Staying Secure, Staying Poor: The “Faustian Bargain”

World Development Vol. 31, No. 3, pp. 455–471, 2003 Ó 2003 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0305-750X/03/$ - see front matter

www.elsevier.com/locate/worlddev

doi:10.1016/S0305-750X(02)00213-9

Staying Secure, Staying Poor: The ‘‘Faustian Bargain’’ GEOF WOOD University of Bath, UK Summary. — The determining condition for poor people is uncertainty. Some societies perform better than others in mitigating this uncertainty. In such societies we observe welfare regimes which reduce the uncertainties of the market to provide for all citizens minimum conditions for reproduction. Such societies are in a minority. Elsewhere, destructive uncertainty is more pervasive. Under these conditions, the poor have less control over relationships and events around them. They are obliged to live more in the present and to discount the future. Risk management in the present involves loyalty to institutions and organizations that presently work and deliver livelihoods, whatever the longer term cost. Strategic preparation for the future, in terms of personal investment and securing rights backed up by correlative duties, is continuously postponed for survival and security in the present––the Faustian bargain. Ó 2003 Elsevier Science Ltd. All rights reserved. Key words — South Asia, poverty, insecurity, vulnerability, political economy, livelihoods

1. INTRODUCTION The key condition under which poor people enact their capabilities for self-improvement is the need for social security (Nussbaum, 2000; Sen, 1982; and the UNDP Human Development Reports). Governments, which are genuinely poverty-focused, try to create enabling conditions for people to move individually, or collectively, from the condition of being poor into a more secure, sustained, nonvulnerable state of well-being. Most governments fail, however usually because they and their supporters are implicated in reproducing the social, economic and political conditions which create the uncertainty and insecurity barriers to that movement. When governments fail in this way (rights, social protection etc.), poor people have to rely upon more personalized resources in the pursuit of security. This becomes a desperate search for security (Wood, 2001) in which longer term goals of autonomous improvement (Doyall & Gough, 1991) have to be put on hold, maybe forever. Alienation becomes a deliberate strategy of survival in present time. This paper continues to explore a line of thought opened up in Wood (2001), and in a more recent paper trying to calibrate the western origins of social policy to the immature capitalisms and predatory states of postcolonial economies (Wood, 2003). The argument is developed through some critical engagement with the literature on vul455

nerability, livelihoods, poverty and social development. The main conclusion from the critique of present livelihoods and vulnerability literature is that the vast majority of poor people face chronic rather than stochastic insecurity. Epistemologically, the paper then offers a phenomenology of insecurity located within assumptions about surrounding hostile political economy, drawing upon the idea of the peasant analogue. This is then represented as a tradeoff between the freedom to act independently in the pursuit of improved livelihoods and the necessity of dependent security, illustrated through what might best be termed as ‘‘stylized ethnography,’’ drawing from three decades of field level interaction with mainly rural, but sometimes urban, poor people in South Asia. 1 It is clear from these encounters that poor men and women are dominated by dysfunctional time preference behavior, in which the pursuit of immediately needed security places them in relationships and structures which then displace the longer term prospects of a sustained improvement in their livelihoods. This is the reference to the ‘‘Faustian bargain’’ of the title.

2. SOCIALLY CLASSIFYING POVERTY In the vast literature on poverty, there is a curious reluctance to distinguish between types

456

WORLD DEVELOPMENT

of poverty in terms of agency and social action. The capabilities literature get close, but do not really offer an anthropological insight into capacities for social action in a hostile political economy. We need that insight because in the circumstances of inequality and the associated arbitrary exercises of power, ‘‘graduation’’ (the individual or collective movement away from poverty and vulnerability) has to include confrontation and struggle as well as personal human resources investment (i.e., a mixture of Marx and Herzen 2 respectively). Thus, when considering the social and political routes to poverty eradication, 3 the poor can and should be significantly classified by the extent of their capacity for social action. Not all types of poor people can effectively act for themselves, even if supported externally to do so. Thus we have idiosyncratic, chronic poverty comprising varied but extreme incapacity for social action: the elderly; orphans (thus perhaps only transitory); widows in patrilineal and patriarchal societies; people with disability (physical as well as learning difficulties); people with long-term illness and morbidity. To these groups, we might add outgroups of various kinds whose exclusion reduces their capacity for social action: migrants; ethnic minorities; minority religious sects within dominant cultures. 4 Thus people may have reduced agency either for individual or for broader ascriptive reasons. In the absence of other help, such people really do have to rely upon responsible and accountable governments, prepared to uphold a broad concept of human rights and prepared to offer meaningful social protection via affirmative action and welfare. They are not well placed, however, to bring about such responsibility in government, and have to rely upon the agency of others, who are capable of social action to this end. Failing this, they have to rely upon the direct agency of others bound to them by some sense of morality and community to offer social protection through informal arrangements. But, even for the capable poor the constant probability of retributive action, by those in power over them, sets severe limits to their social action for poverty eradication in the sense of structural confrontation for institutional reform. Thus while it is easily understandable that the idiosyncratically, chronic poor have to opt for dependent security as their only survival option, we are making the more generic argument that the socially inherent characteristic of most poverty requires the same equation: staying secure, staying poor. Dependent for-

tunes are better than the risk of no fortunes at all. Why should this be so? A further defining characteristic of poverty is uncertainty: being unable to guarantee future stocks and flows. This induces short-term behavior and the avoidance of riskier, longer term investments in economic or personal projects. To be poor means, inter alia, to be unable to control future events because others have more control over them. This is why a sense of political economy is essential to understanding the constrained choices and options facing the poor. People are poor because of others. Securing any kind of longer term future requires recruiting the support of these others, but this only comes at a price: of dependency and the foreclosure of autonomy. Becoming a client, in other words. This involves the acceptance of truncated ambitions of self-improvement and advancement in order to secure basic welfare. Perversely, therefore, we encounter the deliberate strategy of choosing a coping level of poverty as the social condition of securing a sustained, albeit low level, livelihood.

3. STRUCTURAL POVERTY: NONIDIOSYNCRATIC COVARIANCE When understanding the poor in a poor country context, the peasant is a stronger analogue than that of employed worker. 5 Insecurity and uncertainty induce risk averse behavior, by leaving poor people more exposed to livelihood threatening risk. This insecurity and risk is partly an issue of time and partly an issue of social capital and social resources. ‘‘Time’’ refers to the discounting issue and is elaborated elsewhere in Collard (2001), Wood (2001) and JID (2000). Certainly the ‘‘snakes and ladders’’ analogy is useful here (Room, 2000). Across the tenuous uncertainty of time, precarious families are stretching out a survival strategy, but always dominated by a hand to mouth reality, which prevents the necessary preparation to cross the chasms and avoid the traps that reside en route. Room discusses this in terms of endowments as resources and relationships which can offer ladders to bliss if aided by passports, or which offer only snakes leading to social exclusion unless inhibited by buffers. Passports and buffers are created through the public/private partnerships of social policy, but unfortunately the institutional

