Practicing actuarial analysis without a license

Neal Boortz comments on Social Security, making a common error by using the life expectancy at birth to estimate how many years of Social Security someone will receive.

Explanation: Divide the people who have been paying Social Security taxes for the past 40 years into four basic groups. White Males, black males, white females, black females. Statistics will show that of these four demographic groups white females have the longest life expectancy; black males the shortest. In 2002 the life expectancy for a newborn white female was 79.9 years. The life expectancy of a newborn black male was 68.8 years. If you were born after 1960 your Social Security full retirement age is 67. This means that a black male can expect to get Social Security benefits for about two years, while a white female can expect to receive those benefits for almost 13 years. You do the math. even if you went to a government school you can figure out that the average white female will receive Social Security benefits 11 years longer than the average black male. It has been estimated that during his lifetime the average black male will lose about $10,000 in income that will be forcibly transferred to a white woman.

I can't comment on how much money on average is transferred from one group to another, but the average black man does not get 1.8 years of Social Security benefits just because the average life expectancy at birth is 68.8 years.

Consider two groups of people: In the first, all live to age 70. In the second, half live to age 100, half to age 40. Both have an average life expectancy at birth of 70. However, in the first, all receive 3 years of Social Security benefits. In the second, half receive 33 years of benefits, half receive none. The second group receives a lot more Social Security benefits.

The proper analysis is to take the group of people who enter the workforce, take the percentage of that group that lives to retirement age, then calculate the life expectancy from retirement age. The multiplication of the two is the average number of years of retirement benefits received.

But at least Neal is in good company. William F. Buckley, Jr. made the same mistake in a December 2, 1988 op-ed piece, according to an old article by former Social Security actuary Robert Myers.