13s(P
Human Resources Development and Operations Policy
AHRO
Working Papers
Labor Market Insurance
and Social Safety Nets
Deepak Lal
October 1994
HROWP 41



Papers in this series are not formal publications of the World Bank. They present prcliminary and unpolished results of analysis
that are circulated to encourage discussion and comment; citation and the use of such a paper should take account of its
provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s)
and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of
Executive Directors or the countries they represent.



Labor Market Insurance
and Social Safety Nets
by
Deepak Lal



I



Abstract
This paper makes a distinction between social safety nets and welfare states and argues
that while the former are needed because of the ubiquitousness of risk in human life, the case
for the latter is more controversial. It emphasizes that "social" refers to cooperation, and does
not prejudge whether this should be provided through private or public action. It provides a
typology by stage of development of the type of labor market risks faced by workers in different
types of developing countries. It then goes on to examine the relative efficacy of private versus
public transfers in dealing with these risks. The importance of private inter-household transfers
in alleviating destitution, conjunctural poverty and providing the merit goods of education and
health are emphasized. While the empirical evidence based largely on World Bank sources is
used to show that because of reasons concemed both with problems of information and political
economy, the private transfer route may be preferable to the public. Various principles for the
design of social safety nets which are financially and politically viable are then derived and used
to provide broad guidelines for the provision and financing of various components of the social
safety net: merit goods, direct transfers to deal with destitution and conjunctural poverty,
unemployment insurance, severance payments and pensions.



I



Contents
Introduction  ...................................                                                 1
I. Labor Market Risks in the Process of Development  ........ .               .  .  .  .  .  .  .  .  .  .  .    .   1
Safety Nets and Welfare States  .1...................  .....  .....                        .  I
Labor Market Risks  .....................................  2
A Typology of Risks and Stages of Development  ........  .             .  ..  .  .  .  .  .  .  .    .  4
II. Private Versus Public Transfers in Social Safety Nets .5...... .         .  .  .  .  .  .  .  .  .  .  .    .  S
Private Transfers .5......................................   .  S
Public Transfers  .......................................   .  8
Political Economy of Transfer States ........................... . 10
III. Classical Liberal Principles and Labor Market Insurance  .......  .          .  .  .  .  .  .  .  .    . 13
IV. Guidelines for Public Action on Social Safety Nets .................... . 14
Bibliography                  ............................ . 19



I



Introduction
The purpose of this paper is to provide operational guidelines on the design of social safety
nets that are fiscally and politically sustainable. Various components of proposed safety nets - the
provision of merit goods, direct transfers to alleviate poverty, insurable labor market risks,
unemployment benefits and severance payments are distinguished - are discussed.
To anchor the discussion, it is necessary to briefly outline the difference between a 'welfare
state' and a 'social safety net' and to provide a typology by stage of development of the type of
labor market risks faced by workers in different developing countries which requires some collective
action. This is done in section I. Even if a social safety net embodying such collective action is
required, there is the open question whether this should be left to individuals to organize through
private institutions or should be imposed on them through the coercive power of the State -
including its power to tax. This question is discussed along with the available empirical evidence
in section II. This then allows us to set up some general principles for the design of social safety
nets in line with classical liberalism, in Section III. The fourth and final section then applies these
principles to provide some operational guidelines for the individual components of the safety net
distinguished above.
l. Labor Market Risks in the Process of Development
Safety Nets and Welfare States
The subject of this paper are transfers - which allow the income/consumption of an
individual to be higher than it would otherwise be. The distinction between the "welfare state" and
a 'social safety net" then turns essentially upon the universality of coverage of transfers under a
welfare state as opposed to the restriction of collectively provided benefits under a social safety net
to "those among the poor who are unlikely to benefit from economic growth or human resource
development. The safety net includes income transfers for those chronically unable to work -
because of age or handicap - and those temporarily affected by natural disasters or economic
recession" (World Bank, 1992, 4-4). The same report also then notes two essential elements in any
design of a social safety net: "The issues concerning transfers and safety nets include identifying the
groups in need of assistance, and the means of targeting assistance to those groups cost-effectively.
Are these questions for public policy, or are they adequately addressed by the traditional family
network?" (World Bank, 1992, 2-13)                                              -
By contrast, welfare state advocates (e.g., see the recent World Bank conference papers on
"targeting") favor universality as it alone, in their views, provides a feasible means to achieve the
ends sought to be subserved by a social safety net. Some (e.g., Barr, 1992a) have argued that,
because of the ubiquitousness of imperfect information, markets for risk will be inherently imperfect.
Hence, universal welfare states are required as part of an efficient solution to deal with "market
failure." We do not need to deal with this last argument as it was dealt with in a previous paper (see
Lal, 1993b). Suffice it to say that this 'nirvana economics" provides no credible justification for a
welfare state.
An implicit objective of those who argue against targeting and in favor of universal welfare
states is distributivist. This is not surprising as they are by and large socialists who subscribe to the
common socialist end of egalitarianism. But as set out in Lal (1993a), classical liberalism necessarily



2
eschews egalitarianism.' Nor is egalitarianism a universally accepted moral code that can be used
to justify the universalized public transfers of a welfare state on ethical grounds.
The need for a social safety net - to be found in most economies - is not necessarily a
reflection of morality, nor of public action seeking to correct "market failures," but is due to the
ubiquitousness of risk in men's lives and the possibility of reducing its individual burden through
various forms of mutual assurance. This could take various forms: through market processes such
as insurance, as well as social institutions like the family. The term 'social" needs to be clarified
in this context. Though it has become coterminous with public (state) action, in its original sense
it refers only to co-operative action - private or public. In this sense, to say that there is a need
for a "social safety net" does not prejudge whether this should be provided through private or public
action.
Labor Market Risks
So what are the risks in labor markets against which mankind has sought some form of
insurance through social co-operation and how have these changed with different stages in economic
development? As most economies were agrarian organic economies, in which (until fairly recently)
labor was scarce relative to land, there were two major types of risks that were endemic.
'The technocratic approach to public policy has been based on the welfare economics pioneered by Bergson
and Samuelson and which Sen has labetled "welfarism.' The trouble with this has always been: how are the
judgments about the social good which form the social welfare function to be derived? As Sugden (1993) notes:
"most welfarists think of social welfare judgments as being made by a particular individual, but from a neutral
standpoint." This basic idea can be found, for example, in Arrow's (1963, p.l07) ethically neutral 'public
official ...it can be traced back to Adam Smith's 'impartial spectator'* (p. 1949). In classical utilitarianism, pleasure
provided the measure of goodness. But this view flounders on the impossibility of finding a metric for pleasure,
in particular one which is interpersonally comparable. Nor, as Sen (1982) has shown in his penetrating critique of
"welfarism," is the revealed preference version any more coherent. He has then attempted to argue for his own
conception of the social good based on "capabilities" and *functionings, * which unlike revealed preference welfarism
"does not automatically assert that whatever the individual chooses is good for him. Then, by aggregating in some
way the good of all individuals, we can arrive at a conception of the social good" (Sugden 1993, p. 1951). But as
Sugden notes "given the rich array of functionings that Sen takes to be relevant, given the exten$ of disagreement
among reasonable people about the nature of the good life, and given the unresolved problem of how to value sets"
it is not operational, and provides no alternative to the measurements of real national income and practical
cost-benefit analysis based on Marshallian consumer theory that is the bread and butter of applied economics.
The alternative to this attempt to define the social good - which Platonic Guardians then maximize - is
an alternative vision of public policy where "society is seen as a system of cooperation among individuals for their
mutual advantage. On this view, the primary role of government is not to maximize the social good, but rather to
maintain a framework of rules within which individuals are left free to pursue their own ends" (Sugden 1993, p.
1948). This is the classical liberal vision of the state as a civil association in Oakeshott's terms (see Lai 1993a).
Its contemporary exponents are the Virginia public choice school and the neo-Austrians like Hayek (1945). Among
philosophers it is reflected in the contractarian tradition of the American liberal Rawls (1971), and the libertarian
Nozick (1974). It is the viewpoint from which these series of papers have been written. But it should be noted that
Sen (1992) has claimed that even this tradition can be subsumed into his own and that its proponents are also
egalitarians with respect to the good "liberty." But as Sugden (1993) argues convincingly, this is a misreading of
this contractarian and classical liberal position which cannot be subsumed, as Sen suggests, into a theory of the
social good.



3
The first, which was a form of systemic risk, was related to the need to tie labor down to
land where various forms of intensive agriculture were feasible and profitable (largely in alluvial
plains, e.g., the Indo-Gangetic plain in India). Without this tied labor, less intensive and productive
forms of agriculture would have had to be adopted. Various institutions - feudalism in Europe (see
Bloch, 1961) and the caste system in India (see Lal, 1988) - evolved to deal with this systemic risk.
More ubiquitous were the cyclical risks associated with changing climate.  Various
institutional arrangements like the jajmani system in India (see Lal, 1988), sharecropping (see
Bardhan, 1989), interlinked contracts in other factor markets (see Bardhan, 1980), provided the ways
to cope with these risks. Moreover, in these traditional societies, unemployment and destitution as
normal' states were virtually unknown (see Garraty, 1978). Feudal societies were designed to
provide a place for every member, and the local 'society' - village, clan or tribe - provided the
requisite social safety net. The main risk, as in a famine, was of not being able to spread highly
covariant risks across the local group. In traditional Indian village society, for instance, this risk was
partly dealt with by acquiring relatives through marriage in geographically distant areas whose
climatic risks would not be correlated with their own. They could then expect the necessary transfers
from their spatially distant relatives when they were suffering climatically induced falls in their
income. (see Rosenzwig, 1988 for the continuing relevance of this feature in modem Indian village
life).
In other parts of the world: 'Among tribes, no doubt, a rough and ready concern for the sick
and old marked most peoples. In settled communities, an essential part was the role played by the
lord. A typical feudal provision was that of the Prussian code of 1795: the lord had to see to it that
poor peasants were given education, that a livelihood for such of his vassals as had no land must be
provided and, if they were reduced to poverty, he had to come to their aid." (Thomas, 1979, p.
577). In Europe, the breakdown of medieval society, and the subsequent agrarian and industrial
revolutions led to major changes. Not least, because population expansion led not only to the
destitution of those without any labor power (the handicapped and the old without any families), but
also to "the poverty of the able-bodied who lacked land, work, or wages adequate to support the
dependents who were partly responsible for their poverty' (Iliffe, 1987, p. 5). They were the
paupers, and altruism apart, it was the danger to civil order from vagrancy which lent urgency to
the alleviation of their poverty once the link between poverty, crime, and vice was perceived. Their
numbers swelled by another form of conjunctural poverty arising from the Industrial Revolution's
trade cycle and the unemployment that ensues in its downturns.
Finally, in most pre-industrial economies self employment was and remains the dominant
form of employment. A self-employed worker combines in his person and personal enterprise (or
household) all those characteristics which, due to the division of labor, are separated in industrial
firms. These are labor, entrepreneurship, and capital. A variation in the demand for the output
produced by these factors of production will be reflected in an instantaneous change in the implicit
value marginal products of the various 'factors'. There cannot be any 'involuntary' unemployment,
therefore, of the self-employed, and hence no question of unemployment insurance (UI) for them.
Only the income transfers to alleviate conjunctural poverty and those that may be deemed necessary
to provide merit goods as part of the social safety net will be relevant for them.



4
A Typology of Risks and Stages of Development
This suggests that we can combine these differing structural features of labor markets, as
countries move from being rural organic economies to industrial ones, with the three fold
categorization of poverty into destitution, conjunctural poverty and mass structural poverty, outlined
in Lal (1993b). This can provide a typography of the differing components of the social safety net
that will be relevant at different stages in development.
First are those countries (mainly in sub-Saharan Africa) which are still largely pre-industrial
and where labor is still scarce relative to land. There may be a great deal of mass structural poverty,
but no social safety net can deal with it. Its only cure is rapid, efficient growth, following the
classical liberal recipe.
The causes of destitution are likely to be a lack of labor power (because of physical
handicaps, age, and the lack of any able-bodied family members). Direct transfers to alleviate such
destitution are unavoidable. Whether these should be public or private is taken up in the next
section.
The major causes of conjunctural poverty are likely to be climatic crises or political turmoil.
Its most dramatic manifestation is a famine. Since the Indian Famine Code was devised by the
British Raj in the late 19th century, the remedy for a famine is also well known. Because of the
common fall in output and income over a geographical area encompassing private transfer networks,
the purchasing power which is required to ward of the famine will have to provided by state action
(through food for work schemes, for instance). As the Indian example shows, apart from one
exception, this administrative solution has eliminated famines. But this solution does require political
stability and a relatively competent administration. Thus in post-colonial Africa, apart from climatic
factors which have periodically put some Africans (particularly pastoralists) at risk, political violence
in a series of civil wars has created an army of destitutes. These African civil wars have created 12
million 'mass distress migrants' between 1964-84 (Oliver's review of Iliffe). Since then, the
Sudanese and Eritriean famines were due to civil wars in which famine was a political weapon used
by the State against secessionists, and the Somali famine which resulted from the collapse of that
country into a Hobbesian state of nature. Similarly, the worst famine in human history occurred in
China as a result of the politically determined Great Leap Forward. It cost 63 million lives (actual
and prospective through lost or postponed births - see Lin, 1990). In all these cases the State,
rather than being the solution, has been the problem.                        I
Second are those countries (mainly in South Asia) where labor is abundant relative to land.
There, in addition to mass structural poverty, destitution is likely to be due not to a shortage, but
an excess of labor power. Its alleviation most often merges with that of mass poverty, for which the
only long term solution is efficient growth. But there will now be a group - landless labor - which
is able-bodied but may not be able to find enough work throughout the year. This creates a risk of
seasonal unemployment, for which many agrarian institutions such as tied farm workers have evolved
(see Lal, 1989). Whether some public action can provide an alternative and better form of insurance
will be discussed in the last section. The risks of conjunctural poverty will be similar in these
primarily agrarian economies to those discussed above.
Third are those countries in which non-farm employment is predominant in either industrial
or mining enterprises or in plantations (mainly in Latin America and parts of East Asia). The main



5
additional source of labor market risk in these economies, where the division of labor has gone much
further than in the primarily agrarian economies of Africa and South Asia, is linked to the industrial
trade cycle and the accompanying unemployment in its downturns.   As the process of
industrialization also often entails a constantly changing pattern of industrial output and hence
industrial structure, frictional unemployment as workers move from declining to rising industries will
also appear. Questions of unemployment insurance and severance payments will be relevant in
promoting flexibility and thence efficiency in these more 'formal' labor markets. It is these aspects
of a form of conjunctural poverty, and of destitution which are likely to be most relevant in thinking
about social safety nets for such economies.
Finally, in a class by themselves are the former socialist economies now attempting to
transform themselves into market economies. Their problem is not the creation of social safety nets,
but the efficient dismantlement of the system of publicly guaranteed entitlements that their full-scale,
socialist, universal welfare states had created. Their problems are shared by those countries (e.g.,
India, Sri Lanka) whose flirtation with socialism created inefficient public sector enterprises, offering
their workers politically determined entitlements which they now want to rescind. For some socialist
countries, as large parts of their existing industrial capital stock is nearly worthless at world market
prices, economic liberalization also entails at least a temporary drop in industrial output. This raises
the scepter of mass structural poverty, whose cure again is efficient growth. However, unlike the
agrarian economies where "the poor have always been with us," this incipient structural poverty
involves a massive frustration of existing expectations (however economically unviable) which could
be politically explosive. Unlike the other set of countries, therefore, for ex-socialist countries,
designing some transitional arrangement of public transfers to deal with structural poverty may be
a political imperative. We deal with these issues in the final section.
AI. Private Versus Public Transfers in Social Safety Nets
As the risk of income shortfalls over an individual's lifecycle is ubiquitous, it would be
extraordinary if most societies had not found means of insurance against these risks. Historically,
destitution and conjunctural poverty were dealt with through five means. The first was through
institutions like the Church which took one of its primary tasks to be the care of the poor. Individual
charity, most often through interhousehold transfers from an extended family, provided a second
means. A third was through organizations of the poor themselves: through self-help organizations
(e.g., rotating credit associations like the contemporary Grameen bank) and the mutual friendly
societies of 18th and 19th century Britain discussed in Lal (1993b). Fourth were various underworld
organizations engaged in crime. Finally, various forms of insurance embodied in interlinked
contracts in factor markets have historically been the major way of dealing with conjunctural poverty
in traditional village economies (see Platteau, 1991).
Private Transfers
Of these the role of private interhousehold transfers is particularly relevant. For though
transfers through religious channels have probably been significant, I have not been able to find any
scholarly analysis of their size or effectiveness. This is surely an area for profitable future research.
Kin-based transfers, reciprocity arrangements and interlinked factor market contracts have been the
major way that traditional societies have dealt with income risk. As Platteau (1991) concludes:
"Even though empirical evidence is scanty (but not altogether absent), the case can reasonably be



