financial plight.

Share |      By Randall Parker at 2004 August 13 02:27 AM  Immigration Economics

John Brothers said at August 13, 2004 4:43 AM:

Great question. Based on your analysis, it looks like people who earn over 75k on average over their working lifetime are part of the solution. (or working couples who earn an average of 55k each). Having said that, I suspect the likelihood that well-off immigrant men will have working wives is fairly low.

Some comments:
1) You can't include currently retired folks in your calculations of the total US working population. I think you can count the kids, since they will be working over the course of the problem area. So the cost per working person goes up a fair amount.
2) Future laws may reduce these unfunded liabilities (I hope)
3) If we were to import 100 million immigrants tomorrow, each of whom earned on average $75k, we could theoretically pay off that 70 trillion in 20 years. But of course, some significant portion of that extra 3.7 trillion per year would have to go to infrastructure for those immigrants - schools, roads, government services, police, fire, sewer, etc (unless you want to count only the first 37k as covering all of those things). Hard to say how much of an impact that would have.

A Berman said at August 13, 2004 6:29 AM:

Randall, you're usually either spot-on or close to the mark, and I agree with much of what you say about immigration, but I have to disagree with your math at a fundamental level. You make a losing argument.

Key point: The contribution of work is NEVER expressed accurately in salary. Someone who earns X per year is almost certainly contributing far more than X to the US GDP.

How much does a team of $30K/year firemen contribute when they save a burning building? How much does a $40K/year physical therapist contribute when she helps someone have a few more productive years? A $6/hour janitor is needed to keep a McDonalds in business, and how much efficiency is gained by letting on-the-go people have fast food? The matrix of jobs in this country is important and it includes very low income jobs and illegal immigrants are willing to do those jobs. Maybe we shouldn't, but the fact is, simply looking at their utility according to their salary is just incorrect.

Randall Parker said at August 13, 2004 10:32 AM:


You are making a number of errors in your argument.

Contributions that are not included in salary: First of all, I'm trying to solve an economic problem. Contributions that are not financially rewardable do not generate a transaction that can be taxed to pay for the unfunded liability.

When the population increases we need more firemen. The fact that the building is there to burn down in the first place is because there are people to go into it. So the existence of the additional people creates the need that the firemen fulfill.

Similarly, if you increase the population as a whole you will increase the need for physical therapists proportionately to that increase in population. So where's the net gain?

As for the need to keep a McDonalds in business: There is no real need of that sort. If the McDonalds stores have fewer illegals to hire then there will be higher salaries for the janitors, slightly higher McDonalds prices, and slightly less demand for McDonalds food. But since the population will be smaller there will be fewer people to want to buy the McDonalds food in the first place.

You are missing the point that the janitor also will be eligible for Medicare and Social Security some day and hence his $6/hour job makes our financial situation worse.

noone said at August 13, 2004 2:51 PM:

Revenue positive immigrants would look pretty asian,for the most part.

John S Bolton said at August 13, 2004 7:04 PM:

Immigrants who are not on net public subsidy are almost certainly in the top ranks of English-language abilities. They need to connect to the high per-capita income mainstream American economy, not be stranded in a low-income ghetto economy. They are also not just somebody's relative, but more independently qualified.

John S Bolton said at August 14, 2004 1:14 AM:

Asian-nation immigrant characteristics, on the web, gives the median personal income of foreign-born from the 2000 census, as ~$15,000. This is around 30% below the native-born figure. The personal income figure is the relevant one, especially for estimating the taxes paid.

gc said at August 18, 2004 7:34 AM:

I think there are a few points missing from the analysis:

1) Revenue positive citizens reduce the amount of spending required - up to a point. Too much homogeneity and you start getting pressure for socialism a la Sweden. The best possible situation would be a small, visible minority that would disproportionately benefit from redistribution - but wouldn't have enough political clout to put in place.

2) The other assumption is that those "unfunded liabilities" are ever going to be funded. They're not. Someone is just going to get cut off.

Randall Parker said at August 18, 2004 10:03 AM:


There is no plausible scenario under which we will have to worry about too much homogeneity.

I agree that some of the unfunded liabilities will never be paid. However, we are also going to get stuck with much higher taxes. Of course, this will make the economy grow more slowly and therefore the tax revenues will not grow as much as the tax rates.

John S Bolton said at September 29, 2004 9:29 PM:

The US Census 2000, 'population profile of the US', gives the 'average income per household member by race' and hispanic origin, which is one way to look at the economic effects of immigration. The hispanic figure was less than half the non-hispanic white figure, 12k vs 25k. This disparity is like that between blacks and whites in the 1930's, yet for some reason, it is little spoken of. Is this because it makes our immigration policy look like an abysmal failure? The asian figure was also several thousand dollars below the non-hispanic white average, which doesn't quite fit the model-minority picture which has been given. Revenue-positive immigrants are likely to be quite the rare bird, and it would take very strict standards of discrimination to recruit only, or mostly, them, from abroad.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright