PensionReforms' summary and comments
When savers reach retirement, they should logically start spending their savings to replace lost income from employment.  This 'decumulation' period replaces the 'accumulation' phase of their working lives.

Those with private Defined Benefit pensions don't need to worry.  The pension starts and effectively runs down that piece of retirement wealth over the pensioner's remaining life.  If Defined Contribution savings have a compulsory pension purchase, again retirees have little choice other than with regard to the terms of the pension (guarantee period, automatic increases, survivor benefits).  In either case, the decumulation decision is taken care of.  State-provided pensions perform a similar function.

Where retirees have a lump sum or can draw their retirement savings out in cash, they have decisions to make - how and where to invest it; how much to keep accessible; how much of the capital component to spend to top up investment income.  Those decisions need to be made each year of retirement otherwise one of two things might happen - the retiree will run out of money or will die leaving an unintended bequest.  Either possibility is relatively inefficient.  Perhaps, they should use some or all of that money to buy an annuity but voluntary annuities are very unpopular in the US.

The price of an annuity reflects a number of components - the age and health of the buyer (that will drive the years it is expected to be paid); the specific terms (-guarantee period? -continuing after death to a dependent?  -increases during payment?); expenses and profit to the provider; investment returns on the invested assets.  The annual amount could be payable for decades so, for a given purchase price, is very sensitive to the guesses made by the provider for each of the components making up the price.

The provider, like any insurer, does not expect to make a profit on every sale but to make a profit on average across all annuities sold.  That depends on the provider's getting a cross-section of mortality risks into its annuity portfolio, keeping expenses down and earning at least the assumed returns on investments (there is then the re-investment risk).  Providers of voluntary annuities run the risk of adverse selection.  Those who think they might live longer than average probably like annuities more than those who expect a shorter retirement who should prefer to invest their assets directly. 

"Many explanations have been offered for retired households' reluctance to annuitize.  Prominent is that annuities suffer from a considerable degree of actuarial unfairness.  That is, for the average household, the expected value of the income, discounted by a rate of interest and annual survival probabilities, is considerably less than the premium paid.  But it seems likely that households are also influenced by a possibly not wholly rational reluctance to give up access to their life savings." [PensionReforms notes there is more to the "actuarial unfairness issue than this.]

The report suggests that this unpopularity derives from the unattractive design of current annuity products so looks at an alternative approach:
"This paper evaluates a proposal, first brought to the attention of the academic community by [Moshe] Milevsky, for an innovative annuity product - the Advanced Life Deferred Annuity (ALDA).  The ALDA is an inflation protected annuity that would be purchased at retirement or even earlier.  But in contrast to a traditional annuity, income payments would only start at some advanced age, (say) 75, 80, or 90."

The report suggests that "a household planning to smooth consumption through its retirement would need to allocate only 15 percent of its age 60 wealth to an ALDA with payments commencing at age 85, holding the remainder of its wealth in unannuitized form to finance consumption from age 60 to 85."

PensionReforms notes that the effect of having a 'deferred annuity' is that the suggested 15% purchase price is more insurance against longer than expected life than an investment for retirement income.

"The ALDA's attractiveness is that it provides a lot of longevity insurance at a relatively low cost.  It also makes decumulation much simpler during the period before the ALDA payments commence."  That's because it makes the length of that period more certain.

So will the relatively small amount involved with an ALDA overcome retirees' perceived aversion to buying annuities?  The report doesn't suggest an answer but anticipates a problem that needs a solution:
"One possible solution might be to make the purchase of an ALDA the default in 401(k) plans."

By "default", the report actually means "compulsory".  In other words, there would be no choice about the suggested 15% of 401(k) proceeds.  But what of the obvious case where the saver will die before the deferral period ends (perhaps because of retirement with a terminal illness)?

"Our calculations indicate that, even when evaluated in money's worth terms, defaulting high mortality households into ALDAs would cause relatively little harm.  This is because although ALDAs would have a low money's worth in the hands of such households, they would only invest a small proportion of their total wealth in them.  But in expected utility terms, even high mortality households might be better off, indicating that concerns that the ALDA might be inappropriate for high mortality households may be misplaced."

PensionReforms thinks there is more wrong with voluntary annuities than the report notes.  Annuities are expensive to sell and buyers have to pay for the costs involved.  They are very long-term commitments for the provider that has to allow a considerable margin for unknown future events, such as the distinct probability of regulatory changes, including changes to accounting standards.  Again, the buyer must effectively pay each year from the outset for those uncertainties through lower annual payments.  Then there is the credit risk - how does the retiree know that the provider will even be there to meet its commitments when they start in 15-20 years?  That is a matter of particular moment in the current environment.

So, PensionReforms wonders how the suggested ALDA addresses these concerns.  There is less at stake but with a very long period before the retiree gets a return on the investment, PensionReforms suspects that, without some element of compulsion, an ALDA is probably unlikely to be popular.  That's probably why the report suggests there should be no choice.

From a public policy perspective, the government can justify imposing the suggested ALDA purchase.  Taxpayers have provided large subsidies towards the cost of the retiree's 401(k) final balance and there are significant potential welfare advantages to the state in having income payable to retirees at the oldest ages.  That doesn't, however, mean that an ALDA is necessarily in every individual's best interests.  The fact that a relatively small proportion of a retiree's savings are at stake really means that a compulsory ALDA will lose only a relatively small amount of money for many.

Annuity providers will support the idea.  (File size 191 KB;  28 pp) 328