6
made that, barring exceptionally unfavorable circumstances (such as repeated crop failures or crop
diseases affecting entire communities), traditional methods for controlling the risk of falling into
distress have usually enabled the people to counter natural and other hazards in a rather effective
way' (p. 156). With the inevitable erosion of village communities it is feared that these private
insurance arrangements will break down and that no private alternative will be available to counter
destitution and conjunctural poverty in increasingly industrial economies.
It is in this context that the role of private inter-household transfers is of great importance.
Cox and Jimenez (1990) provide evidence to show that they are of considerable quantitative
importance. 'For example, among a sample of urban poor in El Salvador, 33 percent reported having
received private transfers, and income from private transfers accounted for 39 percent of total income
among recipients. Ninety-three percent of a rural south Indian sample received transfers from other
households. In Malaysia, private transfers accounted for almost half the income of the poorest
households. Nearly three-quarters of rural households in Java, Indonesia gave private transfers to
other households. About half of a sample of Filipino households received private cash transfers' (p.
206). (Also see Rempel and Lobdell, 1978; Knowles and Anker, 1981; Collier and Lal, 1986;
Oberai and Singh, 1980; and Lucas and Stark, 1985 on the significant size and effects of remittances
within the rural and between the rural and urban sectors in Ghana, Liberia, Nigeria, Pakistan,
Tanzania, Kenya, India and Botswana). Moreover, since the oil price rise of the early 1970's, the
poor in South Asia and parts of Southeast Asia have found remunerative employment in the newly
rich oil states. Their remittances to their Third World relatives has helped to alleviate their poverty
(see Swamy, 1981).
The motivation for these transfers is of some interest. If they were purely altruistically
determined (as in Becker's famous 'rotten kid' theorem),2 then it would imply that with
intergenerational transfers between parents and children there would be dynastic families which
would behave as though they were a single infinite lived individual. Barro's famous Ricardian
equivalence would then hold, with public policies such as debt financing and social security being
completely neutralized by countervailing private action. Warr (1983) and Bemheim and Bagwell
(1988) went further and showed that as "propagation requires the participation of two traditionally
unrelated individuals,...there will be a proliferation of linkages between families." This gives rise
to even stronger neutrality results. "In particular, no government transfer (including those between
unrelated members of the same generation) has any real effect, and all tax instruments (including so
called distortionary taxes) are equivalent to lump sum taxes. In essence, the government can affect
the allocation of real resources only by altering real expenditures. The efficiency role of government
is thus severely limited, and the distributional role is entirely eliminated. More generally,... if all
linkages between parents and children are truly operative, then market prices play no role in the
resource allocation process: the distribution of goods is determined by the nature of intergenerational
altruism" (Bernheim and Bagwell, 1988, p. 309-10).
2 The rotten kid theorem states that: 'when one member [of a family] cares sufficiently about other members
to be the head, all members have the same motivation as the head to maximize family opportmities and to
internalize fully all within-family 'externalities,' regardless of how selfish (or, indeed, how envious) these members
are. Even a selfish child receiving transfers from his parents would automatically consider the effects of his actions
on other siblings as well as his parents. Put still differently, sufficient 'love' by one member guarantees that all
members act as if they loved other members as much as themselves" (Becker 1974, p. 270).



7
As these implications seem to be highly unrealistic, attempts have been made to explain
private transfers as part of an exchange process involving an implicit, mutually beneficial contract
between parents and children, for instance, who, in exchange for their educational expenditure, may
be committed to looking after their parents in their old age (see Kotlikoff and Spivak, 1981; and
Bermheim, Shleifer and Summers, 1985). Lucas and Stark (1985) have developed an intermediate
model in which both altruism and self-interested exchange are the motives for transfers, and found
that it applies satisfactorily to Botswana. They found that the prediction of the pure altruism model
that lower income households will receive higher transfers is not borne out and that instead, as the
exchange model predicts, there is 'a positive association between amount remitted and per capita
income of the household from other sources" (p. 910). In the exchange model the 'greater wealth
of the family should increase its relative bargaining strength' (p. 906), and thus leads to a higher
demand on its 'migrants'.
But as Lucas & Stark recognize, their data - which is cross-sectional - does not allow the
altruistic motive for transfers to be tested in a dynamic context. Rosenzweig (1988) does so. In a
longitudinal study of six villages in three different agro-climatic regions in the semi-arid tropics of
India, he found that "kinship in a risky world not only tends to bond family members in a single
location (in a particular way) but kinship ties are able to be sustained over time and space in implicit
insurance-based transfer schemes which contribute to consumption smoothing in the face of covariant
income risks' (p. 1167). "It is kinship, and common (family) experiences [which] induce trust,
knowledge and altruism among family members, [hence] such income pooling implicit contracts may
be feasible even if spread across wide areas" (p. 1152).
Thus the empirical evidence on the motives for private transfers is mixed. As Cox and
Jimenez (1990) summarize it: "Some studies find an inverse relation between recipients' resources
and transfer amounts received (for instance Kaufman and Lindauer for El Salvador, Kaufman for the
Philippines, Ravaillon and Dearden for rural households in Java, and Tomes for bequests in the
U.S.). But others (Lucas and Stark for Botswana, Cox for [inter vivos transfers in] the U.S.,
Ravaillon and Dearden for urban households in Java, and Cox and Jimenez for Peru) find a positive
relation, which contradicts the altruism hypothesis" (p. 216).
There are some empirical studies which directly estimate the crowding out effect of public
on private transfers. Most of these have been done for the U.S. (see Lampman and Smeeding, 1983;
Cox and Jakubson, 1989; Rosenzweig and Wolpin, 1985; and Gale, Maritato and Scholtz, 1992),
and find some small crowding out effect. For developing countries there are only'two available
studies. For Peru, Cox and Jimenez (1992) found that in the absence of social security in urban
Peru, private inter-household, old age support would have been higher by 20 percent. So there is
considerable but not complete crowding out of private by public transfers. A study of the Philippines
by Cox and Jimenez (1993) is probably more relevant. As they state: "part of the reason for the low
estimates of the degree of crowding out of private transfers by public ones might be due to the fact
that the estimates discussed above are derived in environments [in OECD countries] where public
transfers are already substantial. These transfers may have already crowded out private transfers to
a large extent, rendering the small samples of private recipients uninformative. In contrast, the
Philippines has almost no public welfare payments, which makes it an ideal case study for gauging
the strength of private transfers" p. 96). They found that transfers were widespread and large. They
used the available data to simulate the effects on these private transfers of three public policies;
unemployment insurance, social security and income grants targeted to the poor. For unemployment
insurance they find: "the reduction in private transfers is nearly as large as the boost in income that



8
unemployment insurance gives to households. Ninety-one percent of the increase in household
income from unemployment insurance is offset by reductions in private transfers' (p. 19). For
retirement income they find that "private transfers would be 37 percent higher" if retirement income
did not exist. On a program to completely eliminate poverty by giving each household the difference
between its actual income and poverty line income, they find that after private transfers adjust, 46
percent of urban and 94 percent of rural households below the poverty line before the program would
still be below the line after the program! Moreover, they give reasons to believe that their estimates
of crowding out are biased downwards. This study should certainly give anyone seeking the public
transfer route to deal with labor market risks considerable cause to pause.
These doubts are further strengthened when we consider that private transfers, by relying on
locally held information (see Hayek, 1945) and, in part, on extra economic motivations like trust and
altruism, can overcome many of the problems of adverse selection, moral hazard, etc., which have
so exercised the "nirvana" economics market-failure school. For as Cox and Jimenez (1990)
summarizing the empirical evidence conclude, "private transfers equalize income; private transfers
are directed toward the poor, the young, the old, women, the disabled and the unemployed" (p. 216).
Public Transfers
Perhaps public transfers can do even better, so that we should not worry if they crowd out
private transfers? Public subsidization of the two merit goods - health and education - are the
major public transfers in nearly all developing countries. In addition, social security is important
in many Latin American countries.
One question on which there is some empirical evidence is the incidence of the benefits from
subsidies for merit goods. Beginning with the pioneering studies of Meerman (1979) for Malaysia,
and Selowsky (1979) for Colombia, a number of other studies have addressed this issue. Jimenez
(1989) summarizing the studies done up to 1987, concluded, "students from the highest quartile of
the income distribution profile in Chile, Colombia, Indonesia, and Malaysia receive between 51 and
83 percent of all public expenditures on higher education, whereas those from the lowest 40 percent
receive between 6 and 15 percent. The effect is only partly counterbalanced by the concentration
of primary education subsidies among poor families, which have most of a country's younger school
children. The net result is a distribution of overall educational subsidies roughly proportional to each
income group's population share, with the exception of the Dominican Republic where the poor's
share is still less. The income bias is less for health. Health subsidies for Colombia and Malaysia
are roughly proportional to each income group's population share. But in Indonesia, the poorest 40
percent capture only about 19 percent from public health centres and hospitals" (p. 114).
More recent studies of the effects of health subsidies in Indonesia (see Deolalikar, 1993; van
de Walle, 1992) find that, "with the sole exception of reported morbidity, all the evidence points to
the children of high-expenditure households benefitting more (in terms of the marginal effects on
both health utilization and health outcomes) from government health spending than the children of
low expenditure households. Even in the case of reported morbidity, the greater benefits derived
by the poor relative to the non-poor from government health spending is quite small." Moreover:
'if government health spending largely improves publiciy-provided health care opportunities for the
non-poor but also crowds out private providers (say, traditional healers) that are used typically by
the poor, it could reduce the total amount of medical care opportunities available to the poor. If the



