1. Introduction

 

 

The 20^th century was the century of the labouring man. It was also the century when the working class scared rulers almost everywhere, was twice decimated by world conflagrations, trudged out in support of two competing socio-economic systems ostensibly dedicated to its interests, and ended the century by splintering in disarray. One cannot reconsider social policy and development without taking the history of labourism into account.

There is another, classical way of looking at what has happened. Recalling Karl Polanyi�s The Great Transformation, and risking the appearance of functionalism in presenting the process in brief, industrial capitalism has evolved through eras of stability and eras of upheaval. An era of stability is when the economy is embedded in society, that is, when the state legitimises and facilitates mechanisms of redistribution. The distributive system, and the state regulations associated with it, gradually over decades lead to "rigidities", which choke economic dynamism. Then there is an era of upheaval, when the economy is disembedded from society, when new forms of production and work organisation spread, and new forms of inequality and insecurity emerge. After a while, those become so great that they threaten the sustainability of the economic system. At that point, if economic progress is to resume, the state acts to re-embed the economy by introducing new forms of redistribution and means of ensuring enough economic security for those near the bottom of society to ensure legitimacy and sustainability.

With that imagery in mind, it is not too fanciful to see the 20th century as dominated by two labourist models by which states attempted to embed the economy -- state socialism and welfare state capitalism. Both, in different ways, made labour the fulcrum of their development strategy. In order to try to facilitate analysis of the challenge before us during "the second great transformation" (globalisation), this note briefly considers relevant characteristics of the welfare state development model, the factors that undermined it, and the subsequent contextual trends which, it is suggested, should shape a vision of distributive justice on which to build.
 
 

2. The Era of Statutory Regulation

 

 

"Labour is not a commodity."
ILO Philadelphia Declaration, 1944

The essence of the two models that competed during the middle decades of the last century was that they promoted labour security at the cost of constraining liberty, albeit in different ways. Both were initially based on the needs and aspirations of labouring man. Under both communism and welfare state capitalism, Full Employment was seen as the major instrumental goal. Adherents of both models tried to export their model to developing countries. The most extreme version of labourism was the Soviet Constitution, which declared that "he who does not labour shall not eat." We need not dwell on the characteristics of the state socialism model here. Although there have been several variants, the welfare state had a rather similar orientation, often linking entitlement to state-provided social protection to current or past performance of labour or willingness to perform it.

The essence of social and development policy in welfare state capitalism was the promotion of seven forms of labour security, summarised in Box 1.

Labour market security � Adequate employment opportunities, through state- guaranteed full employment;

Employment security � Protection against arbitrary dismissal, regulations on hiring and firing, imposition of costs on employers, etc.;

Job security � A niche designated as an occupation or "career", plus tolerance of demarcation practices, barriers to skill dilution, craft boundaries, job qualifications, restrictive practices, craft unions, etc.;

Work security � Protection against accidents and illness at work, through safety and health regulations, limits on working time, unsociable hours, night work for women, etc.;

Skill reproduction security � Widespread opportunities to gain and retain skills, through apprenticeships, employment training, etc.;

Income security � Protection of income through minimum wage machinery, wage indexation, comprehensive social security, progressive taxation, etc.;

Representation security � Protection of collective voice in the labour market, through independent trade unions and employer associations incorporated economically and politically into the state, with the right to strike, etc.

The economic, regulatory and social policies were intended to achieve progress on all these forms of security, and generally did so. The model, underpinned by Keynesian economics, was based on a set of premises. Technical change was expected to be steady and moderate, the production structure was expected to be stable or only changing slowly, with enough time for employers and workers to adjust, without major fluctuations in employment. The labour force was expected to be overwhelmingly in full-time, regular jobs, with men comprising the primary workforce. It was presumed that the economy of any country was essentially closed, in that trade was either a small part of the economy or where it was not mainly involved trade in competitive goods conducted between countries with similar labour rights, and trade in complementary goods according to an international division of labour in which developing countries mainly exported primary goods. The model implicitly depended on dividing the global economy into "three worlds" � industrialised, communist and the Third World � with a static international division of labour.

The regulatory framework was predominantly pro-collective and based on a mix of statutory regulations, epitomised by increasingly complex and comprehensive laws and procedures, coupled with voice regulation, exercised by legitimised trades unions and employer organisations, protected by measures to extend freedom of association, often backed up by neo-corporatist "tripartite" mechanisms. Because the system relied primarily on laws and regulations, one might call the period from about 1945 to 1975 the era of statutory regulation.

There was a rough distributional quid pro quo, in which, in return for leaving the managerial right to manage virtually intact, there was to be �redistribution from growth�, gradually shifting income in favour of workers and lower-income groups. Workers and unions struggled for what amounted to the "decommodification of labour", through raising the social income -- not just raising money wages but shifting a growing share of remuneration onto enterprise and state benefits, as well as giving workers� implicit income in the form of labour protection. This was the focus of the distributional struggle. There was never a distributional consensus, because unions and workers always pushed for a more rapid improvement in social income while employers tried to arrest that process. It was the employers who made most of the concessions.
 