STAYING SECURE, STAYING POOR

prospects of achieving them are lowest where the need is greatest. With social resources so unevenly distributed within the society, the problem for the poor is that they are exposed to the weaknesses of social capital (as a public good), without any prospect of meaningful social resources (personalized networks) to compensate. Their claims across the Institutional Responsibility Matrix (IRM) 6 are weak, as a result. Therein lies their major source of risk. Moreover, without the social options to manage that risk, they have to rely more heavily upon their immediate family and less upon transactions with less intimate others. Hence the shrinking moral universe, and the validity of the peasant analogue. This exposure to risk is multiple and covariant. The World Development Report 2000/ 2001 (WDR) discusses risk in various places, distinguishing between micro-, meso- and macro-levels of risk in Chapter 8, and covariant risk, particularly induced by conflict in Chapter 3 (see especially Box 3.2) and Chapter 7 (World Bank, 2000). The WDR claims to encompass the risk terrain by also distinguishing between covariant and idiosyncratic risk (Chapter 8), such as illness, injury, old age, violence, harvest failure, unemployment and food prices (World Bank, 2000, p. 136). But nowhere does the discussion acknowledge the chronic 7 aspects of risk induced by inequality, class relations, exploitation, concentrations of unaccountable power and social exclusion as absence of ‘‘community’’ membership. In other words, an institutional and relational account of risk is missing. This is a key objection to much of the contemporary livelihoods discourse: it fails to explain the microcircumstances of poor people in terms of meso- and macro-institutional performance, which express political economy and culture. A ‘‘resources profile’’ approach to vulnerability (see Lewis, Glaser, McGregor, White, & Wood, 1992 for the original discussion), wherein households are understood as having a range of material, human, social, cultural and political resources to deploy, is basically a framework of nonidiosyncratic covariance, with weakness on one dimension triggering weakness on another with an unraveling affect on livelihood security as a whole. Some elements of this covariance are more familiar than others. For example, we expect poverty, seasonality of incomes and food availability, nutrition levels, morbidity, acute illness and loss

457

of employment to go together, leading to loss of assets and further spiraling decline. Until recently, we might not have connected strength in social resources to human resources such as health and education; or labor participation (under a ‘‘material resources’’ heading: i.e., income flows) to common property access where an ability to contribute labor constitutes membership. It is this covariance of risk which needs to define policy process and content: understanding in context which type of intervention offers the most leverage on strengthening the livelihood portfolio as a whole via identification of key risk linkages. The WDR addresses this partially in its ‘‘Mechanisms for Managing Risk’’ (World Bank, 2000, p. 141). Here it distinguishes between reduction, mitigation and coping objectives on the one hand; and between ‘‘IRM’’ categories of response divided between informal (individual/household and group based) and formal (market and publicly provided) on the other. But since the WDR is derived from and therefore limited by its prior anti-septic, depoliticized treatment of the issue, it continues to be na€ıve about social and political institutions and the constrained opportunities for positive action via the rules embedded in those institutions. Thus an analytic focus upon physical incapacity of male adults for the labor market might not matter if there is no labor market, or if access is heavily segmented ethnically and culturally. Prices on essential agricultural inputs, or a leveling out of seasonal fluctuations in local level food prices, could be much more significant than capabilities for local peasants and casual, agricultural wage laborers in determining real income entitlements. Of course prices are themselves outcomes of an only partially commodified political economy, characterized by segmentation and embedded interlocking, in which the poor are, by definition, price takers rather than price makers. These arguments set up a proposition. The behavioral imperatives of risk aversion in the present may deliver short-term security while reproducing the conditions for long-term insecurity in the future. This causation exists particularly strongly in the social domain. The individual and household needs for functioning social resources in the context of overall weak social capital require either overstrong reliance upon internal family relations (which may be reciprocal or hierarchical) or allegiance to other providers (hierarchical) at the cost of dependent

458

WORLD DEVELOPMENT

and sometimes bonded loyalty (adverse incorporation). But such risk avoidance extends to other behavior, which has to favor meeting immediate needs over future ones within a peasant analogue. Examples might be crop diversification and subsistence preferences regardless of prevailing prices which might reward specialization; or a deliberate strategy of debt acquisition foreclosing future investment options (e.g., raising dowry capital versus investment in education); and so on (Wood, 2001). Thus, when poor social actors are negotiating their institutional landscape, their cognitive maps comprise: discounting; managing immediacy within severely constrained choices; awareness about long-term loss for short-term gain; and frustration about never being able to get ahead of the game for long enough to really commit resources for the future. How can these time preferences be altered to convince poor people of sufficient present security to invest in their future? Certainly we can conceive of social protection, safety nets and welfare more generally, whether delivered formally or informally, as having a fundamental development function by altering time preferences. In policy terms, there seems to be a basic choice of response, depending on whether the diagnosis emphasizes a security or capabilities stance, leading to social protection or social investment respectively (see Matin and Hulme, in this issue, for an analysis of a program that seeks to combine ‘‘livelihood protection’’ and ‘‘livelihood promotion’’ to help poor people escape the Faustian bargain). But, given the security approach pursued in this paper, how far is it appropriate and ethical 8 to shift the balance of effort in social policy from the stance of intervention to compensate for market outcomes (social protection) to the stance of supporting poor peopleÕs higher level of entry point into labor, commodity, services markets (social investment). The implication of such a shift is to leave it to peopleÕs agency for the performance of real markets to improve poor peopleÕs lives. In response to this elementary problem of agency and time preference behavior, the condition of insecurity reflects vulnerability and the pursuit of a secure livelihood entails rights and correlative duties, which are likely to be informal in the context of poor governance. First, insecurity and vulnerability; followed by a discussion of the precariousness of rights and correlative duties in overcoming insecurity. This leads into the idea of a phenomenology of insecurity, illustrated by the notion of the

‘‘peasant analogue’’ before turning to a general description of the conditions which produce insecurity and how that plays out in different searches for dependent security as the default option for insecure individuals and families. 4. INSECURITY AND VULNERABILITY In making the link between vulnerability and insecurity, there is no need simply to reproduce the thinking of other recent texts on livelihoods and vulnerability (see Loughhead, Mittal, & Wood, 2000; Room, 2000; Wood & Salway, 2000 for critical summaries of recent stances). But a few presumptions, arising in that literature, will inform our position. First, vulnerability is not synonymous with poverty, so that the nonpoor vulnerable need to be included in a pro-poor social policy. Second, therefore, we are dealing with dynamic categories: individuals and families are always in a process, somewhere along their domestic cycle, experiencing troughs and peaks of financial and other security. The WDR refers to the transitory poor, others to poverty churning (see Hulme and Shepherd, in this issue, for further discussion of these categories). Thus, third, in trying to catch that sense of movement, livelihoods thinking should use the present continuous categories of: improving, coping and declining. This is not entirely satisfactory unless one is clear that we are confining these terms to an analysis of the poor, and are not seeking to apply them to people who are in nonvulnerable, secure conditions. 9 Fourth, the better known attempt to capture aspects of this dynamism has been CarneyÕs model (Carney, 1998) of sustainable rural livelihoods (SRL), which has its own roots in the work of De Waal (1989), Lewis et al. (1992), Moser (1998), Swift (1989) and Wood (1994). The SRL has been recently critiqued in Wood and Salway (2000) for: overprivileging social agency; unrealistic assumptions about poor peopleÕs social action; confusion between social capital and social resources; having a stochastic approach to shocks as a decontextualized rather than chronic feature of poverty; and an inadequate account of political economy and power (similar criticisms, in effect, to those of the WDR, above). As a counter to some of these criticisms, Room (2000) offers ‘‘snakes and ladders’’ as a more obvious framework for understanding the dynamic ‘‘careers’’ of people in contrasting trajectories of poverty: improving or ‘‘gradu-