9
health outcomes of the poor are highly responsive to medical care inputs, this could have a
detrimental effect on their health" (Deolalikar, 1993, pp. 26, 28).
Jee Peng Tan (World Bank, 1992, Box 3.9) cites the conclusion from a 1990 World Bank
study of Asian government's spending on health that: "The findings on targeting of public
expenditure on health to the poor in Asian countries are not encouraging... For the countries foi
which data could be assembled, there is little evidence of such targeting, either by income group or
geographical area. The data suggest that the same infections and parasitic diseases that have been
killing people in Asia for centuries continue to do so."
For Brazil (World Bank, 1992, Box A3.2), McGreevey concludes from a 1988 World Baik
report "the poorest 19 percent of the population receives only about 6 percent of social benefits. A
large share of social expenditure benefits higher-income groups. Regional inequalities are also
severe." (Also see Maddison et al., 1992.) For Costa Rica, Riboud (World Bank, 1992, Box A3.3)
concludes on the basis of a 1990 World Bank report, "Total benefits are fairly evenly distributed
among income groups.  This is the result of two opposing and compensating forces: the
regressiveness of the distribution of education benefits and the progressiveness of the distribution of
health benefits.. .Taking account of state pensions, however, social program benefits per capita are
62 percent higher for the richest 20 percent than for the poorest 20 percent of the population."
On social security in Latin America, Mesa-Lago (1983) noted that a stratification of social
security had occurred with each occupational group seeking its own arrangements. "Generally, the
most powerful groups (militarymen, civil servants, the labor aristocracy) exercised pressure to obtain
the best systems so that social security became a pyramid in which the higher strata enjoyed better
systems than the lower strata, while the base of the pyramid was made up of the non-insured. The
more powerful the pressure group, the earlier in time it received protection, the greater the degree
of its coverage, the less its financial cost, and the more generous its benefits' (p. 89). The net result
is that, "the distributive objective of social security (on behalf of the needy) is rarely fulfilled in
Latin America, since in the majority of countries' social security plays either a regressive or null
role. The extreme poverty group is deprived of protection (except for public health and social welfare
programs) and, to add salt to injury, it contributes through taxes and prices to the welfare of those
insured. Within the latter, especially in stratified systems, the lower income groups generally receive
a net social security transference inferior to that accrued by the higher income groups" (p. 95).
Nor are the various public transfers indirectly targeted to help the poor (such,as agricultural
and food subsidies) any more effective in achieving their objectives. Thus Pamela Cox (World Bank,
1992, Box 3.7), summarizing the findings of a 1991 World Bank report on the incidence of
agricultural subsidies in India, found that, "benefits from agricultural input subsidies have gone
overwhelmingly to wealthier and agriculturally advanced regions and to larger farmers." Liberman
(World Bank, 1992) summarizing a 1989 World Bank report's findings on India's safety net
programs concluded, "The public distribution system has high costs and weak targeting, including
a strong urban bias. The national employment schemes have a small positive impact on the income
of poor households, but with relatively low coverage in some very poor states."
Next, what have been the effects of social expenditures on health and education on various
social indicators? The evidence again is mixed. But there are two revealing pieces of evidence that
suggest that the comnmon presumption that a rise in these social expenditures will improve literacy
and life expectancy and reduce infant mortality is not secure. Thus Grosh (World Bank, 1992, Box



10
3.4) summarizing a 1990 World Bank study which 'traced public social sector expenditures for nine
Latin American countries in the 1980's...found that real per capita public social spending on health,
education, and social security fell during some part of the 1980s in every country in the study. The
share of health and education expenditures in total government expenditures also fell, even as that
of social security rose. In spite of lower funding, and no apparent increases in equity and efficiency,
social indicators generally improved in the 1980's." Apart from obvious statistical and other biases
which might explain this anomaly, the most plausible explanation provided is that it might be due
to 'the growing role of non-governmental organizations, and the response of the market-oriented
private sector to enhanced expectations and demand." That is, there was probably a "crowding in"
of more equitable and more efficient private transfers to replace the decline in public ones!
The second piece of evidence is a simple regression we ran on the state level data for India
on per capita public expenditure on health and education between 1976 and 1986 and the changes
in literacy rates and life expectancy and infant mortality rates, given in Ravallion and Subbarow
(1992). In these cross-sections, we found there was no statistically significant relationship between
changes in state level health expenditures and health outcomes, and a statistically significant negative
relationship between changes in educational expenditure and literacy!
Political Economy of Transfer States
So we could go on, but enough has been said to suggest that public transfers are clearly not
the panacea being touted by socialists of various hues. (See Ahmad, 1991 for a representative
sample of this type of viewpoint.) Their efficacy in achieving ends like improved educational and
health outcomes are dubious, and the incidence of their benefits tends to be regressive, certainly as
compared with the evidence on private transfers summarized earlier. The general conclusion about
these transfers is a repetition of a World Bank report's conclusion for Honduras: "most social
prograrns benefit primarily the middle class and rich, through spending on curative hospital care,
pension benefits and higher education. Social spending pays for services that might be financed by
the private sector" (World Bank, 1992, Box A3.5).
This "middle class capture" of the benefits of social expenditure is not confined to developing
countries. It has also been documented for the welfare states of the OECD (see Goodin and Le
Grand, 1987). A systemic process is clearly at work. The new political economy is useful in
understanding it. It is most clearly seen in terms of the political economy of redistribution in
majoritarian democracies. In a two-party, majoritarian democracy, politicians will bid for votes by
offering transfers of income from some sections of the populace at the expense of others. Models
of this political process (which do not require democracy, but rather the interplay of different
pressure/interest groups - see Stigler, 1970; Meltzer and Richard, 1981; and Peltzman, 1980) show
that there will be a tendency for income to be transferred from both the rich and the poor to the
middle classes - the so called "median voter." Even if social expenditures are initially intended to
benefit only the needy, such programs have inevitably been "universalized" through the political
process in democracies, leading to what are properly called transfer rather than welfare states, which
primarily benefit the middle classes.
The poverty alleviation that may occur as a by product of the expansion of the transfer state
is moreover bought at a rising dynamic cost. With the universalization of various welfare schemes,
political entitlements are created whose fiscal burden is governed more by demography than the
conjunctural state of the economy. With the costs of entitlements rising faster than the revenues



11
needed to finance them, the transfer state finds itself in a fiscal crisis. This process is discernible
both in developing and developed countries.
For developing countries Lal and Myint (forthcoming) show how this process is clearly
visible in those countries in their sample (Uruguay, Costa Rica, Sri Lanka, and Jamaica) that, under
the factional pressures of majoritarian democracies, have created and expanded welfare states. All
four welfare states were financed by taxing the rents from their major primary products. With the
expansion of revenues during upturns in the primary product cycle, political pressures led to their
commitment to entitlements, which could not be repudiated when revenues fell during the downturn
in the price cycle. The ensuing increase in the tax burden on the productive primary sector (to close
the fiscal gap) led to a retardation of its growth and productivity, and in some cases to the 'killing
of the goose that laid the golden egg." Thus, while there was undoubtedly some poverty redressal
as a result of the expansion of these welfare states, over the long run the entitlements created
damaged economic growth on which they were predicated, and hence eventually became
unsustainable. Similar processes leading to the fiscal crisis of the state are to be found in many other
developing countries (see Mesa-Lago, 1983; 1989 for Latin America). Not surprisingly, many of
these countries with over-extended welfare states are now seeking to rein them back. The most
dramatic being the case of Chile (see Castaneda, 1992).
Very similar problems are also visible in the more mature welfare states of the OECD. For
though the public provision of transfers to the "deserving poor" in the UK goes back to the
Elizabethan Poor Law, it was not till the late 19th century that, beginning with Bismarck's
introduction of social insurance against sickness, accident and old age, and soon followed in the UK
by Lloyd George's 1911 introduction of national insurance (which in addition, covered
unemployment), public transfers began to expand in all industrial countries (in the U.S., the New
Deal was the watershed). In the post second world war period, these public transfers exploded; in
the 1960s and 1970s the coverage of health and pensions benefits was made universal and their levels
increased. The share of social expenditures in total public expenditure rose from 44 to 60 percent
between 1960 and 1980, and the share of social expenditures in GDP rose from under 14 percent
to nearly 25 percent over the same period (see Hakim and Wallich, 1986). In an earlier study (see
Lal and Wolf, 1986) the deleterious effects of this expansion of the welfare state on the public
finances and on the economy's productivity had been charted. (See Lindbeck, 1990 for a discussion
of the productivity-damaging effects of the Swedish welfare state, which ultimately brought it to its
knees). With the stagflation of the 1970's in part flowing from these trends and with growing
uneasiness about the unintended social consequences of the welfare state (see below),, many OECD
governments took measures to stem the growth of transfers. In some countries which had gone
furthest down the public welfare route, the late 1980s and 1990s saw a growing questioning of the
welfare state in the West, and in some cases (e.g., New Zealand), its partial or virtual dismantling.
One final aspect of the welfare state which is relevant to the theme of these interconnected
papers (Lal, 1993a; Lal, 1993b) is its effects on what de Tocqueville (1835a), in his Denocracy in
America, called the "habits of the heart" - the cultural underpinnings for both a democratic society
and the market economy. The most sustained though controversial critique of the decadence in
private habits engendered by the transfer state is in Murray's (1984) examination of the results of
the U.S. war on poverty. The cultural consequences of the welfare state may be as momentous as
the economic or political ones. This is a very large subject beyond the scope of this paper. But as
Lee Kwan Yew has pointed out, welfare states have, in the least, sapped the family bonds which
provide the safety nets in East Asia.