 

3. The Era of Market Regulation

 

 

In the 1970s, while state socialism was easing into stagnation that brought it to collapse, the institutional arrangements of welfare state capitalism began to unravel. The model was eroded by several factors, which included:

• inflationary pressures and the oil crises,
• the changing international division of labour (which undermined the presumption of �competitive� trade between countries with similar labour rights and costs),
• the emergence of rational expectations theory and supply-side economics that undermined the Keynesian consensus among policymakers,
• the rise of libertarianism (which weakened the values of social solidarity that underpinned the redistributive welfare state),
• the slowdown in economic growth that disrupted the distributional quid pro quo,
• the rise in unemployment in much of the world,
• a widely perceived "fiscal crisis" of the state,
• a technological revolution, which ushered in something akin to a new ("fifth") Kondratief upswing, disrupting the stability of production and the norms of full-time stable employment.

 

 

For our purposes, it is not particularly helpful to look for a �smoking gun�, a primary factor. It is sufficient to note the consequences for the last quarter of the last century. These are the most relevant characteristics of what we might call the era of market regulation, which began in the mid-1970s and looks unlikely to last long in this century, because the economy has been dis-embedded from society, and because the resultant insecurities and inequalities are creating unsustainable pressures.

The emergence of supply-side economics in the 1970s and 1980s -- which underpinned IMF, World Bank and OECD programmes, and shaped the "structural adjustment" and "shock therapy" approaches -- reversed the Keynesian targeting of policy instruments. Whereas under Keynesianism, governments used macro-economic policy to produce so-called Full Employment (male), now macro-economic policy is expected to focus on the control of inflation and monetary movements, primarily by cutting public spending. Whereas micro-economic policy was to limit inflationary pressures, now it is expected to encourage employment, through removing institutional, behavioural and regulatory "rigidities". Whatever politicians say, governments have surrendered the pursuit of Full Employment through economic policy. This is epitomised by the chosen independence of central banks, and delegation of authority over monetary policy to non-elected bodies. Orthodox economic policy places responsibility for unemployment on the behaviour of workers and firms and on "regulations". Meanwhile, governments have been trying to cut "budgetary deficits" and public spending. The public sector has ceased to be "employer of last resort", and is regarded as "crowding out" private investment, employment and growth. Attempts to cut public spending include measures to make social protection more selective, conditional and privatised.

There has been a re-orientation of labour regulations. There is no such thing as a deregulated labour market. However, whereas in the previous era, labour regulations were predominantly protective of workers, and were both pro-collective and socially redistributive, in the succeeding era the orthodox view was that regulations are justifiable only if they promote economic growth, following the Chicago school of law and economics. In general, there has been a tendency to use regulations to increase the role of "market forces", which is why it is apt to call this the era of market regulation. The shift has been characterised by increasing emphasis placed on fiscal regulations and pro-individualistic statutory regulations, with a campaign against protective regulations, which have been depicted as labour market "rigidities" and the source of non-wage labour costs undermining national competitiveness. One of the ironies of the 1990s was that many of the securities that had been regarded as desirable objectives of socio-economic (development) policy came to be regarded as undesirable obstacles to be overcome.

In the 1990s, no longer was there competition to sell welfare state capitalism and state socialism in developing countries. In their place, there was the almost hegemonic imposition of the ubiquitous "Washington consensus". This offered a model consisting of eleven main elements, with more being added as its �success� spread. Briefly, they are trade liberalisation, financial liberalisation, privatisation, "deregulation", foreign capital liberalisation (elimination of barriers to FDI), secure property rights, unified and competitive exchange rates, diminished public spending (fiscal discipline), public expenditure switching (to health, schooling and infrastructure), tax reform (broadening the tax base, cutting marginal tax rates, less progressive tax), and a "social safety net" (selective state transfers for the needy). A twelfth element, expressed in World Bank and IMF (and OECD) reports, is labour market flexibility, by which is meant decentralised labour relations coupled with cutbacks in protective and pro-collective regulations.

When the former Soviet bloc countries became "countries in transition" and part of a broad group of "emerging markets", they were subject to the same set of prescriptions, albeit repackaged and with additional elements under the title of "shock therapy". The social and economic costs of the attempted transformations of these countries have been enormous. Although there was some economic recovery in central Europe in the late 1990s, everywhere poverty, inequality and lack of entitlement to social protection spread.

 
1. Contextual Outcomes: Globalisation, Informalisation and Flexibility

Among the outcomes of the era of market regulation to keep in mind is the evidence that poverty and inequality have increased in much of the world. But there are a few other contextual trends that may be highlighted for our purposes.

