Chapter 5


* Regular (infrastructural) projects, designed to create or enhance productive assets by
 tapping available labour while providing employment opportunities for poor households;

* Long-term employment-generation projects, designed to tackle chronic un- and under-
 employment by offering continuous job opportunities, particularly to the urban poor and
 the landless.

Compared to other forms of targeted assistance (e.g. income, cash or food transfers), such
programmes have two additional advantages:

* The assets created through the economic activities of the people employed, e.g. rural
 roads, dams, land conservation, etc.;

* The programme can be self-targeting in the sense that, if properly designed. it will attract
 only those people who have no alternative source of income and employment.

Targeting is, however, only effective if the wages for the people employed in the programme
are below market wages. Otherwise, public work schemes may attract people other than the
poor and unemployed, displacing private employment.

Box 5.3: The Arithmetic of Cost-Effectiveness in Public Works

A simple example shows how sensitive the cost-effectiveness of public works programs is to the wage and type
of programme. Assume that a programme begins in a region in which 1,000 households subsist on an average
of $ a day and other households earn in excess of $ 2 a day. The objective is to raise household income of the
first group to at least $ 2 a day. Further assume that working in the programme costs participants 50 cents a day
out-of-pocket for added food, travel to and from work, and so forth. A wage of $ 2.50 would then be enough to
attract the target population and keep away other workers who earn more than $ 2 a day. If the value of a day's
work performed under the programme is $ 2.50 and it costs $ 3.50 for wages and materials, the public cost of
augmenting incomes is $ 1. For each dollar spent on the programme, the target population's income would
increase by $ 1. If the value of a day's work were $ 3, each dollar spent on augmenting income would increase
the target population's income by $ 2. Clearly, this programme would be cost-effective.

Public works programs rarely achieve this cost-effectiveness. In many cases they create assets of low value. In
the example above, if the value of a day's work were $ 1 rather than $ 2.50, it would cost the state $ 2.50 for
each dollar of income transferred to the target population. If the daily wage offered were $ 3.50 rather than $
2.50, the programme would attract many workers from outside the target population. It might then cost the state
as much as $ 5 to $ 10 to augment the income of the people in the target population by $ I a day.

Source: World Bank, Poverty and Hunger, Washington D.C. 1986

For an evaluation of public works programmes, the dual objectives of providing targeted
support to vulnerable groups and of generating investments and development assets need to
be taken into account. Therefore, in assessing the cost-effectiveness of such programmes,
two factors have to be considered (see also Box 5.3):

1) The net income (in cash or kind) conveyed to the target group. This is the difference
 between wages paid and the additional costs accruing to the participants (e.g. costs for
 travelling to the work site, for staying away from home, opportunity cost in terms of
 income forgone from other activities such as reduced food production at home. etc.).


- 204 -