12
This is of relevance because of the oft repeated claim that 'ageing populations, growing
urbanization and the rising number of nuclear farnilies have weakened the traditional support
provided by the family and increased the need for formal provision' (Ahmad, 1991, p. 106). But
are extended families in the Third World necessarily likely to be extinguished? And is the decline
of not only extended, but also nuclear families in the West due to some inevitable process associated
with economic growth, or as Murray (1984) and many others maintain, is it the unintended
consequence of well-intentioned welfare policies which subsidize teenage mothers, promote single
parent families, and make the type of reciprocal exchange relationships - outlined in our discussion
of private transfers - more and more redundant?3 If the family has been an institution which has,
to some extent, been created and preserved as a form of mutual insurance against life's risks, is the
transfer of these insurance functions to the State not likely to undermine the very institution whose
decline politicians in all the Western welfare states are currently bemoaning? Nor can one assume
that the type of individualistic as opposed to dynastic motivation for marriage which seems to
predominate in the West is a universal human characteristic, nor that it is a necessary accompaniment
of growth. Many Third world cultures (e.g., India and China) seem to be relatively immune to these
Western cultural norms. As Lee Kwan Yew suggests, the family in many Third World cultures may
be relatively immune to the specifically Western social developments which have undermined it in
the West.  But, of course, as economic incentives matter, Third World farmilies could be
undermined, as they have in the West, by similar welfare state policies.
Thus, seeking ways to strengthen traditional methods of dealing with destitution and
conjunctural poverty, and finding new ways to foster the civic virtues and the growth of civil
associations on which (for classical liberals) the health of sustainable democracies as well as market
economies depend is likely to serve the Third World better than finding arguments for a vast
expansion of State provision/financing of social expenditures. For as sociologists (e.g., Nisbet) have
noted, the conversion of welfare into transfer states in the West (and some developing countries),
has led to an attack on civil society from both above and below. From above, the intermediate
institutions of civil society are forced to surrender their functions and authority to professional elites
and the bureaucrats of centralizing states. From below, "rights chatter" - the clamor for numerous
and newly discovered individual rights (see Lal, 1993a) - undermines the authority of those
traditional civil institutions - family, church, school, neighborhood - which in the past have
promoted both private benevolence and the lower order 'vigorous virtues." The acceptance of claims
' The evidence from the U.S. on the effects of the welfare system on incentives is summarzed in Danziger,
Haveman and Plotnick (1981) and Moffitt (1992). The latter also incorporates many of the findings of the former.
Moffits's general conclusion is:
'The literture on the incentive effects of the U.S. welfare system.. .has shown unequivocal evidence
of effects on labor supply, participation in the welfare system and on some aspects of family
structure.. .Yet the review has also shown that the importance of these effects is limited in many
respects. The labor supply effects, whilst statistically significant, are not large enough to explain the
high rates of poverty among female heads.. .In addition, the econometric estimates of family structure
effects are not large enough to explain long-run declines in marriage rates and, in any case, are
incapable of explaining recent upward trends in female headship because welfare benefits have been
declining...Some of the evidence assembled in the review suggests that family-structure issues appear
to be at least as important in understanding the economic status of low income female heads as labor
supply issues...Unfortunately, the research on family structure remains in its infancy compared to the
voluminous research on labor supply.' (Moffitt 1992, p. 56-7).



13
to various welfare rights, substitutes public for private benevolence - sapping the latter, which for
classical liberals is the highest (though scarce) virtue.
It is these longer-run, unintended, social and fiscal consequences of the welfare state which
are now leading to its partial dismantlement in many OECD countries, of which the most dramatic
example is the virtual abandonment of the New Zealand welfare state by its chastened socialist party,
and the most poignant is the gradual extinction of that social democratic beacon of hope the Swedish
"middle way." It is then particularly ironic that, at a time when the welfare state is coming to be
repudiated by its progenitors, international institutions such as UNDP, UNICEF and WIDER are
seeking their extension in the Third World. Hopefully the World Bank will not follow suit!
III. Classical Liberal Principles and Labor Market Insurance
The refurbished classical liberal principles for public action in insuring labor market risks
then follow naturally from the above discussion.
The first principle is that nothing should be done which would damnage the existing private
institutions and channels which provide for private transfers to deal with destitution and conjunctural
poverty. "Forbear" should be the watchword for every proposed scheme which seeks to alleviate
poverty through public transfers.
The second is that, if for whatever reason, public money is sought to be transferred to the
Nneedy," this is best done through private agencies. In particular for the merit goods of primary
health care and primary education, whatever the case for public financing there is none for public
production. As the Bolivia public sector expenditure review noted: 'in the health sector, NGOs
deliver the most effective service. In education, several communities have asked the largest NGO
to manage their public schooling, indicating the perceived better quality education offered, even with
much higher student-teacher ratios in NGO schools... The public education system spends more than
10 times more per child than the largest NGO, which provides high-quality education" (World Bank,
1992, Box A3.1). (Also see Jimenez, Lockheed and Paqueo, 1991.)
The third principle is that the very problems of moral hazard, adverse selection and
monitoring cited by "nirvana economics" for public insurance in fact argue for fostering the
alternative private route which capitalizes on the comparative informational advantage of private
agents with local knowledge. (See for instance the evidence on the relative efficacy of private credit
agents over centralized public agencies in rural credit markets: Binswanger et al., 1985; Braverman
and Guasch, 1989; von Pischke, Adams and Donald, 1983). These private welfare channels can be
promoted by various methods of cofinancing them with public funds. (Though some of the pitfalls
in this public embrace of private NGOs outlined in Lal, 1993a should be borne in mind).
For the World Bank it may be worth considering a radical proposal for channeling all of its
IDA aid for "safety nets" for alleviating destitution and conjunctural poverty through a World Bank
sponsored consortium of international charities. This could be modelled on the highly successful
World Bank CIGAR initiative for agricultural research.
Fourth, in cases where for political reasons public responsibility for destitutes and the
conjuncturally poor has to be accepted, public policy should be concerned with alleviating real



14
hardship and not with equalizing people. The socialist distributivist end must be forsworn. State help
should be concentrated on the minority in absolute need, and the categories of prospective
beneficiaries should not be allowed to multiply as "'need' assumes an elastic dimension in the name
of 'relative poverty'" (Harris, 1988, p. 18).
Fifth, universal provision of welfare and social security benefits should be eschewed. For
particularly in majoritarian democracies, such 'universalization' will corrupt the polity with
competing politicians showing their compassion by indiscriminately buying votes with other people's
money. This in turn will lead to the likely "middle class" capture of the transfer state, and could
lead to endemic growth and fiscal crises.
IV. Guidelines for Public Action on Social Safety Nets
If public action is considered desirable to deal with various labor market risks, what are the
guidelines for specific types of risks that the application of the above principles can provide. For the
political reasons mentioned earlier, the dismantling of the universalist welfare states in former
socialist countries will necessarily involve some continuing public provision of the "safety net."
Moreover, in those countries (e.g., the Soviet Union) where socialism has completely extinguished
or greatly weakened those channels of private insurance through the family or mutual aid agencies,
it may be a long time before private substitutes for public welfare become feasible. Hence the
following discussion may be of particular relevance for such "economies in transition." We deal
with the various elements of the safety net in sequence.
For the merit goods of education and health, as noted in the last section, there may be a case
for financing the poor but none for public production. As J.S. Mill (1848) put it: 'If the country
contains a sufficient number of persons qualified to provide education under government auspices,
the same persons would be willing to give an equally good education on the voluntary principle,
under the assurance of remuneration afforded by a law rendering education compulsory, combined
with state aid to those unable to defray the expense" (p. 161). For health we have argued (see Lal,
1993b) that the poor are best helped by various means to allow them to purchase private care. These
means include mutual aid societies, as well as private charities and NGO's. Matching public funds
for these private institutions would be the method of channeling public funds to the poor. For both
type of merit goods, if the country has an adequate administrative capacity, vouchers earmarked for
purchasing these merit goods given to the preferred purchasers will be more efficient than subsidies
to producers. If, faut mieux, public provision of these merit goods is undertaken, then differential
pricing to recover the costs of provision while subsidizing the poor on the lines developed and
recommended by the World Bank in the past is desirable (see Jimenez, 1989 for. a powerful
restatement of the case for social sector pricing against the universalist welfarists).
Social security pensions form the other major component of social expenditures, at least in
the more advanced developing and former socialist countries. As in most Western welfare states,
they are pay as you go schemes. In the light of demographic trends in most ex-socialist countries,
they are, as in the West, an economic time bomb (see Lapidus and Swanson, 1988 and IMF, IBRD,
OECD, and EBRD, 1991 for socialist countries, and Lal and Wolf, 1986 and Boskin, 1986 for the
West). In countries where the welfare ethos has become widespread, there is the danger that it may
be rational to be feckless in providing for one's old age as "the rotten kid meets the good samaritan'
(see Bruce and Waldman, 1990). Some state compulsion in ensuring everyone provides for a basic