STAYING SECURE, STAYING POOR

ating;’’ coping; and declining through irreversible ratchets/traps (unless assisted by social protection in general and safety nets in particular). Although ‘‘snakes and ladders’’ offers only limited additionality to its predecessors, 10 its importance to a security perspective is: ––it tracks individuals and groups through gateways of opportunity and disaster, with some indication of what happens to them on the way (it is a ‘‘career’’ approach); ––it is thus consistent with a poor actor-oriented epistemology of agency negotiating structure, expressed as a hostile political economy; ––and it therefore offers a model for understanding how people negotiate the institutional responsibility matrix. 11 The main conclusion from this critique is that the vast majority of poor people face chronic rather than stochastic insecurity. The basic conceptual issue here is that shocks are not shocks but hazards. 12 In policy terms, this prompts a distinction between ‘‘relief’’ type interventions (when shocks are shocks, whose unpredictability requires rapid mobilization of short-term response) and ‘‘preparation’’ for the more predictable hazards which affect sub-sets of the population chronically. Thus ‘‘relief’’ responds to situations where security has broken down in surprising ways: e.g., flash floods in Italy. But ‘‘preparation’’ should be in position for the predictable hazards of flooding in Bangladesh, in a well-functioning, nonhostile political economy (i.e., one with good governance, economic growth, public revenues and policies of redistribution via state instruments). This notion of ‘‘preparation’’ is thus directed at improving the capacity of poor people to negotiate their institutional landscape, which features hazards more significantly than it does shocks. It consists of the creation and maintenance of security, especially when that security is predictably threatened by life-cycle events. 5. INSECURITY, RIGHTS AND CORRELATIVE DUTIES Who performs the functions of either relief or preparation? The key question for poor people in poor countries is: where do security inducing rights and correlative duties most securely and predictably reside if neither the state nor the market can be relied upon? The argument is that in the context of societies with poor governance, nonlegitimate states and political in-

459

security, poor people have to rely on (in short hand) a gemeinschaft rather than gesellschaft basis of rights. For the market, we have the classic debates between Granovetter (1985), Platteau (1991) and Scott (1976), and now Fukuyama (1995), Putnam (1993) and dozens of others, about the capacity of emerging societies to evolve moralities of exchange, and therefore trust, beyond face to face, personalized transactions with perfect information toward generalized exchange between strangers under conditions of imperfect, asymmetrical information. Clearly the issue of rights and entitlements in the market incorporates a notion of fairness into the relationship between scarcity, quality and price. And as we know from Sen (1981), the principle of fairness can break down catastrophically for the poor, as sudden scarcity, whether real or manipulated, dominates the exchange equation. Under such conditions, there are no ‘‘moral economy’’ restraints to speculation or concern over disastrous outcomes for some, since there is a low level of moral proximity between the gainers and losers in distant markets, as entitlements collapse. So, rights in markets beyond those circumscribed by the moral economy are contingent and precarious. The prevailing political conditions of insecurity in many parts of Africa reflect such fragility. But in the peasant analogue, applicable to other parts of Africa as well as South Asia and other poor regions and pockets, significant proportions of poor people have significant proportions of their economic transactions within more localized, moral arrangements where a residual sense of fairness persists and therefore rights to entitlements exist. In a welfare regime approach to poverty alleviation 13 (Esping-Andersen, 1990, 1999), the amoral market, producing unsustainable outcomes for poor people, is compensated for by the state: residual in the ‘‘dualist’’ liberal regimes; targeted in the conservative; and universal in the social-democratic. The main criticism of this approach is that it represents a decontextualized view of the state, benignly delinked from the morality of the prevailing political economy. Thus the idea of rights emanating from the legitimated state, serviced by intermediary organizations (social capital in the Putnam, 1993) constituting civil society, is almost laughable as a responsible description of the evolution of political institutions of power and authority in sub-Saharan Africa. South Asia evolved differently, with respect to civil

460

WORLD DEVELOPMENT

society, with a longer and more educated basis to an organized critique of the colonial state, alongside an incorporated set of intermediary institutions. But even in South Asia, the tradition of an expected set of rights defined by the state, realized and maintained through civil pressure, is much stronger in contemporary India than its neighbors in Bangladesh and Pakistan. Thus the idea of rights enshrined in the state remains a weak and contestable phenomenon in the cognitive maps of social actors (rich and poor alike). Rather, rights have a tenuous position in the state and market arenas of the domestic IRM given the real history of actually existing social capital. Where do they actually reside most securely in the minds of social actors in poor countries? Where are the universal rights and needs most nearly provided for in the domestic, and sometimes globalized, IRM? If the domestic state and market domains continue to be problematic, then clearly we have to look to the ‘‘community’’ and household, kin and clan domains. A small logical digression is necessary before proceeding further. It is axiomatic that we cannot conceive of society without some established acknowledgement of need, and a notion of rights to ensure the meeting of need. In other words, rights and perceptions of need have to reside somewhere to fulfil the conditions for calling a collectivity of people ‘‘a society.’’ Otherwise, we are only looking at anarchy and war; i.e., such a state of insecurity that no oneÕs interests can be met, so that there is no reason for that particular collectivity to exist. Bevan (2001) gets close to that position when being pessimistic about Africa, though, of course, she is arguing that the insecurity conditions support the interests of a few powerful people and some of their followers for a while. Outside the domain of the state and therefore law, such rights and entitlements are usually referred to as ‘‘informal’’ in contrast to formal. As suggested above, the alternative institutional domains to the state and the market can be designated as ‘‘community’’ and household. Of course, there can be many ambiguities in terms such as community and household. Both are heavily contested terms. ‘‘Community,’’ for example, can stretch from imagined ones (Anderson, 1983) to closed, locational and residential ones, with other variants in between. Some communities can be more moral than others (Bailey, 1966). They can be characterized by reciprocity and hierarchy, sometimes si-

multaneously. For social actors, the experience of community is variable, according to circumstances of conflict and unity, of fission and fusion. In other words, they are fluid, not fixed, and our understanding of them has to depend upon an actor-oriented epistemological perspective. Sometimes, the construction of community reflects kin structures, thus blurring the distinction between community and household. Indeed, we have preferred the term ‘‘household’’ over ‘‘family’’ precisely to enable kin relations to appear under a community heading, while reserving the term household to refer to much closer, hearth-bound, interdependencies and senses of responsibility. 14 Thus, within ‘‘community’’ we can include clans and lineages which offer social actors crucial identities as well as the social frame within which rights of allocation of scarce resources occur, such as land, water, access to pasture, places to build homesteads, and so on. This would apply with equal force to say Northern Pakistan and rural East Africa. Kin dimensions of community also offer a key basis of ‘‘membership,’’ and with membership goes rights, which are connected to prevailing presumptions about needs and entitlements. To lose ‘‘membership’’ is to be excluded. People lose membership under various conditions: migration; resettlement; urbanization (until ‘‘membership’’ is regained, or reestablished); failure to conform or perhaps contribute; being elderly or infirm (e.g., in the West); being cast as a minority in the context of larger-scale events (e.g., Jews in Germany in the 1930s and 1940s); or being outcasted in various ways in South Asia and Africa. Within the political economies of poor countries, especially in the poorest ones of South Asia and sub-Saharan Africa, what general informal rights pertain? Given the vested interest of the rich in prevailing structural inequalities, we need to distinguish between rights through hierarchical relations and reciprocal ones. Rights through hierarchical relations basically operate within relations of adverse incorporation and clientelism. They can be classed more as welfare ones than developmental. That is to say, clientelist relations will offer relief but rarely invest in the long-term security of others through, for example, education or capital transfers on easy terms. Relief or welfare transfers, for example, through tied labor, long-term debt or interlocked tenancy agreements, occur in return for loyalty and other ‘‘dependent’’ favors which contribute to