We may as well begin with "globalisation". This has come to mean greater openness of national economies, in which capital mobility has increased (size, volatility, speed and, perhaps most importantly, potential mobility), in which national economic and social policy has been more constrained by international pressures, and in which the balance of bargaining strength has shifted from �labour� to �capital�. Globalisation is seen as threatening the viability of certain forms of welfare state and regulatory system, and has contributed -- to a smaller or greater degree according to one�s perception � to the growth of economic inequality and labour insecurity. This may be a reasonable assessment, although there are still sceptics. But there is one sense that must be resisted � globalisation has been used by some commentators to tell proponents of redistributive policies that there is no choice, except to adapt and to accept permanently greater inequality.

Linked to globalisation has been the global technological revolution, the most relevant points of which are:

Decentralisation and multi-site production is more feasible than when �heavy� technology predominated ;

More managerial options in the way firms can organise labour arrangements, payment systems, and so on;

The labour market, organization and technological changes have facilitated the feminisation of employment.

New technologies, combined with globalisation, have contributed to the global growth of flexible systems of production and flexible labour markets.

The growth of flexible labour systems has been pervasive. One can identify seven main forms, which give a context for assessing mechanisms of social protection and security:

Organisational flexibility � more turnover of firms, more use of sub-contracting and production �chains�, and a tendency to contract the employment function;

Numerical flexibility � more use of external labour, such as contract workers, outworkers, homeworkers, agency labour, temporary workers, and teleworkers;

Functional flexibility � greater change in work tasks, job rotation, and skill;

Job structure flexibility � more changes of restructuring jobs and occupational structures in production;

Working time flexibility � more continuous working, flexible hours, etc.;

Wage system flexibility � a shift from fixed to flexible wages, monetisation of remuneration, greater use of bonuses, etc.;

Labour force flexibility � less attachment to sectors, companies or occupational groups, erosion of �collective labour�, and greater tendency for workers to move in and out of the labour market and labour force.

• Those locked out of the �modern�, �formal� part of the economy in petty forms of production, including smallholder farmers and other rural workers, and in such peri-urban activities as shops, repair sheds, and the like. They are outside the state social protection system, without realistic prospect of being protected by social insurance, mainstream labour regulations or representative bodies such as trade unions. Their insecurity would be endemic, except that they are dependent on family and kinship communities, and any local "grass roots" agencies that emerge to represent their interests.

• Those protected in the core of the economy, those living off capital or large-scale landholdings, the �salariat� in and around central government, and those employed in large-scale firms, often linked to the international economy. These are highly secure by the standard of those around them, and often by the standard of most groups in more affluent countries.

 
• �Flexiworkers�, consisting of millions of people periodically doing casual labour, including migrants, labour �circulants� (usually moving between rural and urban areas for short periods), and many others in itinerant statuses.

 
• Those involved in bare survival activities, on the edge of legality and socially stigmatised, doing activities such as begging, prostitution or streetside marketing.

What is clear is that the labour market presumptions that underlay the labourist models of the 20^th century have been shown to be invalid. It had long been anticipated that the vast majority of workers would be in stable, full-time wage employment, able to pay insurance contributions or have them paid for them, in order to give them social protection against the "temporary interruptions in earnings power". In developing countries, it was expected that gradually that model would emerge, as surplus labour was absorbed into the "modern" or "formal" sector. In a few countries that has happened to some extent, but mainly it has not, and there is no prospect that it will do so in the near or medium-term future. Not only are �modern� firms not resorting to the standard employment model, but the state sector that was expected to be, and for a long time was, a major source of �formal� employment has been shrinking or stagnant in much of the world. And in industrialised countries, there has been a trend away from standard, regular, protected full-time employment, such that now in some countries only a minority of the employed is in such jobs.

A final contextual point is that the intellectual changes that shaped economic, social and labour market policy in the 1980s and 1990s may be running out of steam. A feature of recent years, culminating in the imagery of globalisation, is that for the first time in history we have been offered a model of economic and social development that is supposed to apply to all countries. In the 19^th century, and for part of the 20^th, colonial powers divided countries into �civilised� and �uncivilised�, with the �natives� expected to provide primary goods. Later �under-developed countries� were expected to pursue policies of import-substitution industrialisation. Then there was the era in which the two labourist models were presented as alternative models for developing countries to adopt, usually under external tutorage. But since the 1980s, the Chicago school � or, as it evolved, the Washington consensus � has been the only serious model on offer.

One predicts that this hegemony will weaken, and it will do so precisely because the inequalities and insecurities are becoming too extensive and severe. There is no reason to agree with people like Sachs, who claimed the upheavals of 1997 were "the end of an era". However, the disembeddedness of the economic system may have become so excessive that it is threatening the stability of national economies and the process of globalisation.

For two decades, the Chicago school of law and economics and libertarianism combined to shape mainstream thinking, but in the late 1990s their advocates� confidence was shaken. Here is some good news. The World Bank�s Chief Economist, Jo Stiglitz, resigned at the end of 1999, having said that the Washington consensus is bunk and wrong. He deserves to be saluted. There is an awkward problem. For over two decades, the World Bank has devoted billions of dollars to induce countries all over the world to adopt the policies of that consensus. If there is an attempt to re-arrange "the financial architecture" (one of the latest euphemisms), it is not too cynical to predict that the restructuring will be devoted mainly to stabilising capital markets, not to addressing the social and distributional concerns of most people. For the ILO and UNRISD, a strategic question is whether we should engage in more open debate with the international financial agencies on principles of human security and on policies to promote it, and if so how that could be done.

