assess the impact of migration on economic development is whether remittances are used to
finance consumption or productive investments10; yet, the latter analysis is not an easy task
either, because of fungibility issue, externalities and lack of reliable measures of (net)
remittances11. Lastly, remittances are endogenous to the migration selection and typically
vary directly with the cost, risk and social dislocation associated with the move (i.e. with the
form of migration). Therefore, the economic effects of remittances are intimately tied to
migration determinants and cannot be evaluated independently of them. Furthermore, the
insurance provision provided by family members working in different labour markets may
influence sending farm households (and their productive capacity) even without remittances
(Stark 1980).
Thus, household behaviour towards migration is crucial in shaping the economic effect of this
process, which may act as a shelter against income and production risks faced by people left
behind. There is considerable evidence in the development literature on the widespread
diversification of farm household income sources as a way to manage risk in developing
countries (see, for example, Morduch, 1995) 12. There is also growing evidence, though, that
entry constraints may limit the usefulness of income diversification strategies. This is to say
that risk-management strategies may imply an efficiency loss for the poor, which the rich -
typically better protected via assets and institutional arrangements - do not have to endure
(see Dercon 2002).
Thus, given incomplete insurance and segmented capital markets, structural characteristics
and wealth of households typically shape liquidity and risk constraints, thereby influencing
the incentive to move and the shadow value of remittances. Moreover, when a farm household
decides to send out a migrant, this has simultaneous implications on its productive capacity
and may modify productivity-enhancement choices, such as a change of agricultural
technology. Farm household decides about its present labour and other inputs allocations, on
10 One perspective is that remittances tend to be used for conspicuous consumption rather than investment: for
house construction or the sponsoring or weddings and the like, rather than improvements that are likely to lead to
increasing agricultural productivity. A common use of remittances, nevertheless, is also to pay for education of
the next generation and that does appear to be a clear investment strategy. However, a clear distinction between
investment and consumption may be difficult to maintain in the context of the use of remittances. There is an
important indirect effect of remittances money in the villages (expenditure on house construction for example
can stimulate local building enterprises etc.).
11 Remittances are notoriously difficult to measure (money and goods) because of official and unofficial
channels (through relatives or when they go back). Moreover, a typical assumption of much of the work on
remittances is that migrants are all self-supporting, that is no economic support is given to migrants leaving
household of origin. Lipton (1980) argues that net remittances are quite small relative to village income (they
are much more concentrated on richer households in the village unlikely to suffer from capital constraints) and
have a negative economic effects at home. Taylor (1992) and Adams (1991), on the other hand, show a more
positive scenario.
12 See Mendola (2004) for a literature review on this.