STAYING SECURE, STAYING POOR

the reproduction of the initial inequality. Such relations are as likely to occur between close kin as between others more distant. 15 With market penetration and a widening of economic opportunities, the shrinking of moral ties further reduces the sense of responsibility for others, especially at the community level. Indeed, with the ‘‘welfare regime’’ ideology, growing in the minds of the rich, of the interdependence between markets and states, the sense that poor people are the responsibility of government grows and community level rights about welfare correspondingly decline. At the same time, especially in rural contexts, community leaders/elders and stronger classes are reluctant to see whole poorer families or poor relatives completely outcast or cast out. Where, in Northern Pakistan, for example, access to common property remains an important part of the family livelihoods portfolio, then those unable to fulfill the conditions for membership are often assisted to that effect: for example widows and the elderly or infirm (including mentally). 16 Other basic needs rights might also be guaranteed locally: ––allocation of homestead land; ––allocation of agricultural and pasture land (where the community or extended kin group traditionally disposes); ––access to drinking water (but much less so to irrigation water); ––access to materials (wood and mud) for building construction; ––access to fuelwood or peat; access to wild fish stocks and forest products (i.e., gathering and hunting rights); ––a share in communally engineered new land (via forest clearing, new irrigation channels, and feeder roads opening up new physical access); ––and, access to primary health care and even education facilities possibly with some means tested adjustment to expected cost recovery. Where the provision of these rights, in the sense of correlative duties to honor them, are hierarchical via the collectivity of richer, dominant families in the community, then we might imagine a collective version of adverse incorporation, where there is a strong expectation for poorer, aided, families to express gratitude by not challenging the basic arrangements of the political economy which reproduces their poverty in the first place. In this way, we observe institutional and plural clientelism at the

461

community level, alongside the more private, individualized form which emphasizes intense personal loyalties and dependencies. Thus we have collective and individual patronage, with the former offering more long-term security than the latter. Rights through reciprocal relations can be problematic in other ways. Basically rights offered and guaranteed by those with little power in the society or local community are not worth much in longer security terms. Thus the poor do not have much to offer each other beyond very micro, immediate transfers plus sympathy. But, such a dismissive remark needs to be qualified in at least two opposing directions. First, reciprocal relations, offering rights between the poor, are further limited by the structural conditions of poverty which place the poor in competition with each other for scarce resources, including those social resources of patronage. Solidarity is not necessarily natural. But secondly, movements to build solidarity among the poor have precisely emerged from the interrelated analysis of structural inequality and the need to overcome the ‘‘within poor’’ competition that results from it. Such movements intend to create the social resources for the poor that are naturally lacking, and thereby contribute to the formation of social capital (i.e., institutions) more conducive to their long term interests. Proshika in Bangladesh would be a good example of this approach. 17 So reciprocal relations can be enhanced, and this has to be a major ingredient of any poverty-focused agency in poor conditions. But, it is these socio-political characteristics of poverty, as revealed through the precariousness of rights and correlative duties, which have to be interwoven into a security perspective.

6. PHENOMENOLOGY OF INSECURITY How then should we construct a model in which security is sought at the price of continuing poverty? Several concepts have to be squeezed into a holistic understanding, encapsulated by the idea of the peasant analogue. The way into this understanding can be described as the phenomenology of insecurity, supported by a ‘‘poor actor-oriented’’ epistemology. 18 In another language, we have to be ‘‘emic’’ in our methodological approach. 19 Poor people in the poorer parts of the world can be characterized

462

WORLD DEVELOPMENT

by discourses such as ‘‘peasant’’ and/or ‘‘client,’’ reflecting the particular embedded nature of livelihood options in a noncommodified socio-political economy. Countries in South Asia and sub-Saharan Africa remain predominantly agrarian and pastoral, and even though this will demographically change over the next quarter century, the associated principles of social organization will persist perhaps indefinitely, since we cannot equate urbanization with a linear route to commodified modernization (see, e.g., Roberts, 1978). At the same time, while these countries have operating nationstates (just) their problems of governance and effectiveness remain (Landell-Mills, 2001; UNDP, 2000; Wood & Salway, 2000). In subSaharan Africa, continuing to be agrarian is only part of the understanding since the general conditions of insecurity through wars, famines and AIDS challenge the very conception of nation-state as a valid territorial entity. Bevan characterizes these situations as ‘‘insecurity regimes,’’ with large-scale political clientelism (not just localized socioeconomic clientelism) as the main framework of social inclusion (Bevan, 2001). The principles of being a peasant extend as a social description into urban, industrial and informal sector life i.e., into urban as well rural labor markets, and into urban as well as rural lifestyles (Roberts, 1978, and later, Loughhead et al., 2000; Wood & Salway, 2000). This ‘‘peasant’’ analogue thus represents a phenomenology of insecurity: risk aversion, discounting, covariance of risk, emphasis upon reproduction (physical and social), significance of the domestic/life cycle, intergenerational forms of transfers and provision, intrafamily dependency ratios, significance of localized social resources in context of weak social capital. Pervasive clientelism takes this phenomenology a step further by emphasising loyalty over voice in the absence of exit options (Hirschman, 1970), but therefore the kind of loyalty which extends and compounds the problem of governance because it is personalized, arbitrary and nontransparent. This analytic stance has to be placed within an actor-oriented epistemological perspective located within a political economy appreciation of power and inequality which inevitably places the political onus of pro-poor institutional change upon the poor themselves. This leads to four key presumptions: ––that only the poor will ultimately help themselves to the point of structural signifi-

cance in relation to the basic terms of control over key societal resources; ––second, that the poor will act according to their perceptions of options, but that these perceptions can be enlarged by their own and othersÕ actions to expand room for manouevre; ––third, that individual acts can be significant at the level of the ‘‘graduating’’ individual, but come up against the usual social mobility arguments in which only a few change status and only in steps rather than jumps; ––so that, fourth, collective action can be structurally significant for wider units of solidarity with many gains on the way to full structural reformation (which may never be). Thus the desperate search for security (Wood, 2001) takes two principal forms: individual agency and collective or social action, connected to alleviation and eradication respectively. Thus poverty alleviation is about reducing the incidence of poverty via individual processes of graduation and successful incorporation into existing social arrangements and patterns of distribution. Whereas poverty eradication relies upon the principle of structural change, and is about cohorts of the poor confronting power and inequality. Mindful, therefore, of context and actor-oriented epistemology, a phenomenological position is adopted by considering the insecurity/security axis as the central link between livelihoods aspirations and the precarious social conditions through which they have to be realized. It links conditions, perceptions and family/domestic life cycles very closely together. If, when considering livelihood strategies, we focus upon key decision-making levels, then the primary points are: individual; nuclear family; and extended family (sometimes joint, but increasingly less so). We must always recognize that individuals and families see themselves as dynamic not static. They intrinsically and inherently operate within a strong sense of life-cycle, the domestic cycle of the family with its notions of stewardship, with shifting dependency ratios over time. And they are perpetually threatened by events, requiring continuous tradeoffs between present consumption (the more poor, the higher the immediate necessity) and future investment (the more poor, the less certain the conditions for any investment).

STAYING SECURE, STAYING POOR

7. CONTINUING RELEVANCE OF THE PEASANT ANALOGUE Despite rapid economic change in many parts of the poor world entailing polarization and depeasantization in the context of expanding labor markets, the peasant analogy continues to inform our basic understanding of these decision-making levels. This remains true even when we acknowledge structural change in the direction of: marketization and commodification; wage employment; industrialization and the new service economy; urbanization; rising incomes; and the nucleation of families (see Kearney, 1996 on reconceptualizing the Mexican campesino). In poor countries, we cannot assume that these changes will inexorably lead us toward commodified modernization. Contemporary survival options include seeking to reproduce ‘‘peasant’’ behavior and client status under conditions of urbanizing labor markets (Khan, 2000), characterized by segmentation and informal sector attributes (Opel, 2000) as well as increasingly unstable families (Jesmin & Salway, 2000). This peasant mentality, perforce, has to focus more upon reproduction than production as the central motivation for managing the domestic cycle whether annual or intergenerational. The peasant, with low technological control over the environmental conditions for production and always a price-taker rather than a price-maker, is pervaded by a sense of market insecurity. But since the peasant is rarely, if ever, self-sufficient, then any exchange, as a prerequisite for survival, is fraught with danger. In addition, with the state historically functioning more as a predator–protector than as a market compensator and enabler, the state represents part of the problem of insecurity rather than a moderator of it. Thus the peasant has to rely for survival more upon those institutions where membership is more complete, acknowledged and legitimate: i.e., community and kin. This, arguably, is where the peasant has more chance of controlling events and reducing insecurity. This only works up to a point. Where the community level is itself characterized by severe inequality, then class or other exclusions undermine the value of these institutions to the excluded as a basis for dealing with uncertainty and insecurity. Certainly, under such conditions, any notion of reciprocity has to give way to hierarchical, adverse incorporation: i.e., the

463

embracing of client status and a consequent reinforcement of clientelism as the pervasive political settlement.

8. DIMENSIONS OF INSECURITY Although the analogue of the peasant has been deployed above, it is also important to reflect upon wider conditions of insecurity derived loosely from a countryÕs poverty status as well as the family level, life-cycle conditions. The wider conditions also contribute to the uncertainty at the local, community and family level and undermine the viability of institutions at that level to perform secure reproduction functions. These points are made more with reference to contemporary Africa, than South Asia or elsewhere. But, when considering them, let us not forget Afghanistan, Kashmir, Burma (and the border areas with its neighbors), East Timor, other parts of Indonesia and the Philippines, parts of Colombia (despite its overall paradoxical progress on economic and social indicators) (Barrientos, 2001), and the Balkan region. Recent history would add a few more examples outside Africa too. Thus consider the following list: ––the socially embedded state producing discretionary, even arbitrary outcomes with legitimacy based upon authority rather than accountability, and sometimes straightforward coercion as a substitute for legitimacy; ––such states, and the challenges to them, as a cause of war, with large clan and ethnic factions capturing the state and excluding others; ––intense, all or nothing, competition over scarce and valuable natural resources (i.e., minerals and watersheds); ––war as pervasive civilian dislocation negatively impacting upon different points of the life-cycle over short-term periods (crop seasons), medium-term periods (theft of livestock) and longer periods (undermining a predictable basis for human, natural and productive investment); ––such dislocation undermining the incentives for large-scale, long-term, public investment which might, for example, manage water more securely and therefore crops; ––colonial re-structuring of agrarian and pastoral economies undermining flexibility of response to climatic conditions (e.g., the

464

WORLD DEVELOPMENT

starving pastoralists in the Karamoja famine of 1982); ––thus, for some countries especially in Africa, periodic threats to food security and collapse of markets; ––under such conditions, a shrinking morality which emphasises the significance of ever decreasing circles of more immediate kin as the corollary of the breakdown of public trust; ––enforced mobility of populations as refugees from war zones and disasters, sometimes the outcome of arbitrary colonial borders; ––long-term oppressed minorities under siege (e.g., Southern Sudan, parts of Afghanistan); ––legacies of indentured labor (especially in Southern Africa) with migrant, male cultures and machismo gender outcomes interacting with trade routes to socially underpin pandemics like AIDS, now a major source of insecurity at all institutional levels. Under these conditions, it is hardly surprising that the poor in poor countries have a high discount rate. The demands on the present are too extreme to warrant sacrificial investment in a highly uncertain future. This can be exacerbated by global economic conditions as well as thoughtless domestic macroeconomic policy which removes certainty from skill-based labor markets, and thereby removes the propensity for private or public human capital investment in them. (See Kanbur, 2001 for the all important distinction between growth, which is always positive for poverty reduction; and particular growth-oriented economic policies, which maintain poverty where it might otherwise have been reduced.) Better, therefore, for the poor to operate in the spheres of the known, familiar and controllable. The survival algorithm is stronger than the optimizing one (Lipton, 1968). 9. SEARCH FOR DEPENDENT SECURITY: STYLIZED ETHNOGRAPHIES (a) The supplicant beggar The most obvious and perhaps extreme example of the process is the beggar, seeking preferential client status from a single or range of loyal patrons. They may also seek stable sites and pay rents for the privilege to the police or

other informal brokers in the expectation of building up stable support from regular passers-by. In order to introduce an element of certainty into their precarious hand-to-mouth existence, the tactic has to be to draw the potential patron into the inner place of the concentric circles of moral proximity. To move a patron from the instrumentality of a single stranded transaction (which can easily be broken at any time) to multistranded ones, thereby gaining wider moral attention. To gain a commitment not just to a ‘‘case’’ (labeled as a beggar like any other) but beyond to a ‘‘story,’’ where the relationship becomes more personalized and individual, thus cementing a more certain moral bond. Such commitments might be achieved by introducing wider features of the story for example, referring to illness and medical expenses, or to support for childrenÕs education. Thus the beggar is clearly exercising agency in the attempt to create more certain surrounding social and cultural conditions through which entitlements might become more predictable. Clearly this beggar analogy is a metaphor for a more generic process of drawing transacting parties (which can include formal institutions like the state) further into a moral basis for the relationship, making it harder for rights and services to be subsequently withdrawn. But this kind of agency entails a particular performance as a worthy and deserving client. Need has to be continuously demonstrated, thus foreclosing other forms of agency such as autonomy, independence and self-reliance. (b) The urban mastaan A key story of the next two decades is the rapid urbanization of the large poor countries, especially in South and East Asia. As indicated above, the livelihoods dimension of that urbanization can still be understood through applying the peasant analogue. Roberts (1978) pioneered this anti-modernity argument for South America, but it remains with equal force. Embodied within the notion of ‘‘peasant’’ is weakness and vulnerability in the social and cultural conditions through which needs have to be met. Poor urbanites are thus like peasants in being weakly positioned in relation either to the state or different markets (commodities, labor and services). They require networks and patron-brokers to link them more successfully to essential opportunities in these state and market domains. In urban Bangladesh, such

