Political parties that emphasise the role of families in society seem also to prefer lower
spending on the welfare state, whereas political preferences for high welfare spending
seem to go together with weaker concerns for ‘strengthening’ family ties. New research
by Rafael Di Tella and Robert MacCulloch, published in the latest issue of the
Economic Journal, suggests an explanation: these are more efficient combinations in
the sense that, for example, proposals for a generous welfare state and measures
designed to strengthen families would be offering individuals too much insurance
against adverse economic circumstances, notably unemployment.
The researchers note that a great deal of economic research has analysed the problem
of the optimal level of provision of unemployment benefit and the impact of benefits on
the unemployment rate. But an important factor has been largely ignored: the potential
of families as providers of insurance against unemployment. Yet in a world without
government, families may provide much of the ‘social insurance’ available to people.
This raises several questions: if family members are bound together by insurance
provision, will a more generous welfare state increase total insurance available to
people or will it ‘crowd out’ insurance provided by families? Could the state make things
worse by destroying informal insurance to such an extent that total insurance falls? And
what are the implications for the optimal size of the welfare state?
The researchers attempt to answer these questions. They present analysis designed to
capture the idea that the state has an advantage vis-à-vis the family in the provision of
insurance because it can tax individuals, whereas the family must rely on self-enforcing
agreements. In such a case, the effect of state transfers on transfers within the family is
more than one-for-one. Thus, somewhat perversely, both informal transfers and total
insurance transfers to the unemployed fall as the state's generosity increases.
This does not imply that the optimal size of the welfare state is zero. The results still
hold when families are assumed to be better than the state at monitoring the job search
activities of the unemployed.
The research also provides a rationale for the birth of the welfare state. A reduced role
for religion, as well as a more tolerant view of divorce and wider availability of birth
control methods may have all affected the temptation to ‘defect’ from the family. As the
benefit level that can be informally enforced by the family falls, it may become optimal
for the state to intervene and take over responsibility for social insurance provision from
the family. But if strong family ties exist naturally, the state's best response is to keep
these ties (with the associated gains from monitoring of individual behaviour by peers)
and opt out of welfare provision altogether.
On the other hand, the case of single mothers may call for different action. This case
introduces the possibility that the cost of risk involves an indivisibility. If a pregnancy
occurs, the mother may require a high minimum transfer (below which transfers are not
valued by the recipient). It is possible that the amount of money/help the mother will
need (in terms of lodging, food, etc.) is so high that becoming a single mother is not an
insurable risk for the family. But if the state provides some help to single mothers, it can
make the risk insurable. Just below that crucial level, increases in the generosity of the
welfare state can increase informal transfers within the family.
ENDS
Notes for Editors: 'Informal Family Insurance and the Design of the Welfare State’ by
Rafael Di Tella and Robert MacCulloch is published in the July 2002 issue of the
Economic Journal.
Di Tella is at Harvard Business School; MacCulloch is at the London School of
Economics.
For Further Information: contact Robert MacCulloch on 020-7955-6960 (email:
r.j.macculloch@lse.ac.uk ); or RES Media Consultant Romesh Vaitilingam on 0117-983-
9770 or 07768-661095 (email: romesh@compuserve.com).