Beyond these minor internal considerations, it is perhaps symptomatic of the intellectual crisis that the cacophony that goes for international debate is replete with mischievous euphemisms. In most countries the package of policies that has been "globalised" has produced �stabilisation� without stability � epitomised by the periodic crises that break out somewhere in the world, such as Mexico in 1994, east Asia in 1997, Brazil in 1999, and Russia every six months. There has been "deregulation" that has involved many new regulations, and there have been �safety nets� without safety, as millions more people are pushed into poverty and as inequalities have grown. No wonder Stiglitz and others are worried. Whether any Bank is the appropriate institution to articulate or implement it, we may indeed need a "new development paradigm", one based on reducing insecurity and inequality, and re-embedding social policy in the economy.
 

1. Labour Insecurity and "Social Safety Nets"

Of course, the majority of people in the low-income countries of Africa, Asia and Latin America have long suffered from chronic economic insecurity � the slightest mishap could push many into an abyss. Yet we also know that a country�s income level is not the only factor that produces collective or individual insecurity. Labour and social relations of production and distribution play a major part in ensuring entitlements (long known before Sen�s study of famines), and the social structure will usually determine whether a community survives an economic or ecological shock.

There are reasons for thinking that the fragility of security in low-income developing countries has been increased, because of the erosion of traditional kinship systems of social support, as well as by mass urbanisation and the economic restructuring. But all forms of labour security have been eroded around the world. Briefly:

• Labour market insecurity has grown almost globally, with much higher unemployment, slower rates of employment growth and higher �labour slack�.
• Employment insecurity is high and rising, with growing proportions of those in the labour force having insecure employment statuses and with more workers lacking employment protection.
• Work insecurity has become greater, due to more people being in work statuses without coverage by protective institutions and regulations.
• Job insecurity has worsened, with more workers having to switch jobs and learn new tricks of working.
• Skill reproduction insecurity is considerable, in part because skills become obsolescent more quickly and because few workers are receiving career skills.
• Income insecurity is greater for those in employment, due to flexible wages and so on, and for those outside employment, due to explicit and implicit disentitlement to benefits.
• Representation insecurity is growing due to de-unionisation, erosion of �tripartite� institutions and the changing character of collective bargaining.

On labour relations, decentralisation is part of a trend to employment and wage �bargains� made at the individual or small-unit level. Globalisation, with the shift in power and influence from the representatives of labour to capital, and with more labour flexibility, has contributed to this, and it is a global phenomenon. From the days of comprehensive �incomes policy� in the 1960s and 1970s, the vision of national-level tripartite bargaining has faded, only kept going in a few places such as the Netherlands, where the underlying rationale has been concession bargaining by unions.

Sectoral bargaining has gone into decline, because of the ability of corporations to switch production more easily and to indulge in whipsaw bargaining, and because workers� allegiance to sectors is diminishing (and the location of firms in specific sectors has also become more flexible). On the workers� side, lack of sectoral identity partly reflects the decline in job security and employment security, as well as the plunging decline in representation security, measured by the share of workers belonging to trade unions. In most developing countries, any trend that had existed towards improvement has gone into reverse.

Occupational bargaining has long been divisive, since Taylorism largely destroyed the base of craft unionism, while bodies to protect �professionals� have been mechanisms for guarding the incomes and privileges of the privileged. The growing sphere of bargaining is at the levels of enterprise and firm, and increasingly at the individual level of the worker and employer. A consequence is that company unions and the like have been boosted, which have been instrumental in intensifying labour market fragmentation. The weaker elements are less likely to be protected by collective voice, while firms are more able to splinter their workforces into groups with dissimilar benefits, social entitlements and social protection.

The ongoing changes in the character of national systems of social protection are complex, but for our purposes the principal changes underpinning the mischievous euphemism of social safety nets are:

• increased selectivity of state transfers,
• multi-tierism in modes of provision of social protection,
• partial privatisation of social policy.

For our purposes, the main significance of these trends in the character of labour markets, labour relations and social protection is that they seem to be accentuating the fragmentation of societies into groups with sharply different opportunities and access to elements of social income.
 

1. The Concept of Social Income

 

 

If we talk of a need for income security, we need a reasonable idea of what counts as income, as well as what counts as security. Does the following help? Would the proposed disaggregation facilitate analysis, policy consideration and data gathering?

The suggestion is that we think in terms of social income. In any society, every individual has some source of income, or dies. The total may be inadequate or grotesquely large. There are various sources, and the composition determines not just the level but also the overall security of income.

Basically, any individual in any society has up to five sources of income, which together constitute the person�s social income. This may be defined as follows:

SI = W + CB + EB + SB + PB

where SI is the individual�s total social income, W is the money wage or income received from work, CB is the value of benefits or support provided by the family, kin or the local community, EB is the amount of benefits provided by the enterprise in which the person might be working, SB is the value of state benefits provided, in terms of insurance benefits or other transfers, including subsidies paid to workers or through firms to them, and PB is private income benefits, gained through investment, including private social protection.

