You buy homeowners insurance in order to shift money from financially non-scary scenarios (e.g., those in which your house doesn’t burn down) to financially scary scenarios (e.g, those in which your house does burn down).
And you buy life insurance in order to shift money from financially non-scary scenarios (in which you live long enough to enjoy a normal-length career) to financially scary scenarios (in which you die at a young age, leaving your dependents with insufficient sources of income).
And the same thing goes for several other similar decisions: buying health insurance, buying auto insurance beyond that mandated by law, including low-risk asset classes in your portfolio, etc.
A key point about each of these decisions is that they actually worsen your average outcome. That is, on average, you will collect less money from an insurance policy than you pay in premiums (because insurance companies have costs as well as profit margins). Yet such decisions are generally regarded as prudent, because they protect you from a significant financial risk.
What This Has to Do With Social Security
Delaying Social Security is a choice that shifts money from financially non-scary scenarios (i.e., short retirement due to early death) to financially scary scenarios (i.e., long retirement due to living longer than average). What’s so compelling about this decision, however, is that in most cases it doesn’t actually worsen the average outcome. In fact, for more than half of retirees (namely, most people other than lower-earning spouses in married couples) the decision to delay actually improves the average outcome.
To state that again: For most retirees, delaying Social Security simultaneously reduces risk and improves the average outcome (i.e., the amount of dollars you can spend over your lifetime). That type of opportunity is very rare in finance. And that’s why you see so many experts saying that it’s a good idea to hold off on taking benefits, if you have the ability to do so.
Point of explanation #1: Delaying Social Security improves the average outcome for unmarried people because the system includes a built-in interest rate that exceeds current market interest rates. That is, in order for taking benefits at 62 to be as good on average as taking benefits at 70, an unmarried person would have to be able to invest his/her early-received benefits at a roughly 2% real interest rate — which you cannot safely do right now, with interest rates as low as they are.
Point of explanation #2: When the higher-earning spouse in a married couple delays claiming Social Security, it dramatically improves the average outcome because (in addition to the above point about interest rates) the increased monthly payout that comes from delaying will be applied for as long as either spouse is still alive.