15
pension may be desirable. As in Chile (see Castaneda, 1992) this could be done through earmarked
taxes that are put into fully funded and actuarially fair, private pension funds.
This leaves various forms of income suport programs -  including disability and
unemployment insurance, and transfers to alleviate low-end poverty - which are common in many
Western welfare states, and which are being recommended for the transitional socialist economies
(see Barr, 1992b and Paull, 1991). These schemes involve unavoidable 'tax- cum-subsidy
distortions' because lump sum taxation and subsidization is normally not feasible. The econometric
attempts to provide quantitative evidence of the effects on labor supply, consumption and savings of
these distortions has been inconclusive (see Atkinson, 1987) - another exarnple of the emenrging law
that "all econometric evidence is equivocal!"
Unemployment insurance is of little relevance for the majority of developing countries in
which self employment predominates over wage employment. Even in the so-called modern sectors
of these economies where wage employment is the norm, measured open unemployment rates are
relatively low, and the unemployed are typically highly educated, relatively wealthy urban youth.
Their unemployment reflects, in part, the availability of 'reserves" (from their relatively better off
families) to finance job search in the high-wage, modern sector and, in part, the over expansion of
higher and secondary education because of economically unjustifiable public subsidies (see Squire,
1981). In addition, relatively high institutional wages in the modern sector encourage what can be
termed Harris-Todaro type unemployment. Given these "typical' features of developing country
labor markets, unemployment insurance is only feasible for wage employees in the modern sector.
In an excellent overview of the literature on unemployment insurance, Hammermesh (1992)
argues that the only rationale for unemployment insurance programs is the 'individual based goal of
consumption stabilisation... For this goal to be valid it must be true that workers either estimate the
probability of becoming unemployed correctly, but are so myopic that they do not save for the
eventuality; or they underestimate the probability and thus have insufficient precautionary savings"
(p. 3). But this still does not establish whether such programs should be private or public. Beenstock
and Brasse (1986) have argued that private unemployment insurance would be feasible. (But see
Barr, 1992a who again uses "nirvana economics" arguments to counter this claim). But as
Hamermesh notes: "Before the British unemployment insurance program was introduced in 1911,
several private firms offered UI as an insurance option... The early British companies went bankrupt,
and the classic argument in favor of social insurance for unemployment is that private carriers cannot
insure against the common risk of a nationwide recession. A compulsory privately operated
program, with very large carriers that have sufficient reserves or borrowing capacity to weather a
recession, might not have such problems" (1992, p. 13).
If however, for whatever reason, a publicly funded program of unemployment insurance is
introduced for the urban employees in the formal sector, what should be its features? We need to
minimize the side-effects in terms of the higher urban unemployment/expansion of the urban informal
sector and increased rural-urban migration that may be induced. These effects will depend upon how
the "tax" to finance the benefits is raised. If the tax is levied on modern sector employers while their
wage employees receive the benefits, the net effect will be to raise the 'effective" wage of such
employees and this will, through the familiar Harris-Todaro type dynamic process, lead to a
reduction in modern sector employment, increase the size of the urban "informal" sector, and also
rural-urban migration. If, instead, the "tax is assessed on worker's earnings, there is no effect on
the size of the urban sector, just as there is no impact on the employment or the unemployment rate.



16
[Hence] to the extent we believe that labor-market dualism characterizes developing economies, UI
benefits should be financed by taxes on the earnings of workers in the modern sector"
(Hanimermesh, 1992, p. 35).
From a poverty viewpoint, the seasonal unemployment of landless labor in rural areas is
likely to be of greater imnportance than urban unemployment. Rural public works schemes like the
Maharashtra Employment Guarantee scheme (see Ravallion, 1991), have been effective both in
preventing famines and in dealing with problems of short run income variability. But their success
lies in the self-targeting that is made possible by offering a wage that only the truly needy will
accept.
Severance Payments. In many ex socialist economies as well as those developing countries
which created over-extended and inefficient public sectors, an important part of structural adjustment
is the privatization of such enterprises. In addition, many bureaucracies which were set up to
manage controlled economies need to be slimmed as they move towards market economies. Such
structural adjustment programs typically face political resistance from the public sector workers who
face retrenchment and/or cuts in their real wages. Such workers can exert political pressure to
prevent the rescinding of their politically determined entitlements to future income streams which are
above what they would be able to obtain in the free market. The capitalised value of the difference
between their expected public sector earnings (including pension and other benefits) and those they
could get in the private sector (adjusted for the probabilities of being hired and fired in the market)
represent the rents public sector workers are currently receiving. If their resistance is to be
overcome they might need to be compensated for these rents. This is a political rather than an
economic argument for severance payments, over and above those that might already exist in the
contractual arrangements that may be in force in the respective labor markets (see Rosen, 1985 for
a survey of the reasons why many labor market contracts will have various forms of severance terms
built into them for efficiency reasons).
Ways to deal with this problem will differ from country to country, depending upon its labor
laws and the government's ability to rescind these rents. If the method of adjustment is to be
voluntary, however, certain common problems will be faced. The most imnportant of these is that,
with a heterogeneous labor force, the rents derived from public sector jobs will differ for different
workers, being highest for the "bad" workers whose market opportunities relative to their
entitlements in the public sector are the worst. With imperfect knowledge of each worker's rents,
and the difficulty in devising perfectly discriminating severance payment schemes, if the severance
compensation is set to persuade the last "bad' worker to leave the public sector, the intra marginal
workers will be receiving more compensation than the capitalised value of their public sector rents.
This could mean a very high cost to the fisc. But in some cases (e.g., where the public enterprise
is producing negative value added at world prices), shutting down the enterprise even with this high
cost may lead to a net gain in GDP. In others, where the enterprise might still be viable after
restructuring and privatization (involving retrenchment), the problem of tailoring a severance package
to these differential rents remains. Papers by Fiszbein (1992) and Diwan (1993) provide detailed
analyses of the various options, as well as discussions of severance payment schemes, in a number
of countries.
The most attractive plan which would meet both the objectives of limiting political opposition
and reducing the fiscal burden would be one limited to the workers who are not hired by the newly
privatized enterprise. This tackles the adverse selection problem whereby the "good" workers take



17
the severance package and the 'lemons" are left with the new firm. The severance package for thomie
made redundant should be based on the principle of tailoring the benefits to the median redundant
worker's public sector rents. This would imply that, if the severance package offered uniform
compensation at the level of the rents to the median retrenched worker, all those with lower rents
would be better off, and they would provide the political support for the scheme to override those
workers who would be worse off because their rents were greater than the median. Little more can
be said in principle about the specific terms of these programs which need to be tailored to local
conditions, and in particular, to the relative bargaining power of public sector workers vis a vis the
state.
Transfers to a]leviate 2overty. We have already discussed the reasons for preferring the
strengthening of private channels and institutions for the alleviation of low-end poverty (destitution
and conjunctural poverty) in countries where the welfare state ethos has not become widespread.
This is best done by the public cofinancing of local and international charities as well as local mutual
self-help associations. But there are a number of countries, mainly the ex-socialist, where, either
because they already have universalist welfare states and/or the institutions for the private sector
option are not present, some form of public provision for income support of the low-end poor may
be necessary. How can this best be done? A large part of the relevant group is likely to consist of
people who, for identifiable reasons, are incapable of earning a living (the handicapped, the mentally
ill, the old and infirm without any families or savings). Their inability to finance their basic needs
is genuinely involuntary, and hence the problems of perverse incentives which bedevil so many social
insurance type programs is not likely to arise. This is the category of people the Victorians labelled
the 'deserving poor." In the absence of private charity, public assistance to provide them some basic
minimum income may be justifiable. But again, for the reasons discussed in previous sections, this
public assistance is probably best channeled through local voluntary associations.
The problems of disincentives and the creation of a dependency culture is particularly acute
for the able-bodied poor (see Murray, 1984). No completely satisfactory scheme is available on the
basis of Western experience to alleviate their poverty without the perversion of incentives. A
negative income tax, or basic income schemes (e.g., those recommended by Friedman and in the
Meade Committee report, 1978) involve unacceptable tax burdens. Thus Brittan and Webb (1990)
estimate that such a scheme which guaranteed a minimum income for all UK families at the current
income support level would require a uniform income tax rate of 40 percent, much higher than the
current basic income tax rate of 25 percent. However, the alternative of targeted means-tested
benefits, which are withdrawn when the unemployed find work or the poor move up the income
ladder, lead to well known disincentives associated with unemployment and poverty traps. There
is no obvious remedy. Decentralization to local communities of the administration and monitoring
of these programs to help the able-bodied poor - as is done in Sweden and Switzerland- may be
part of the answer (see Lindbeck, 1990 and Segalman, 1986). An income maintenance scheme that
distinguishes between the able-bodied poor and the deserving poor has recently been proposed for
Eastern Europe by an IMF economist (see Paull, 1991). It has much to commend it. The deserving
poor are means tested and publicly assisted to bring them up to a minimum income level. The
able-bodied poor are assisted through a series of training, job-search and workfare programs to enter
or re-enter employment. The proposals are costed for Poland and do not seem to imply unreasonable
shares of GNP as compared with Polish social expenditures for 1990, or similar expenditures in
developed countries. If the programs can be decentralized so that they are administered and
monitored locally, they might provide the best that can be done to deal with an intractable problem.



18
But in the long run, even in these ex-socialist countries the hope must lie in the creation of
a civil society where private transfers and institutions arise to take care of the needy. Besides
promoting the supreme moral virtue - private benevolence - extolled by classical liberalism, it
would also prevent that corruption of the polity that the creation of universal welfare states has
invariably caused.