We can disaggregate the elements as follows:

SI = (W_b + W_f ) + (FT + LT) + (NWB + IB) + (C + IS + D) + PB

where W_b is the base or fixed wage, W_f is the flexible part of the wage (bonuses, etc.), FT are family transfers, LT are local community transfers, including any income from charity, non-governmental organisations, etc., NWB are non-wage benefits provided by firms to their workers, IB are contingency, insurance-type benefits provided by firms to their workers, C are universal state benefits (citizenship rights), IS are insurance-based income transfers from the state in case of contingency needs, and D are discretionary, means-tested transfers from the state.

We do not have data on the distribution of social income, or of the weights of the several components. However, based on anecdotal and scattered data, we can guesstimate the modal patterns in the regions of the world during the era of statutory regulation. In a stylised way, we can guess that relative to other regions and as a proportion of total personal income, in the post-1945 era the average values of the components of the social income, as expressed in the first identity, were as follows:

Table 1: Relative Structures of Social Income in Era of Statutory Regulation, by Region

W EB SB PB CB

Africa Medium Low Low Low High

Western Europe Medium Medium High Medium Low

Eastern Europe Low High Low Low Low

North America Medium Medium Low High Low

Latin America Medium Low Low Medium Medium

South Asia Low Medium Low Low Medium

South-East Asia Low Medium Low Low High

For example, in western Europe, for the average person the share of total social income accounted for by the money wage was about average for the world, whereas the share coming from state benefits was high by world standards and the share coming in the form of informal transfers from the local community was low. In eastern Europe, the money wage was a small part of social income, while enterprise benefits comprised a very high share. In east Asia, money wages were not only low in absolute terms but were a small share of social income, while community transfers comprised a high share � wage workers were subsidised by their predominantly rural kinship communities.

When commentators said that "labour is not a commodity" and referred to "de-commodification" in the middle decades of the 20^th century, they meant that a growing proportion of social income was coming from non-wage, non-monetary elements. By contrast, although one would have to document it, it is not too fanciful to depict recent trends as increasing "labour re-commodification", since for many groups in various ways there has been a shift away from non-wage, non-monetary forms of social income to money wages.

Perhaps the most important ways have been the erosion in informal social support systems, particularly in developing and �transition� countries, and removal or loss of entitlement to enterprise and state benefits. But there are other ways too, often harder to evaluate. Suppose, to give an illustrative example, a man was receiving $100 as a wage, and had a long-term secure employment contract, guaranteed healthcare, subsidised canteen food and membership of a group pension scheme. The total value might be $200. How do we measure the value to the worker of the employment guarantee? One imagines many people would be prepared to sacrifice, say, 10% of their income to retain such a guarantee. If an employer takes the guarantee away and gives the worker a 10% pay rise, our statistics would show wages rising by 10%, which is misleading because all that has happened is that the structure of social income has changed.

Although we do not have good data to demonstrate it, something like this seems to have been happening all over the world � except that for many there has been no compensation for loss of security or the hope that they could obtain it has diminished. Another example is where there is a shift from SB to PB, or a shift to IB in which workers have to pay a higher share of the costs to secure the benefit. In such cases, the cost might account for proportionately more of the social income of the low-paid, insecure worker, and the actual cost might be higher as well.

In short, under the impact of economic liberalisation and "globalisation", the structure of social income has been changing. Overall, one may hypothesise that the median changes taking place in the designated regions are as in Table 2, where a plus sign implies a rise in the relative contribution to individual income, a minus sign implies the opposite. Where both a plus and a minus sign are given, one surmises that part of the growth of income differentiation is due to a shift in one direction for some groups and in the opposite for others. Table 2 suggests that in most parts of the world there has been a shift to money wages and a shrinkage in the share provided by state benefits and services. Perhaps most significantly � and this is a guesstimate rather than a statement based on substantial statistical information � there is increasing reliance on private provision (personal investment and saving) and community support (voluntary provision), or an increased need for those sources to fill voids opened by diminishing public provision.

However, the disaggregation of Table 2 is not as revealing as when we consider what has been happening to the sub-components of social income and to distinctive socio-economic groups. Before considering this, we need to have an image of the labour fragmentation that has been taking place internationally.

Table 2: Trends in Components of Social Income since 1970s, by Region

W EB SB PB CB

Africa -/+ -/+ - 0 0

Western Europe +/- +/- - + +

Eastern Europe + - +/- + +

North America +/- +/- - + +

Latin America + + - + +

South Asia 0 ? - +? 0

South-East Asia + + 0 + -
 
 

2. Fragmentation: The Social Income under Strain

 

 

It is of limited use to consider social protection and human security as if there were no social groups or �classes�. Our conventional labour statistics are not very helpful. For instance, there are no peasants in international statistics. And we have no information on classifications based on the range of income sources and controls by which patterns of economic security or insecurity are reproduced or undermined.