STAYING SECURE, STAYING POOR

patron-brokers are referred to as mastaan (Khan, 2000). They operate mafia-like through an emphasis upon loyalty. They function as intermediaries between vulnerable families/ individuals and more formal institutions. Within the slums, they manage shelter and services (electricity, water etc.), charging ‘‘rents’’ for provision and protection even on governmentowned sites. They control access to employment, charging laborers for the links and charging employers for delivering and maintaining quiescent labor. Poor urbanites, and especially new migrants, have no option but to gain membership of such networks and patronage. The price for such loyalty is not to challenge the structural conditions, which in turn deny them long term autonomy and rights. (c) Traditional patrons: lineages and feudal legacies Within societies dominated by kinship and clan structures in various forms, the networks of loyalty and patronage are partially coterminous with blood ties of varying intensity. Extended, joint families or wider lineage formations may generate patron figures from among themselves either through a deliberate ‘‘Beckerian’’ (Becker, 1981) strategy of investing in particular individuals with natural talent or as a result of serendipity. Traditionally such figures have then taken on responsibility for the social protection, safety nets or human capital investment of lineage and clan relatives. The definition of ‘‘relative’’ in some societies can extend widely, becoming almost indistinguishable from feudal paternalism. Thus nephews and nieces and even the children of cousins and in-laws may benefit from the patronage of successful family leaders in the large families of Bihar, India where the reliance upon the state or markets is precarious due to high levels of preferentialism and rent-seeking. Thus we observe family structures comprising poorer, client members obliged to accept the authority of family patrons in exchange for livelihoods support. As a result their prospects of personal graduation are remote, and likely to be reproduced for the following generation as well. The literature and drama of the region (including Bengal) is full of such references to oppressed poorer family members, essentially alienated within these combined but unequal relationships. Broader, perhaps more subtle, versions of these dependent relationships can be seen in

465

the mountains of Northern Pakistan (and also elsewhere in Pakistan, especially NWFP and the Sindh), with traditional elite families continuing to offer direct and indirect patronage (i.e., from employment to introductions) to ever widening circles of kin on an agnatic-affinal continuum. Indeed honor is gained and maintained by deploying oneÕs superior social and cultural resources in the service of others more weakly positioned. Such agency functions to recreate the structures of dependency and subservience to the point where the tradeoff between autonomy and poverty is complete––in other words, the worst condition for a poor person is not to be a client. Thus these ‘‘feudal’’ brokers provide employment or the links to it; they put in a good word (the institution of safarish); they enable the poor eventually to get to the front of the queue; they get files shifted (e.g., on a land case); they give or guarantee loans; they can procure legal services; they can get the rent-seeking police to withdraw an enquiry; they can mediate conflict, perhaps by paying off the aggrieved party; and so on. In return? Unswerving loyalty, even re-affirmed through kissing the back of the hand. There is not much room for resistance under such conditions. The risk of challenge is too great. (d) From diagonal to vertical: reconfiguring the moral political economy Such extremes of personalized patronage may amount to a moral political economy, but can it be relied upon indefinitely? The emphasis immediately above was upon traditional elites––but perhaps this is a fading aristocracy with only some successfully moving into a commercial–political nexus which enables a continuation of such style via newly meaningful services. Alongside, the poor are also increasingly threatened with a loss of opportunity to be a client, to be included even on adverse terms. As traditionally strong families (or subbranches) shift the basis of their livelihoods toward commercial and professional activities, so is the moral political economy reconfigured. The threat to poorer branches of the family or poorer parts of broader lineage and clan identities is that the successful among them will increasingly concentrate their resources upon immediate descendants. In effect the pattern of moral responsibility shifts from vertical and diagonal to vertical only. In other words indirect ‘‘diagonal’’ obligations (e.g., toward

466

WORLD DEVELOPMENT

nephews) recede in favor of only direct vertical ones to sons and perhaps daughters. Informants in Bihar have complained of this trend toward their declientelization, and emerging professional groups in North Pakistan have also complained about the difficulty for them of disengaging from these wider obligations in order to focus upon their nuclear family members. (e) Interlocked client status: credit dependency The poor can also be victims of interlocked client status. This can be most clearly observed in the context of credit relations. The rising microfinance industry worldwide represents an attempt to break into the interlocked relationships of informal moneylending. It is axiomatic that the poor need money; that their earnings and production are not adequate to their needs; that they need to borrow; that they have declining assets to deploy as collateral; and that therefore they have increasingly to commit labor and loyalty as a substitute for asset collateral. It is also invariably true that the basis of their income is highly seasonal, thus generating short-term liquidity crises. Their vulnerability is also reflected in their exposure to life cycle crises––especially illness and death of key earners, and the medical expenses entailed in trying postpone such events. But they also have problems in maintaining minimal community membership standards through rites de passage. Thus they have many welfare and apparently unproductive demands for funds (even though they are essential for the maintenance of social and cultural resources) which do not attract formal institutional lenders, including the so-called flexible microfinance institutions. If the poor cannot provide these needs through their own reciprocal arrangements (i.e., the chronic poor are rarely able to create or participate in ROSCAs and ASCAs, for example) then dependency upon patron moneylenders is inevitable. Remaining land assets, as collateral, become exhausted often via mortgage/high interest conditions which prevent any realistic prospect of reclaiming mortgaged out land. Then the price of such borrowing shifts toward committing in advance labor services at low wages even during high-wage seasons, or committing the unpaid labor of family members in forms of bonded labor (often this may be the domestic labor of females or children). Indeed in the Indian subcontinent, children

may find themselves obligated to remit the debt of their deceased parents through ongoing bonded labor services. Other ‘‘prices’’ are: highly exploitative tenancy arrangements; client loyalty in times of dispute and conflict; loss of rights to common property resources; loss of the ability to protect the honor of family females; loss of claim to ancestral residential sites as expanding patron lineage members seek to settle on prime sites. In short, an overall loss of autonomy and a deepening of alienation through the re-affirmation of dependency. (f) Migration: the degraded exit option When considering ‘‘induced loyalty’’ in this way, it is logical to consider exit options (on the assumption that ‘‘voice’’ is ruled out for many unless a strong NGO or other organized social movement has intervened locally in a sustained way). These usually take the form of migration, but which can also easily lead to degraded outcomes. The dependency of migrants upon mastaans in Bangladesh or their equivalents in other societies has been noted above. Migrants extracting themselves from adverse incorporation are also leaving behind an architecture of social and cultural resources essential for the pursuit of their livelihoods. How do they reconstruct these without re-entering the dependencies from which they have been escaping? Furthermore, migration is rarely the move in toto of the entire moral unit. Those left behind have to be cared for, and thus continue to rely upon the extant moral political economy. Indeed the prospective migrant in North Bihar villages has to borrow and commit the labor of family members to the moneylender as the condition for gaining the resources to migrate in the first place. Their destination in the agricultural labor markets of Punjab, a 1,000 miles away, may deliver relatively higher wages but at the price of high social and cultural vulnerability as their new employers can abuse them at little long-term cost to themselves (long working hours, beatings for low productivity and insubordination, entrapment into contractor-led labor gangs, high interest on wage advances which can thereby never be paid off). In addition, the business of sending remittances back to family members is precarious: formal ‘‘Post Office’’ methods are subject to rent-seeking officials; informal transfers also carry a ‘‘protection’’ tax against robbery. The migrant might

STAYING SECURE, STAYING POOR

then need to return to the home village. This happens frequently as destination opportunities fail to work out for various reasons. The prospect of needing to return reduces the value of the exit option as a counterweight to alienating loyalty. (g) Marriage, membership and access Since Hirschman, we are attuned to think in terms of exit, voice and loyalty. Schaffer and Lamb (1974) however substituted access for loyalty. Both terms, access and loyalty, can be understood as precursors of inclusion strategies as responses to social exclusion. It has been interesting to track family marriage strategies with reference to the search for security via inclusion. The rise in value of dowries in rural North Bihar relative to other prices in the local economy provides us with some insight. Poor peasant fathers, or elder brothers, are prepared to mortgage substantial proportions of their remaining landholding in order to finance effective levels of dowry. Of course, many factors affect the level of a dowry settlement, not least the gender composition of the brideÕs siblings and expectations of ‘‘return’’ on sons as bridegrooms. But leaving such complexity aside, I have come to understand that this apparently nonproductive, short-sighted, antimodernist tradeoff between dowry value and loss of cultivable land has indeed a clever rationale. Poorer families suffer from lack of access to scarce but essential agricultural inputs to increase the productivity of their land. The problem of access is not just an ability to meet price demands but an inability to operate in highly imperfect markets where price is not the only or even the most significant variable of access. This is reinforced where the local state is involved in rationing and licensing dealers. Social networks and connections are everything in this context. Poor peasant fathers do not have these networks and have to connect themselves to others who do. Thus they have to inflate their dowry offers in order to get their daughters married into stronger families, thereby establishing the affinal claims for the brideÕs natal family upon the networking of the bridegroomÕs family. They consider that increasing the productivity of their remaining land is more important than running the risk of never reclaiming the land mortgaged out to fund the dowry in the first place. In this way, they are in a position of paying for the patronage necessary for essential access to livelihood

