Wednesday, June 3, 2009

The Social Security Statement: New and Improved

The Social Security Administration has made some improvements to the Social Security Statement, the annual benefit estimate sent to every worker each year.

The first change is a mention on page 1 of SSA's new When to Start Receiving Benefits pamphlet, which I think is a much improved explanation of the pros and cons of delaying benefit claiming. In the past I'd thought there was a bias in SSA's procedures toward encouraging people to claim at 62. The new material is much more balanced and I think very informative.

Second, in a bit of behavioral economics at work, the benefit estimates on page 2 now begin with benefits as of the Full Retirement Age rather than age 62, the age of earliest eligibility. People often gravitate toward the first thing they read, so it makes sense to make that the Full Retirement Age rather than earliest one, at which benefits can be cut by 25 percent for life.

Click here to see a sample statement.

Congrats to all at SSA who worked on this – a job well done.

5 comments:

Unknown said...

The new material is much more balanced and I think very informative.

-------------

Andrew,

The problem with it (still) is that most people - I suspect the great majority of people - won't know what to make of the difference in payouts. Waiting until you're 70 makes more economic sense the longer you expect to live (which might favor women, incidentally). However, if you're in ill-health or have lower life expectancy (because, for example, you've been a smoker all your life), taking the money sooner make more cents (sorry, couldn't help myself!). Other factors include what you're doing with the money - investing it or spending it as the checks arrive.

SS should provide some calculators to help people understand the difference in payouts based on different possible scenarios.

Andrew G. Biggs said...

James,

SSA used to provide these kind of calculators but no longer does, and I must admit I'm probably the main reason it doesn't. For years SSA has pushed the "break even" age approach, which says how many years you'd have to live to "break even" by delaying benefits from, say, 62 to 65.

But this is the wrong approach. The real value from Social Security benefits is longevity insurance, that is, insurance against outliving your assets. This has VERY big welfare gains. The break even approach treats it as a gamble, which is not what the real issue is.

So I understand your concerns, but think it's better to think about delaying retirement in terms of "buying" more insurance against outliving your assets. Even someone whose life expectancy is below average would benefit from this, since even that kind of person (say, an African American male) still faces a lot of uncertainty regarding when he'll die.

James said...

Andrew,

Have studies been conducted that look at the fiscal viability of SS - impact on yearly surpluses and long-term liabilities - under various scenarios regarding when people choose to begin receiving payments using various distributional models. On the extremes would be everyone at 62 or everyone at 70. In particular, are there models under which the progam's long-term fiscal problems are meaningfully improved?

Andrew G. Biggs said...

In general it doesn't matter to the system when people claim. The benefit formula that adjusts for early/delayed claiming is just about actuarially neutral, meaning that the present value of expected lifetime benefits is about the same no matter when you claim.

That said, retirement isn't the same as claiming. If people choose to work longer and then claim benefits, Social Security's finances will improve a bit. The reason is that the extra taxes paid into the system don't produce comparable levels of benefits (technically speaking, the marginal rate of return is below the interest rate on the trust funds). I don't know how big this effect would be, but I could check it out.

James said...

I don't know how big this effect would be, but I could check it out.

----------

I'd be very interested in seeing this data, if possible. Thanks so much.