For our immediate purposes, we might find it useful to consider the following image of social fragmentation taking place internationally. The point of any such exercise is to identify groups that have distinctive sets of entitlements and patterns of social protection and security, and that are likely to have a particular attitude to various forms of social protection.

A feature of the fragmentation is that growing numbers of people are detaching themselves or being detached from mainstream national regulatory and protective systems. It is presented in descending order based on average social income.

* The elite. At the zenith of the globalising economy is an elite, consisting of a tiny minority of absurdly rich and high-earning people, whose impact is out of all proportion to their number. Some of these individuals have long since reached the stage of seeing their incomes rising almost exponentially. They are global citizens. Expanding the stratum down to multi-millionaires, the relevant points are that they are detached from national regulatory and social security systems, neither needing nor contributing to them, either psychologically -- not feeling committed to their maintenance or improvement -- or politically. The elite has very strong income security, and whatever they need of other forms of security. Their biggest danger is hubris, and being caught in criminality.

* Proficians. These are the new craftsmen of the global flexible economy. As the name implies, they are a mix of professional and technician, mostly working as consultants or in short-term employment contracts. They operate in a climate of insecurity, but are well compensated for this. Perhaps their main form of insecurity is work insecurity, epitomised by the frenzied pace of their erratic work schedules, stress and burn-out. They are often able to evade or avoid taxation, and are at least partially detached from state-based social protection systems.

* The salariat. This consists of salaried employees, including those working in civil services, large corporations, para-statials and other bureaucracies. They have a high degree of labour security, but probably suffer from some job and skill reproduction insecurity because they may be moved around and/or gain promotion in their enterprises by leaving technical skills behind them. Because of their reasonably high incomes and a tendency to identify with managements, employers and the elite and profician strata, members of the salariat typically feel detached from the state social protection system, seeing their future and income security in terms mainly of private insurance benefits and earnings from judicious investment.

* Core workers. These are the bulwark of what those of us with long memories used to call the working class. Welfare states were created to serve the needs of core workers, those in full-time, regular, typically unionised jobs, usually with manual skills. During the era of statutory regulation, it was implicitly presumed that these workers represented the norm and that a majority of workers in all countries would eventually belong to this stratum. The larger the proportion of people belonging to it, the more people in a position to support and benefit from the mainstream, insurance-based social protection system and the mainstream regulatory system.

The trouble is that although the legitimacy of a redistributive welfare state depended on core workers, they never comprised a majority of the economically active in most countries and since the 1970s have been shrinking. This is not just a reflection of "de-industrialisation" in industrialised countries, and the associated dispersion of manufacturing wage labour around the world. It is also because of various forms of labour market flexibility. Core workers traditionally benefited from most forms of labour security, but with the growth of wage system flexibility a growing proportion of their income has come in insecure form. They also suffer from increasing job insecurity and employment insecurity, while their unions have been weakened almost everywhere. Above all, with core workers dwindling in numbers and not expected to grow, their agenda has lacked legitimacy.

* Flexiworkers. These comprise a disparate group of people in non-regular work statuses, including casual workers, outworkers, sub-contracted and contractor labour, agency workers and domestic workers. Their common characteristic is labour insecurity in almost all respects. In the era of statutory regulation and welfare state capitalism it was presumed that these "informal" forms of employment would decline as economies developed. In recent years, they have appeared to be the future. Not only have the number of people trapped in petty activities in rural and peri-urban areas grown, but the flexible labour processes have boosted many other forms. Among the associated trends, growing proportions of labour forces have lacked entitlement to mainstream statutory protection and have been disentitled to social transfers.

* The unemployed. The number has risen extraordinarily in the era of market regulation. They suffer from labour market insecurity, given higher unemployment, and from greater income insecurity than in the past, because the level of benefits has been cut, duration of entitlement has been shortened, and conditions for entitlement have been tightened.

* The detached. This is a growing minority of the population in many countries, cut off from mainstream state benefits, lingering in poverty, anomic and threatening those above them in the income spectrum simply because others fear falling into their ranks. In recent years, politicians have been inclined to treat these, many of whom are victims of economic liberalisation, as in need of "re-integration". They linger in the streets, in bus and train stations, in city parks. They make those above them in the social order feel uncomfortable or smug, depending on where they fit. The detached represent fear. And it is fear that induces concessions from the near poor � the ultimate tool of inequality.

If one divides societies (and the international economy) into these seven fragments, one sees that the top three strata are increasingly detaching themselves from state-based social protection, while the bottom three are increasingly being detached by explicit and implicit disentitlement to its benefits and services. Although one may choose a different way of stratification, the analytical device may also help us to picture the growing inequality of social income and deterioration of economic security.