Bibliography
Ahmad, E. 1991. 'Social Security and the Poor: Choices for Developing Countries." World Bank
Research Observer 6(1): 105-127.
Ahmad, E. et al. (eds.) 1991. Social Security in Developing Countries. Oxford: Clarendon Press.
Arrow, K. 1963. Social Caoice and Individual Values. Second Edition. New Haven: Yale.
Atkinson, A.B. 1987. "Income Maintenance and Social Insurance." In A.J. Auerbach and M.
Feldstein (eds.) Handbook of Public Economics, Vol. 2. Amsterdam: North Holland.
Atkinson, A.B. and J. Micklewright. 1991. "Unemployment Compensation and Labor Market
Institutions: A critical Review." Journal of Economic Literature 24, no. 4(December): 1679-
1727.
Bardhan, P. 1980. "Interlocking Factor Markets and Agrarian Development: a Review of Issues."
Oxford Economic Papers 32(1).
_   (ed.) 1989. The Economic Theory of Agrarian Institutions. Oxford: Clarendon Press.
Barr, N. 1992a. "Economic Theory and the Welfare State: A Survey and Interpretation." Journal
of Economic Literature 30, no. 2(June): 741-803
1992b. "Income transfers and the Social Safety Net in Russia." Studies of Economies in
Transition. Paper No. 4. World Bank, Washington DC.
Becker G. 1974. "A Theory of Social Interactions." Journal of Political Economy 82(6): 1063-91.
Reprinted in G. Becker, The Economic Approach to Human Behavior, Chicago, 1976.
Beenstock, M. and V. Brasse. 1986. Insurance for Unemployment. London: Allen and Unwin.
Bernheim, B. 1986. "On the Voluntary and Involuntary Provision of Public Goods." American
Economic Review 76, no. 4(September): 789-93.
Bemheim, B. and K. Bagwell. 1988. "Is Everything Neutral?" Journal of Political Economy 96,
no. 2(April): 308-38.
Bernheim, B., A. Shleifer and L. Summers. 1985. "The Strategic Bequest Motive" Journal of
Political Economy 93, no. 6(December): 1045-76
Bhalla, S. 1988. "Is Sri Lanka an Exception?" In T.N. Srinivsan and P. Bardhan (eds.) Rural
Poverty in South Asia. New York: Columbia University Press.
Binswanger, H. et al. 1985. "Credit Markets in Rural South India: Theoretical Issues and
Empirical Research." Research Report No. 45. Agricultural and Rural Department, World
Bank, Washington DC.
Bloch, M. 1961. Feudal Society. London: Routledge.
Bodie, Z. 1990. "Pensions as Retirement Income." Journal of Economic Literature 28, no.
1(March): 2849.



20
Boskin, M.J. 1986. Too Many Promises. the Uncertain Future of Social Security. Homewood,
Illinois.
Braverman, A. and J.L. Guasch. 1989. "Rural Credit in Developing Countries.' Population,
Planning and Research Paper No. 219. World Bank, Washington, DC.
Brittan S. and S. Webb. 1990. "Beyond the Welfare State." Hume Paper No. 17. David Hume
Institute, Aberdeen University Press.
Bruce, N. and M. Waldman. 1990. "The Rotten Kid Theorem Meets the Samaritan's Dilemma."
Quarterly Journal of Economics CV(February): 155-66.
Castaneda, T. 1992. Combating Poverty. San Francisco: International Center for Economic
Growth, ICS Press.
Collier, P. and D. Lal. 1986. Labor and Poverty in Kenya: 1900- 1980. Oxford: Clarendon Press.
Cox, D. 1987. "Motives for Private Income Transfers." Journal of Political Economy 95, no.
3(June): 508-46.
1990.  "Intergenerational Transfers and Liquidity Constraints."  Quarterly Journal of
Economics CV, no. 1(February): 187-218.
Cox, D. and G. Jakubson. 1989. "The Connection Between Public Transfers and Private
Interfanily Transfers." Mimeo. Boston College, Boston, MA.
Cox, D. and E. Jimenez. 1990. "Achieving Social Objectives Through Private Transfers:
'areview." World Bank Research Observer 5, no. 2(July): 205-18.
1992. "Social Security and Private Transfers in Peru." World Bank Econonmc Review 6,
no. 1(January).
1993. "Private Transfers and the Effectiveness of Public Income Redistribution in the
Philippines." Mimeo. Paper presented at the World Bank Conference, Public Expenditures
and the Poor: Incidence and Targeting.
Danziger, R., R. Haveman and R. Plotnick. 1981. "How Income Transfers Affect Work, Savings
and Income Distribution." Journal of Economic Literature 19, no. 3(September): 975-1028.
de Tocqueville, A. 1835a. Democracy in America. Paris.
1835b. "Memoir on Pauperism."  Reprinted in S. Drescher (ed.) De Tocquevile and
Beaumont on Social Reforn. New York, 1968.
Deolalikar, A.B. 1993. "Does the Impact of Government Health Spending on the Utilization of
Health Services by Children and on Child Health Outcomes Differ by Household
Expenditure: The Case of Indonesia." Mimeo. Paper presented at the World Bank
Conference, Public Expenditures and the Poor: Incidence and Targeting.
Diwan, I. 1993. "Efficient Severance Payment Schemes." Mimeo. World Bank, May 1993.



21
Fiszben, A.  1992.  'Labor Retrenchment and Redundancy Compensation in State Owned
Enterprises: The Case of Sri Lanka." Internal Discussion Paper No. IDP 121. South Asia
Region, World Bank, Washington, DC.
Gale, W., N. Maritato and J.K. Scholtz. 1992. 'Effects of Public and Private Transfers on Incorne
Variability and the Poverty Rate." Mimeo. University of Wisconsin, Madison.
Garraty, J.A. 1978. Unemployment in History. New York: Harper and Row.
Goodin, R.E. and J. Le Grand. 1987. Not Only the Poor. London.
Hakirn, L. and C. Wallich. 1986. 'OECD Deficits, Debt and Savings. Structure and Trends
1965-81: A Survey of Evidence." In D. Lal and M. Wolf (eds.) Stagflaon, Savings and
the State. New York: Oxford University Press.
Harnermesh, D.S. 1992. "Unemployment Insurance for Developing Countries." Policy Research
Working Papers No. WPS 897. World Bank, Washington, DC.
Harris, R. 1988. "Beyond the Welfare State: An Economic, Political and Moral Critique of
Indiscriminate State Welfare and a Review of Alternatives to Dependency." Occasional
Paper No. 77. Institute of Economic Affairs, London.
Hayek, F. 1945. "The Use of Knowledge in Society." American Economic Revew 35, no.
4(September): 519-30.
Iliffe, J. 1987. The African Poor. Cambridge.
IMF, IBRD, OECD and EBRD. 1991. A Study of the Soviet Economy. Paris: OECD.
Jimenez, E. 1989. "Social Sector Pricing Revisited: A Survey of Some Recent Contributions."
Proceedings of the World Bank Annual Conference on Development Economics, 1989.
Jimenez, E., M.E. Lockheed and V. Paqueo. 1991. "The Relative Efficiency of Private and Public
Schools in Developing Countries." World Bank Research Observer 6, no. 2(July): 205-18.
Kaufman, D. 1982. "Social Interaction. A Strategy for Economic Survival Among the Urban Poor:
Theory and Evidence." (Ph.D. Diss.) Mimeo. Harvard University, Cambridge, MA.
Kaufmann, D. and D. Lindauer. 1986. "A Model of Income Transfers for the tJrban Poor."
Journal of Development Economics 22(July): 337-50.
Knowles, J.S. and R. Anker. 1981. "An Analysis of Income Transfers in a Developing Country."
Journal of Development Economics 8(April): 205-6.
Kotlikoff, L.J. and A. Spivak. 1981. "The Family as an Incomplete Annuities Market." Jounal
of Political Economy 89(April): 372-91.
Lal, D. 1988. The  indu Equilibrium Vol.l Culural Stability and Economic Stagnation, India c.
1500BC-AD1980. Oxford: Clarendon Press.
1989. The Hindu Equilibrium Vol.2: Aspects of Indian Labor. Oxford: Clarendon Press.



22
_.___* 1993a. "Participation, Markets and Democracy." HRO Working Paper No. 20. Human
Resources Development and Operations Policy Department, World Bank, Washington, DC.
1993b. "The Role of the Public and Private Sectors in Health Financing." Mimeo. Human
Resources Development and Operations Policy Department, World Bank, Washington, DC.
Lal, D. and H. Myint. (In press.) The Political Economy of Poverty, Equity and Growth - A
Comparative Study. Oxford: Clarendon Press.
Lal, D. and M. Wolf (eds.) 1986. Stagflation, Savings and the State. New York: Oxford
University Press.
Lampman, R.J. and T. Smeeding. 1983. "Interfamily Transfers as Alternatives to Government
Transfers to Persons." Review of Income and Wealth 29(March):45-66.
Lapidus, G.W. and G.E. Swanson. 1988. State and Welfare USA/USSR. Berkeley: Institute of
International Studies.
Lin, J. 1990. "Collectivization and China's Agricultural Crisis in 1959-61." Journal of Political
Economy 98(December).
Lindbeck, A. 1990. "The Swedish Experience.' Seminar Paper No.482. Institute for International
Economic Studies, Stockholm.
Lucas, R.E. and 0. Stark. 1985. "Motivations to Remit: Evidence from Botswana." Journal of
Political Economy 93(October): 901-18.
Maddison, A. et al. 1992. The Political Economy of Poverty, Equity and Growth: Brazil and
Mexico. New York: Oxford University Press.
Meade Committee. 1978. The Structure and Reform of Direct Taxation. London: Institute of Fiscal
Studies, Allen and Unwin.
Meerman, J. 1979. Public Expenditure in Malaysia: Who Benefits and Why. New York: Oxford
University Press.
Meltzer, A. and S. Richard. 1981. "A Rational Theory of the Size of Government." Journal of
Political Economy 89:914-27.
Mesa-Lago, C. 1983. "Social Security and Extreme Poverty in Latin America." Journal of
Development Economics 12:83-110.
1989. Ascent to Bankruptcy: Financing Social Security in Latin America. Pittsburgh.
Mill, J.S. [1848] 1970. Principles of Political Economy. Ed. D. Winch. London: Penguin Books.
Mitchell, O.S. 1993. "Trends in Retirement Systems and Lessons for Reform." Policy Research
Working Paper No. WPS 1118. World Bank, Washington, DC.
Moffitt, R. 1983. "An Economic Model of Welfare Stigma." American Economic Review 73, no.
5(December): 1023-35.