467

resources. They have deliberately incorporated themselves, at a price. It is secondary to this argument that I have learned that such land is rarely reclaimed as the mortgage debt is invariably divided among brothers on the death of the father/grandfather, and these brothers have similar strategies to follow with their own daughters. (h) Security before graduation: Afghan commanders Finally, let us return to the extreme circumstances of sustained disruption and conflict which alas applies increasingly to many parts of the world. Although these conditions apply strongly to countries in sub-Saharan Africa, my illustration comes from a recent visit to Afghanistan. It is common in both Africa and Afghanistan to refer to ‘‘warlords.’’ Locally in Afghanistan, the more common term is ‘‘commander.’’ Badakhshan Province in the North East of Afghanistan was a minor invasion route for the Soviet forces in 1979, a significant area of mujadhin mobilization against the Soviet occupation, a key player in the post-Soviet conflict ending in a retreat of its Northern Alliance forces from Kabul in the early 1990s, and finally an area of ‘‘siege’’ as the Taliban forces tried and failed to enter and subordinate the province to its control. During all this time, and especially during the period of ‘‘siege’’ in the second half of the 1990s, the Northern border with Tajikistan was largely closed except for incoming food aid wheat convoys, and the border with Pakistan to the East was virtually impassable. These conditions produced tremendous stress on the local moral political economy. As part of a much larger story, a network of commanders emerged in the absence of a formal state, a collapse of stable currency, increasing scarcity of essential supplies, and deepening poverty. They managed a localized and arbitrary system of law and order, dispensed rough justice, distributed supplies and loans to distressed families in return for mortgaged land, cheap labor and young male recruits to the Northern Alliance militia. Under these conditions, the local peasants and pastoralists had no exit or voice options. Loyalty at any price was the only survival strategy. They had to accept the patronage of these commanders and their absolute authority. The prevailing conditions of insecurity over two decades effectively clientelized the society, where more local democratic, tribal conditions had obtained before. Even though the

468

WORLD DEVELOPMENT

surrounding conditions are now changing (maybe), the future remains highly discounted with local people having little faith (though much hope) in the return of a responsible state, stable currency, contract compliance, wider nontainted market options and security at personal and community levels. The commander system will therefore be slow to unravel even though many clients, in private conversation, would like to see it disappear. But by now, many of these clients are deeply in debt, have lost substantial amounts of cultivable land, are committed to debt–labor bondage, have surrendered community autonomy and even sometimes personal autonomy (women, especially young women, remain heavily veiled less for Islamic-cultural reasons than to avoid droit de seigneur). For them, the dangers of not being a client, of not being protected, of losing ‘‘membership’’ of the local commander led community are immense. Better to be with the devil you know––the Faustian bargain. Security at the price of graduation––individual or collective. Striking out on your own is just too risky.

10. CONCLUSION: SECURITY AS THE FAUSTIAN BARGAIN The determining condition for poor people anywhere in the world is uncertainty. Some societies perform better than others in mitigating this uncertainty for individuals and families under conditions where most people are enjoying a functional security, and are prepared to endorse rights and resources to the vulnerable poor among them. In so doing they are prepared to accept taxation and regulation as the collective compensator of individual venality, and the state is more or less trusted as an honest broker between selfishness and collective security. Thus, in such societies we observe welfare regimes which reduce in various ways the uncertainties of the market to provide for all citizens minimum conditions for reproduction. Such societies are a minority, however and are even themselves threatened by various forces of globalization which can leave their respective governments powerless to act as a compensator of last resort. Elsewhere in the world, destructive uncertainty is more pervasive. Such uncertainty derives from an overall instability and volatility in political, economic, social, and moral environments. Among the factors are: governments

that do not function, with a corresponding weak sense of citizenship; economies that experience price fluctuation in key commodities, including exports, which affect employment levels; hostile political economies, characterized by inequality, privilege and exploitation rather than abstract laws of supply and demand, reinforced rather than mitigated by forces of globalization; rapidly changing identities and loyalties comprising continuous fission and fusion. All these combine to destabilize the moral universe in terms of trust, loyalty and responsibility, entailing a world of cheats, free-riders and predators. Living in such conditions is risky, and the risk is compounded by uncertainty about the future. For those of us who are relatively secure in advanced economies, risk management partly involves preparing for the future because we think it secure enough to warrant such preparation, despite persistent threats of fluctuating financial markets and terrorism. So we save through pension plans, and take out various personal equity policies. We are also prepared to commit to long mortgages, as well as to shorter term borrowing for consumption items such as cars. In the present, we might acquire diversity of risk through dual income families; and we might continuously update our skills to connect to increasingly flexible labor markets. In this way, among others such as driving carefully or conforming to law in the expectation that others will do likewise, we gain our security. None of these strategies are so easy, or indeed rational, under conditions of high uncertainty. The poor, almost by definition, face more uncertainty than others. They have less control over relationships and events around them. They are obliged to live more in the present, and to discount the future. The value to them of the known present exceeds that of the unknown future. In other words, their timepreference behavior is for the present, which leaves them more vulnerable for the future. Moreover, risk management in the present involves loyalty to institutions and organizations that presently work and deliver livelihoods, whatever the longer term cost. Thus multiperiod games are established on the basis of patron-client dependencies, comprising a multiple web of transactions which limit the clientÕs room for maneuver since all ties could be threatened if one of them is allowed to collapse. Preparation for the future is continuously postponed for survival in the present-the Faustian bargain.

STAYING SECURE, STAYING POOR

469

NOTES 1. Most recently (as I write) in the northeast corner of Afghanistan.

10. Given that Chambers (1982) discussed ‘‘downward ratchets.’’

2. The first self-proclaimed socialist in Russian history, though a free thinking critic of Marx, see Isaiah BerlinÕs essays in Hardy & Kelly, 1978.

11. Though incidentally for the present argument, how social policy might assist in the improvement of that negotiation––i.e., not just ‘‘individual graduation’’ passports but ‘‘institutional reform’’ passports for whole societies, with different contents and meanings for different people and situations.

3. Alleviation is a different, lesser, albeit worthy agenda. 4. That is, people with weak social and cultural resources in their agency profile. We can, of course, note that some ethnic minorities and migrants are far from excluded and poor: e.g., South African whites remain highly privileged; Indian IT and health professionals are doing very well in the United States. This qualification reminds us that structural circumstances will determine whether distinctive groups of migrants, minorities or disabled are actually precluded from social action. The argument here is that the overall structural conditions of poor country political economies will in most cases link such ascribed characteristics to impaired capacity for social action. 5. Even in Latin America, where as a generalization only about 50% of the labor force is commodified (see Barrientos, 2001). 6. This expression refers to an institutional landscape comprising, at the domestic level, state, market, community and household; with corresponding international domains to offer an eight-box matrix. Each institutional domain is problematic for social agency, especially under conditions where political legitimacy and the social capital associated with widespread generalized commodity production are fragile (see Wood, 2001, 2003).