Table 3 is an interpretation of what anecdotal evidence and reports suggest are the sources of income received by the separate strata, or the sources on which each group relies for survival. The asterisks indicate the main sources of income for the higher-income strata; the blanks imply that the source is not applicable, or that no guess on aggregate trend seems reasonable. The table�s final row indicates what seems to have been the global trend for the source of income specified by the column. Thus, the base wage has tended to decline as a part of social income, the flexible part of the wage has been rising, income from private savings and investment has been rising sharply, and so on. The +/- sign means the source has risen for higher-income groups, fallen for others.

Table 3: Sources of Social Income, by Socio-Economic Status
 

Sources Stratum	Wb	Wf	FT	LT	NWB	IB	C	Is	D	PB/K
Elite										***
Proficians		*				+	+			*
Salariat	*	+			*	*	+	+		+
Core	+	+			+	+	+	+
Flexiworkers		+	+	+			+	(+)
Unemployed			+	+			+	(+)	+
Detached				+			+		+
Global trend	�	+		+	+/�	+/�	�	�	+	*

Table 4 complements Table 3, in that it indicates how the seven strata experience various forms of security. Thus, for example, proficians have high levels of income security and skill reproduction security, being in control of their own activities, but no employment or job security. Core workers have some employment security (although diminishing), relatively high work security and reasonable entitlement to enterprise and other benefits.

Table 4: Forms of Labour Security, by Socio-Economic Status
 

Security  Stratum	Labour market	Employ-ment	Job	Work	Skill Reprod.	Income	Benefits	Represen-tation
Elite	+	+	+	+	+	+	+	0
Proficians	(+)	�	�	�	+	+	0	�
Salariat	+	+	(+)	+	(+)	+	+	0
Core	0	(+)	0	+	0	0	+	+
Flexiworkers	�	�	�	�	�	�	�	�
Unemployed	�	�	�	�	�	�	�	�
Detached	�	�	�	�	�	�	�	�

Note: +, above average; 0, average or not applicable; -, below average.

Although one may quibble with Tables 3 and 4, they suggest a way of interpreting what is happening to income and income security. The socio-economic fragmentation has produced a situation in which those in -- or identifying with -- the top three strata feel increasingly detached from the mainstream state social protection system. They are thus less inclined to defend its principles of social solidarity, while the bottom three strata feel deprived, detached by disentitlement to the benefits long offered to core workers, to whose ranks they had aspired. To them, there is no solidarity on offer, and for them, to talk about social solidarity would sound like a sick joke. The pervasive detachment and lack of social solidarity have contributed to the loss of legitimation of the welfare state.

An image of fragmentation coupled with the classifications of informal activities and forms of flexibility outlined earlier may be useful for assessing the relevance of policies to give social protection and security, and for assessing the limitations and apparent lack of general appeal of systems of social protection promoted in the 20^th century.
 

1. Welfare State "Crises"

• The fiscal crisis. This has attracted most attention, and has several components. First, as welfare states expanded and the rationale for state-based social transfers and services was developed, the range of categories and of conditions identified for transfers and services was steadily extended. The share of GNP going on social protection in industrialised countries rose, seemingly remorselessly. Then labour market changes have eroded the contributions base and resulted in more people becoming �dependent� on state transfers. For these reasons among others, governments all over the world have been trying to cut back, notably by cutting the level of benefits, increasing resort to �targeting� and �selectivity�, increasing conditions for entitlement to benefit, shortening the duration of entitlement to some benefits, and so on. The "social safety net" has become a euphemism for cutting back on the universal, comprehensive welfare system.

• The legitimation crisis. This arises from erosion of social solidarity, or the diminished realism of such a claim. Intellectually, we may subscribe to the desirability of social solidarity � and no doubt have done their fair share to support it. But the awkward reality is that with flexible labour markets and socio-economic fragmentation, those in higher strata are likely to feel they have a very low probability of needing or qualifying for state transfers, while many in other strata are likely to be disinclined to commit themselves to the state system, given the low value, uncertain entitlement and means-tested or stigmatising character of many transfers. Some political scientists and economists have argued that legitimacy would only be achieved by adherence to the "reciprocity principle" � imposing obligations on claimants to legitimise conditional forms of social protection. But it is hard to envisage a sense of equivalence to make that appealing. A legitimation dilemma also arises over democratic governance of social protection. Around what principles could a consensus be built and sustained? Would a stable coalition emerge that would give minority groups a secure set of entitlements? What would discourage opportunistic politicians from eroding benefits and rights for one minority interest after another? In short, the politics of income security are surely more problematical in societies characterised by flexible labour markets and socio-economic fragmentation.

• The moral crisis. In recent years, social (or public) policy reformers have been drawn down an old road. Dubiously, the poor are being divided into three groups:

• "The deserving poor", to be offered a residual social safety net, through means-tested basic benefits (including basic first-tier pensions);
• "The undeserving poor", offered the carrot and stick of �workfare� and/or conditional low-level transfers;
• "The transgressing poor" (those who fail and resort to unsociable behaviour), offered the stern state prepared to uphold the law in keeping public order.