23
1992. 'Incentive Effects of the U.S. Welfare System - A Review." Journal of Economic
Literature 30, no. 1: 1-61.
Murray, C. 1984. Losing Ground. American Social Policy 1950-1980. New York: Basic BoolM
Nisbet, R. The Quest for Community. New York.
Nozick, R. 1974. Anarchy, State and Utopia. Oxford: Blackwell.
Oakeshott, M. 1993. Morality and Politics in Modern Europe. New Haven: Yale.
Oberai, A.S. and H.K.M. Singh. 1980. "Migration, Remittances and Rural Development."
International Labor Review (March-April).
Paull, G. 1991. "Poverty Alleviation and Social Safety Net Schemes for Economies in Transition."
IMF Working Paper No. WP/91/14. Research Department, International Monetary Fund,
Washington, DC.
Peltzman, S. 1980. 'The Growth of Government." Journal of Law and Economics 25, no. 3:
209-87.
Platteau, J.P.  1991.  Traditional Systems of Social Security and Hunger Insurance: Past
Achievements and Modem Challenges." In E. Ahmad et al. (eds.) Social Security in
Developing Countries. Oxford: Clarendon Press.
Ravallion, M. 1991. "Reaching the Poor Through Public Employment: Arguments, Evidence and
Lessons from South Asia." World Bank Research Observer 6, no. 2(July): 153-76.
Ravallion, M. and L. Dearden. 1988. "Social Security in a 'Moral' Economy: An Empirical
Analysis for Java." Review of Economics and Statistics 70(February):36-44.
Ravallion, M. and K. Subbarow. 1992. "Adjustment and Human Development in India." Journal
of Indian School of Political Economry 4, no. 1: 55-79.
Rawls, J. 1971. A Theory of Justice. Oxford.
Rempel, H. and R. Lobdell.  1978.  "The Role of Urban-to-Rural Remittances in Rural
Development." Journal of Development Studies 14(April):324-41.
Rosen, S. 1985. "Implicit Contracts: A Survey." Journal of Economic Literature 23:1144-75
Rosenzweig, M. 1988. "Risk, Implicit Contracts and the Family in Rural Areas of Low-Income
Countries." Economic Journal 98(December): 1148-70.
Rosenzweig, M. and K. Wolpin.  1985.  "Specific Experience, Household Structure, and
Intergenerational Transfers: Farm Family Land and Labor Arrangements in Developing
Countries." Quarterly Journal of Economics 100(supplement): 961-87.
Segalman, R. 1986. The Swiss Way of Welfare: Lessons for the Western World. New York.
Selden, T.M. and M.J. Wasylenko. 1992. "Measuring the Distributional Effects of Public
Education in Peru." Mineo. Paper presented at the World Bank Conference, Public
Expenditures and the Poor: Incidence and Targeting.



24
Selowsky, M. 1979. W7ho Benefits from Public Expenditure? A Case Study of Colomnbia. New
York: Oxford University Press.
Sen, A.K. 1982. Choice, Welfare and Measurement. Oxford: Blackwell.
_   1992. Inequality Reexamined. Oxford: Clarendon Press.
Squire, L. 1981. Employment Policy in Developing Countries. New York: Oxford University
Press.
Stark, 0. and D. Levhari. 1982. 'On Migration and Risk." Economic Development and Cultural
Change 31, no. 1: 191-6.
Stigler, G. 1970. "Director's Law of Public Income Distribution." Journal of Law and Economics
13, no. 1: 1-10.
Sugden, R. 1993. "A Review of Inequality Re-examined by Amartya Sen." Journal of Economic
Literature 31, no. 4(December): 1947-86.
Swamy, G. 1981. "International Migrant Workers' Remittances: Issues and Prospects." World
Bank Staff Working Paper No. 481. World Bank, Washington, DC.
Thomas, H. 1979. A History of the World. New York: Harper and Row.
Tomes, N. 1981. "The Famnily, Inheritance, and the Intergenerational Transmission of Inequality."
Journal of Political Economy 89(October):309-22.
van de Walle, D. 1992. "The Distribution of the Benefits from Social Services in Indonesia,
1978-87." Policy Research Working Paper mimeo. Country Economics Department, The
World Bank, Washington, DC.
von Pischke, J.D, D.W. Adams and G. Donald. 1983. Rural Financial Markets in Developing
Countries. Baltimore: Johns Hopkins.
Warr, P.G. 1983. "The Private Provision of a Public Good is Independent of the Distribution of
Income." Economic Letters 13:207-11.
World Bank. 1992. "Poverty Reduction Handbook." Washington, DC.



Human Resources Development and Operations Policy Working Paper Series
Contact for
Title                         Author                   Date             paper
HROWP1 1   Operational Education          George Psacharopoulos   September 1 993  L. Malca
Indicators                                                              37720
HROWP12   The Relationship Between the   John Clark                October 1993      P. Phillip
State and the Voluntary                                                 31779
Sector
HROWP1 3   Obstacles to Women's           Joseph Kutzin            October 1993      0. Shoffner
Access: Issues and Options                                              37023
for More Effective
Interventions to Improve
Women's Health
HROWP14   Labor Markets and Market-       Arvil V. Adams           October 1993      S. Khan
Oriented Reforms in Socialist                                           33651
Economies
HROWP1 5   Reproductive Tract Infections,  May T.H. Post           October 1 993     0. Shoffner
HIIV/AIDS and Women's                                                    37023
Health
HROWP1 6   Job Security and Labor         Ricardo D. Paredes       November 1993   S. Khan
Market Adjustment in                                                    33651
Developing Countries
HROWP1 7   The Effects of Wage            Luis A. Riveros          November 1993   S. Khan
Indexation on Adjustment,                                               33651
Inflation and Equity
HROWP1 8   Popular Participation in        Philip R. Gerson        December 1993   L. Malca
Economic Theory and Practice                                            37720
HROWP1 9   Economic Returns from          Edwin Mansfield          January 1994      I. Dione
Investments in Research and                                             31447
Training
HROWP20   Participation, Markets and       Deepak Lal              January 1994      L. Malca
Democracy                                                               37720
HROWP21   Safe Motherhood in               Patricia Daly           January 1994      0. Shoffner
Francophone Africa            Michael Azefor                            37023
Boniface Nasah
HROWP22   Indigenous People and           George Psacharopoulos   February 1 994    I. Conachy
Poverty in Latin America      Harry Anthony Patrinos                    33669
HROWP23   Is Grameen Bank Sustainable?  Shahid Khandker            February 1994    S. David
Baqui Khalily                             33752
Zahed Khan
HROWP24   Concepts of Educational         Marlaine E. Lockheed     March 1994        M. Verbeeck
Efficiency and Effectiveness    Eric Hanushek                           34821
HROWP25   Scientific Research for         Erik W. Thulstrup        March 1994        L. Malca
Development                                                             37720
HROWP26   Issues in Education Finance      Stephen P. Heyneman    April 1994         B. Cassorla
and Management in ECA and                                               37172
OECD Countries



Human Resources Development and Operations Policy Working Paper Series
Contact for
Title                        Author                  Date              paper
HROWP27   Vocational Education and        Julian Schweitzer       April 1994       A. Gonzalez
Training: The Role of the                                              37799
Public Sector in a Market
Economy
HROWP28   Social Security Issues and      Nguyen X. Nguyen        May 1994          M. Espinosa
Elements of Reform                                                     37599
HROWP29   Health Problems and Policies    Mary Eming Young        May 1994          0. Shoffner
for Older Women: An                                                    37023
Emerging Issue in Developing
Countries
HROWP30   Language and Education in       S.M. Cummings           May 1994          M. Espinosa
Latin America: An Overview   Stella Tamayo                             37599
HROWP31   Does Participation Cost the     Jesko Hentschel         June 1994         D. Jenkins
World Bank More? Emerging                                              37890
Evidence
HROWP32   Research as an Input into       Harold Alderman         June 1994        P. Cook
Nutrition Policy Formation                                             33902
HROWP33   The Role of the Public and      Deepak Lal              June 1994        M. Espinosa
Private Sectors in Health                                              37599
Financing
HROWP34   Social Funds: Guidelines for    Soniya Carvalho         July 1994        K. Labrie
Design and Implementation                                              31001
HROWP35   Pharmaceutical Policies:        Graham Dukes            July 1994        0. Shoffner
Rationale and Design         Denis Broun                               37023
HROWP36   Poverty, Human Development  Harsha Aturupane            August 1994       P. Cook
and Growth: An Emerging      Paul Glewwe                               30864
Consensus?                   Paul Isenman
HROWP37   Getting the Most out of         Helen Saxenian          September 1994  0. Shoffner
Pharmaceutical Expenditures                                            37023
HROWP38   Procurement of                  Denis Broun             September 1994' 0. Shoffner
Pharmaceuticals in World                                               37023
Bank Projects
HROWP39   Notes on Education and          Harry Anthony Patrinos   September 1994  1. Conachy
Economic Growth: Theory                                                33669
and Evidence
HROWP40   Integrated Early Child          Mary Eming Young        October 1994     0. Shoffner
Development: Challenges and                                            37023
Opportunities
HROWP41   Labor Market Insurance and      Deepak Lal              October 1994     M. Espinosa
Social Safety Nets                                                     37599