7. In the sense of structurally persistent. 8. Especially for the idiosyncratically, chronic poor. 9. Of course, we are all vulnerable to something: sudden loss of employment, for example. But secure people either have the skills to re-enter the labor market at reasonably similar levels, or they have been incorporated into other protective financial arrangements (redundancy entitlements, pensions, insurance and so on).

12. I am grateful to Sarah White at Bath for pointing this out. 13. In these approaches, eradication is understood as a more distant, less realizable goal. 14. Of course there are many different forms of ‘‘household,’’ characterized by the classic terminology of anthropology: patrilineal, patriarchal, patrifocal, virilocal, matrilineal, matrifocal, monogamy, polygamy, polyandry, and so on. There are some clear stylized differences between a South Asian patrifocal polygamous household sharing the same hearth and a subSaharan African patrilineal, polygamous, matrifocal extended family with scattered matrifocal hearths. 15. See Wood (1994) for descriptions of this process in villages in Bangladesh. 16. I have recently written a report (as yet unpublished) for the Aga Khan Rural Support program on informal social protection in the Northern Areas and Chitral district of Northern Pakistan: ‘‘AKRSP Poverty Policy For 2003-8 Phase: Issues, Strategies And Dilemmas’’ Mimeo 9.3.02. See also the thesis by LawsonMcDowall (2000) in which Chitral compares favorably with the more stratified rural Nepal in this respect. 17. A large, social mobilization NGO, currently organizing in excess of 80,000 landless groups (both menÕs and womenÕs), with a federated structure increasingly significant at district level. 18. i.e., the presumption that actorsÕ perceptions explain behavior, which give us knowledge about outcomes. 19. i.e., gaining insight through empathizing with the way choices and options appear to the subjects of our analysis (Wood, 2002).

470

WORLD DEVELOPMENT

REFERENCES Anderson, B. (1983). Imagined communities. London: Verso. Bailey, F. G. (1966). The peasant view of the bad life. In T. Shanin (Ed.), Peasants and peasant societies. London: Penguin. Barrientos, A. (2001). Welfare regimes in Latin America. Social Policy in Developing Countries Working Paper, University of Bath. Becker, G. (1981). A treatise on the family. Cambridge MA: Harvard University Press. Bevan, P. (2001). Dynamics of African (in)security regimes and some implications for global social policy. SPDC Working Paper, University of Bath. Carney, D. (Ed.). (1998). Sustainable rural livelihoods. London: Department for International Development. Chambers, R. (1982). Rural development: putting the last first. London: Longman. Collard, D. A. (2001). The generational bargain. International Journal of Social Welfare, 10(1), 54–65. De Waal, A. (1989). The famine that kills: Dafur, Sudan 1984–85. Oxford: Clarendon Press. Doyall, L., & Gough, I. (1991). A theory of human need. London: Macmillan. Esping-Andersen, G. (1990). Three worlds of welfare capitalism. Oxford: Polity Press. Esping-Andersen, G. (1999). Social foundations of postindustrial economies. Oxford: Oxford University Press. Fukuyama, F. (1995). Trust: the social virtues and the creation of prosperity. London: Penguin. Granovetter, M. (1985). Economic action and social structure: the problem of embeddeness. American Journal of Sociology, 91(3), 481–510. Hardy, H., & Kelly, A. (Eds.). (1978). Russian thinkers: a collection of BerlinÕs essays. London: Hogarth Press. Hirschman, A. O. (1970). Exit, voice and loyalty. Cambridge, MA: Harvard University Press. Jesmin, S., & Salway, S. (2000). Marriage among the urban poor of Dhaka: instability and uncertainty. Journal of International Development, 12(5), 689–705. JID (2000). The inter-generational bargain: special conference issues. Journal of International Development, 12(4) (See papers by Collard, McGregor Copestake and Wood, and Kabeer). Kanbur, R. (2001). Economic policy, distribution and poverty: the nature of disagreements. Working Paper, Cornell University, (based on presentation to Swedish Parliamentary Commission on Global Development, September 2000). Kearney, M. (1996). Reconceptualising the peasantry. Boulder, Co: Westview. Khan, M. I. A. (2000). Struggle for survival: networks and relationships in a Bangladesh Slum. PhD thesis, University of Bath, UK. Landell-Mills, P. (2001). Better governance for a better future: reforming public institutions in Bangladesh, Report for the National Institutional Review, World Bank, Dhaka.

Lawson-McDowall, B. (2000). Handshakes and smiles: the role of the social and symbolic resources in the management of a new common property, PhD thesis, University of Bath, UK. Lewis, D., Glaser, M., McGregor, J. A., White, S., & Wood, G. (1992). Going it alone: female-headed households in Bangladesh. CDS Occasional Paper, University of Bath, UK. Lipton, M. (1968). Theory of the optimising peasant. Journal of Development Studies, April. Loughhead, S., Mittal, O., & Wood, G. (2000). Urban poverty and vulnerability in India: DfIDÕs experiences from a social policy perspective. DfID, Mimeo. Moser, C. (1998). The asset vulnerability framework: reassessing urban poverty reduction strategies. World Development, 26(1), 1–19. Nussbaum, M. (2000). Women and human development. Cambridge: Cambridge University Press. Opel, A. E. O. (2000). The social context of labour markets in Dhaka slums. Journal of International Development, 12(5), 735–750. Platteau, J. (1991). The free market is not readily transferable: reflections on the links between market, social relations and moral norms. Paper presented to the 25th Jubilee of IDS, Sussex University. Putnam, R. (1993). Making democracy work: civic traditions in modern Italy. Princeton NJ: Princeton University Press. Roberts, B. (1978). Cities of peasants. London: Edward Arnold. Room, G. (2000). Trajectories of social exclusion: the wider context. In D. Gordon, & P. Townsend (Eds.), Breadline Europe: the measurement of poverty. The Policy Press. Schaffer, B. B., & Lamb, G. (1974). Exit, voice and access. Social Science Information, 13(6), 73– 90. Scott, J. C. (1976). The moral economy of the peasant: rebellion and subsistence in Southeast Asia. New Haven CT: Yale University Press. Sen, A. (1981). Poverty and famines. Oxford: Clarendon Press. Sen, A. (1982). Choice, welfare and measurement. Oxford: Basil Blackwell. Swift, J. (1989). Why are rural people vulnerable to famine? IDS Bulletin, 20(2), 9–15. UNDP (2000). Human development report 2000. Oxford: Oxford University Press. Wood, G. (1994). Bangladesh: whose ideas, whose interests? London: IT Publications. Wood, G. (2001). Desperately seeking security. Journal of International Development, 13(5), 523–534. Wood, G. (2002). At the crossroads: development agency in Afghanistan. Mimeo, Centre for Development Studies Report, University of Bath. Wood, G. (2003). Informal security regimes: embedding social policy in developing countries. In I. Gough, & G. Wood (Eds.), Insecurity and welfare regimes in developing countries. Cambridge: Cambridge University Press.

STAYING SECURE, STAYING POOR Wood, G., & Salway, S. (2000). Introduction: securing livelihoods in Dhaka slums. Journal of International Development, 12(5), 669–688.

471

World Bank (2000). World development report 2000–1: attacking poverty. NewYork: Oxford university Press.