I find myself thinking more these days about the fairness of Social Security and other government retirement benefits.

One reason, of course, is because I’m getting close to retirement age—and, as I discover each time I raise the issue with students, young people don’t think about it much.* Another reason is because Social Security (in addition to Medicare, Disability, and other programs) is the way the United States creates a collective bond between current and former workers, by using a portion of the surplus produced by current workers to provide a safety net for workers who have retired.

That represents a kind of social fairness—that people who have spent a large portion of their lives working (most people need 40 credits, based on years of work and earnings, to qualify for full Social Security benefits) are eligible for government retirement benefits provided by current workers. Another aspect of that fairness is the system should and does redistribute from those with high lifetime incomes to those with lower lifetime incomes. While that makes the actual “rates of return” unequal across groups, it’s designed to provide a floor for the poorest workers in society.

Many people consider the U.S. Social Security system fair on those two grounds. That’s true even though some people, by random draw, may live longer than others. However, as Alan J. Auerbach et al. (pdf [ht: lw]) report, that fairness may be put into question if there are identifiable groups that vary in life expectancy, “as this introduces a non-random aspect to the inequality.”

Here’s the problem: retirement benefits in the United States are increasingly unequally distributed on a non-random basis. As I’ve written about many different times (e.g., here, here, and here), there’s a gap in life expectancies between those at the bottom and top of the distribution of income. And the gap has been growing over time.


That result is confirmed by Alan J. Auerbach et al.: for the male birth cohort of 1930, life expectancy at age 50 rises from 26.6 to 31.7—a difference of 5.1 years. For the 1960 cohort, the lowest quintile has a slightly lower life expectancy than the 1930 cohort but then rises a level of 12.7 years higher for the top quintile, “indicating a very large increase in the dispersion.”


Not surprisingly (since benefits rise with earnings), Social Security benefits also rise with income quintiles. Thus, for example, for men in the 1930 cohort, workers in the lowest quintile can expect to receive, on average, $126 thousand in benefits over the rest of their lives (discounted to age 50), while workers in the top quintile can expect to receive $229 thousand, or 82 percent more than the lowest income workers.

What is particularly troubling is how the results change when we move to the 1960 cohort. The additional 6-8 years of life expectancy for the top three quintiles lead to large increases in expected Social Security benefits, with benefits for the top quintile reaching $295 thousand. The difference between the highest and lowest quintiles is then expected to be $173 thousand, or 142 percent of the lowest income workers’ benefit.

According to the authors of the study,

These results suggest that Social Security is becoming significantly less progressive over time due to the widening gap in life expectancy.

Not only does the growing gap in life expectancies undermine the basic fairness of the Social Security system. It calls into question capitalism itself.


*For understandable reasons. I certainly didn’t think about retirement at that age. (I barely thought about getting a job. I just presumed I would—and would be able to—at some point.) However, when students are induced to do think about retirement, as I’ve written before, most take it for granted that Social Security is doomed. While they expect to pay into Social Security, they don’t expect to receive any Social Security benefits when they retire. Then, of course, I explain to them that making only one change—raising the taxable earnings base—would eliminate the projected deficit and keep Social Security solvent forever.


Special mention



Special mention

194500_600 194606_600


Today, thousands of scientists, science educators, and others are protesting in hundreds of cities around the world against what they see as a global political assault on science.

According to Lucky Tran, who serves on the steering committee of the March for Science,

We need another scientific revolution. The first scientific revolution transformed our understanding the natural world and changed civilization as we know it, making us healthier, happier and more productive. Today, we have huge scientific challenges ahead such as climate change, genetic engineering and automation. And yet we are stuck with the hand brake on and unable to act effectively as a society, because a few privileged people are blocking progress at the expense of the rest of us.

Tran might also have invoked Albert Einstein, the most famous scientist of the twentieth century, who in 1949 expressed his support for socialism in the name of social values but with a perhaps more humble view of the role of science.

we should be on our guard not to overestimate science and scientific methods when it is a question of human problems; and we should not assume that experts are the only ones who have a right to express themselves on questions affecting the organization of society.

And the problem?

Production is carried on for profit, not for use. There is no provision that all those able and willing to work will always be in a position to find employment; an “army of unemployed” almost always exists. The worker is constantly in fear of losing his job. Since unemployed and poorly paid workers do not provide a profitable market, the production of consumers’ goods is restricted, and great hardship is the consequence. Technological progress frequently results in more unemployment rather than in an easing of the burden of work for all. The profit motive, in conjunction with competition among capitalists, is responsible for an instability in the accumulation and utilization of capital which leads to increasingly severe depressions. Unlimited competition leads to a huge waste of labor, and to that crippling of the social consciousness of individuals which I mentioned before.

This crippling of individuals I consider the worst evil of capitalism. Our whole educational system suffers from this evil. An exaggerated competitive attitude is inculcated into the student, who is trained to worship acquisitive success as a preparation for his future career.

To which Einstein offered a solution:

I am convinced there is only one way to eliminate these grave evils, namely through the establishment of a socialist economy, accompanied by an educational system which would be oriented toward social goals. In such an economy, the means of production are owned by society itself and are utilized in a planned fashion. A planned economy, which adjusts production to the needs of the community, would distribute the work to be done among all those able to work and would guarantee a livelihood to every man, woman, and child. The education of the individual, in addition to promoting his own innate abilities, would attempt to develop in him a sense of responsibility for his fellow men in place of the glorification of power and success in our present society.

I wonder if today, in addition to defending science, scientists are willing to criticize the existing economic and social institutions and participate in the project of imagining and creating an alternative.


Special mention

194576_600 download (1)


On the eve of their presidential election, the French people and politicians continue to debate how they should respond to the end of “Les Trente Glorieuses,” a period that appears to receding into ancient history.

Except, as it turns out, for those at the very top, for whom the last thirty years have been quite glorious.

According to new research by Bertrand Garbinti, Jonathan Goupille-Lebret, and Thomas Piketty, between 1983 and 2014, average per adult national income rose by 35 percent in real terms in France. However, actual cumulated growth was not the same for all income groups:

the growth incidence curve is characterized by an impressive upward-sloping part at the top. Cumulated growth between 1983 and 2014 was 31% on average for the bottom 50% of the distribution, 27% for next 40%, and 50% for the top 10%. Most importantly, cumulated growth remains below average until the 95th percentile, and then rises steeply, up to as much as 100% for the top 1% and 150% for the top 0,01%.

The contrast with the earlier, 1950-1983 period is particularly striking. In effect, during the “Thirty Glorious Years,” Garbinti, Goupille-Lebret, and Piketty observe the exact opposite pattern: growth rates were very high for the bottom 95 percent of the population (about 3.5 percent per year) and fell abruptly above the 95th percentile (1.5 percent at the very top). However, as is clear from the chart above, between 1983 and 2014, growth rates were very modest for the bottom 95 percent of the population (about 1 percent per year) and rose sharply above the 95th percentile (3 percent at the very top).

As we know, similar patterns hold for the United Kingdom (which voted for Brexit) and the United States (which elected Donald Trump).

The key question in France, in the first and second rounds of the presidential election, is how French voters will respond to a political economy that has generated thirty glorious years only for those at the very top.


Special mention

194539_600 194504_600