Q. The flat, unqualified statement in your recent column— "Receiving (Social Security) benefits before 66 just doesn't make sense"— brought me out of my chair.

I'm sure there are many reasons people take benefits early. But I suggest the major reason people take them early is a kind of "insurance.” A worker at age 66 facing a decision about two different income streams may not understand the "crossover" calculation versus life expectancy. Or he may not appreciate the impact of an assumed earnings rate for money in hand sooner.

What they do appreciate is that no matter which alternative they choose, they have roughly a 50/50 chance of being right. Roughly half the people that take your advice and wait until 66 will die before the "crossover point." So half will have made the wrong decision. Faced with a 50/50 chance most people will go with the "bird in the hand" choice. This is probably encouraged by an innate distrust of the government.  So there’s also a kind of "get mine now while it's there" logic.

I still think the average Joe in average situations should take benefits as early as available. If they "live too long," any "loss" should be treated as an "insurance premium" cost that protected them as they aged. —B.M., by email

A. The only problem with that 50/50 argument is that the odds are incorrect. Let me explain why. In the case I have written the most about— that of a married couple where the husband is older than his spouse and has the stronger work record (demographically typical)— the husband who defers is betting on his life expectancy and the life expectancy of his spouse. The joint life expectancy of a couple in their mid-sixties is about 25 years. So there is a 50/50 chance that one of them will be collecting benefits for twice as long as the 12.5 year break-even period.

The other cases are less dominant. But they still make some amount of deferral a good bet. The US Life Tables tell us that the life expectancy of all Americans at age 62 is 20.0 years. At age 66 it is 17.8 years and at 70 it is 14.9 years. So even if you defer to age 70, deferral is a much better bet than 50/50.

If we get more precise and identify by race and gender those expectancy figures shift some. They shift up for both white and black women. They shift down for white men. Black men have the lowest life expectancy; at age 70 it is 12.3 years— a bit less than a 50/50 bet.

If we get down into the actuarial weeds, we find that life expectancies are still higher for some people. For instance, people who have more education and more income— like newspaper readers. This would improve the bet a bit more.

And, finally, there is the simple investment case. Millions of seniors are earning nothing on their savings. But an easy option will provide a far better return than anything available in the marketplace.        Suppose a worker is to receive a $1,000 per month benefit at age 62. If the worker spends $1,000 a month from savings and defers taking benefits for a year he or she has “invested” $12,000 in a life annuity from Social Security— the extra income for deferral.

If they take benefits after those 12 months, their benefit will be about 7.1 percent higher. (The increase is 8 percent a year after reaching full retirement age of 66, but less before.). So they would be getting $71 a month more income for their $12,000 of deferred benefits.

That’s $852 a year more income.  And it would receive inflation adjustments upward for the rest of their lives. So you’ve got a life annuity payout at 7.1 percent with a lifetime inflation kicker. It’s also likely that this income would not be taxable— unless the retiree had a fair amount of income from other sources.

At www.immediateannuities.com I found that a 62-year-old male would receive a lifetime income of only $62 a month on a $12,000 investment. That amount would be constant, not inflation adjusted, and some of it would be taxable.

The bottom line: People soon to retire should look to Social Security first if they are seeking to safely increase retirement income.