People who buy annuities to finance their retirement are as a group substantially longer-lived than
typical individuals in the UK population. And voluntary annuitants are even longer-lived than
compulsory annuitants. For example, an average 65-year-old man has a 41% change of surviving
until the age of 82. For the average 65-year-old male compulsory annuitant, this probability rises
to 48%, and for the average 65-year-old male voluntary annuitant, the probability rises even
further to 56%.
These research findings by Amy Finkelstein and James Poterba, published in the latest issue of
the Economic Journal, suggest that there is 'adverse selection' in the market for annuities. People
who expect to live longer are more likely to buy them, encouraging insurance companies to raise
their prices and making them even less attractive to people who do not expect to live so long.
What's more, because of the even stronger selection effects with voluntary annuities, they are
more expensive than compulsory annuities. These comparisons have important implications for
the policy debate about the desirability of compulsory annuitisation requirements at a time when
interest rates are low and pensions are shifting to defined contribution systems.
The ageing population in the UK and elsewhere has generated substantial interest in the way
households finance consumption in their retirement years. Annuities - insurance products that
provide a payment stream for as long as the insured individual is alive - offer one way to spread an
accumulated stock of resources over a retirement period of uncertain length. They offer the
potential of providing valuable insurance against the possibility of outliving one's resources and
ending up impoverished in old age. Despite this great promise, a concern for policy-makers has
been the tiny size of the voluntary annuity markets in many nations.
One explanation for the small size of these markets is the possibility of adverse selection. If
individuals have information that the insurance company does not have about how long they are
likely to live, then those who are likely to live a long time and receive annuity payments for many
years may be more likely to buy annuities. If they do, while shorter-lived individuals do not
participate in the market, insurance companies will respond by raising annuity prices. This may
ultimately make them unattractive to some individuals who do not expect to live longer than the
average person of their age.
Finkelstein and Poterba present new evidence on the extent and nature of adverse selection in the
UK annuity market, paying particular attention to the differences in the extent of this selection in
the voluntary and compulsory annuity markets. Individuals who save for retirement through taxdeferred
savings vehicles such as occupational pensions or personal pensions face some
compulsory annuitisation requirements of their accumulated balances. Individuals with additional
non-pension savings may choose to annuitise part of their wealth in the voluntary market.
The researchers consider two types of evidence of the degree of adverse selection:
l First, they compare the mortality experience of individuals in different annuity markets.
The fact that annuitants are longer-lived that the typical individual is consistent with
individuals having private information about their longevity and selecting into the market
accordingly. And the fact that voluntary annuitants are even longer-lived than compulsory
annuitants is what would be expected if the scope for selection were reduced by
compulsory annuitisation requirements.
l Second, they compare the pricing of voluntary and compulsory annuities. Not surprisingly
given the mortality differences, voluntary annuities are priced higher than compulsory
annuities. Indeed, on average, the amount of adverse selection in the compulsory market -
as measured by the increase in pricing associated with the selection - is about one-half of
that in the voluntary market.
These comparisons of the two markets bear directly on policy discussions about the desirability of
compulsory annuitisation requirements to combat the limited size of voluntary annuity markets.
The decline in long-term interest rates in the UK since the early 1990s has coincided with a
decline in annuity payouts relative to annuity premia, leading some to question the wisdom of
compulsory annuitisation in defined contribution pension programmes. But compulsory
annuitisation reduces the degree of adverse selection: this should be weighed against the possible
costs of such a policy.
More generally, annuity markets have attracted attention as part of a global debate on social
security reform. There are proposals in many nations to replace or supplement defined benefit
social security programmes with defined contribution systems in which individuals would
accumulate assets in individual accounts. In such systems, it is not clear how individuals would
draw down their asset balances in retirement. Some proposals call for compulsory annuitisation at
retirement. Others would allow individuals to draw down their account balances in more flexible
ways, either by choosing to purchase annuities from private insurance firms or by taking lumpsum
distributions. The relative attractiveness of these various options depends critically on
whether reasonably priced individual annuities are available in the private annuity market.
Notes for Editors: 'Selection Effects in the United Kingdom Individual Annuities Market' by
Amy Finkelstein and James Poterba is published in the January 2002 issue of the Economic
Journal. Finkelstein is at the National Bureau of Economic Research (NBER) in Cambridge,
Massachusetts; Poterba is at the Massachusetts Institute of Technology (MIT) and NBER.
For Further Information: contact Amy Finkelstein on +1-617-588-0317 (fax: +1-617-868-2742;
email: afinkels@nber.org); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or
07768-661095 (email: romesh@compuserve.com).