• The work dilemma. This is potentially the Achilles heel for the paternalists who have dominated social and labour market policy in recent years. Gradually, we are moving into an era in which all forms of work and work-related activity are being recognised as legitimate and as work, including caring for one�s dependent relatives or neighbours, learning new skills and voluntary community or charitable work, whether in private voluntary organisations or on an individual basis. This poses ethical and administrative problems for social protection policy, particularly for schemes based on labour conditionality. How do you decide whose work should count as work and whose should not? What criteria should be used? Who should be allowed to make such decisions, if anybody? These questions are particularly pertinent for those who believe in gender equality. And the relevance of the need to reconceptualise work � and to refine �labour statistics� � is made even more urgent in the context of aging. The trend towards flexible working lives will further complicate the matter.

• The social dumping dilemma. This issue has generated more heat than light so far. The suggestion is that because of regional or global economic integration governments are lowering social protection standards to deter firms moving abroad, to attract firms to relocate to their country or region, and/or to discourage labour migration to their country or region. There is also the related tendency to find ways of lowering non-wage labour costs in one country to try to make conditions more "competitive" with other countries. In the longer term, it is likely that the question of tax credits designed to facilitate low-wage, "unskilled" labour will be brought into the international debate on unfair trading practices, as part of some future round of the World Trade Organisation.

• The coverage crisis. Ultimately, social protection systems have lurched into the biggest crisis of all. They do no offer the prospect of reasonable income security and social protection for the poor and near-poor. There are a few countries where that does not apply, but they are a shrinking minority. At one extreme are most of the countries of Africa, where a tiny percentage of the population is covered by state-based schemes of any sort. In the middle are the �transition� countries where often a growing majority is denied any realistic prospect of decent coverage. And in industrialised welfare states a growing minority is not covered, or fear that they are not. A recent survey in the USA found that more young people believed in UFOs (unidentified flying objects) than that they would be covered by social security later in life. Diminishing coverage, and the fear of uncertain coverage, seems to be a global phenomenon.

1. Selected Policy Options for Income Security

The Programme just mentioned is exploring the advantages and disadvantages of a range of policy options proposed to extend basic social protection to groups excluded by conventional systems and to provide minimal income security for all, consistent with the commitment of all ILO member countries to these principles, as enshrined in the Universal Declaration of Human Rights and the International Covenant of Cultural, Social and Economic Rights. In doing so, it may be useful to consider each option in terms of the following evaluative questions:

• Exclusion. Does the policy or scheme help to extend protection to groups actually excluded at present?

• Detachment. Does it help to reduce the process of detachment from mainstream economic activities, both at the bottom and at the top?

• Inadequacy. Does it offer individuals and communities a higher probability of obtaining an income that is adequate for their basic needs?

• Uncertainty. Does it offer the poor and near-poor a more dependable source of basic security?

• Efficiency. Does the scheme operate with reasonably low costs of administration?

• Disincentive to work. Does the scheme weaken or strengthen the incentive to work, and to save and invest?

• Inequality. Does the scheme offer the prospect of reducing economic inequalities, and does it pass the Rawlsian Difference Principle test?

• Deadweight. Does the policy involve public expenditure that is unnecessary because the action or something like it would have taken place anyhow?

• Displacement. Does the scheme threaten to drive out other systems of protection or production?

• Substitution. Does the policy offer the likelihood that one group would benefit at the cost of another?

• Liberty erosion. Does the policy reduce the freedom of choice and freedom of action of potential target groups or beneficiaries?

• Sustainability. Is the policy sustainable -- financially, politically and in terms of public legitimacy?

• Extension of social insurance-- including "care" insurance

• Extension of means-tested social assistance
• Categorical targeting
• Introduction or strengthening of a minimum wage
• Public works
• Consumption subsidies
• Wage subsidies
• Employment subsidies
• Workfare and welfare-to-work schemes, to "re-integrate the unintegrated"
• Earned income tax credits � negative income tax-- including childcare tax credits

• Participation income
• Citizenship income, partial or full
• Micro-insurance schemes, based on communal voluntarism
• Delegating responsibility for social protection to private voluntary agencies.

1. Concluding Points

Recognising that something has to be done about the poor and the "losers" in the market economy, they offer a minimalist (residual) "social safety net" for "the deserving poor", which the state should provide through targeted, means-tested social assistance, coupled with fiscal and moral incentives to charitable disbursements from the wealthy. Faith-based organizations will be boosted everywhere, aided by foreign aid and much lusty rhetoric. For the "undeserving poor", they offer the carrot and stick of "workfare", "public works", and conditional, low-level state transfers. For the "transgressing poor", they offer a stern State prepared to uphold the law in keeping public order.

On the left (if that is the right word), the rump of old �labourists� and social democrats are coalescing under a delightful euphemism � "Third Wayism". It seems to offer constrained labour market security with constrained liberty. It seems to lead to a more social, or moral ("responsible"), variant of libertarianism. It accepts and welcomes globalisation, in that it believes in economic liberalisation, low taxation on capital and on higher income groups, and a �need� for subsidies for capital. It favours some statutory protective regulation to strengthen labour rights, and believes that inequality can be redressed by raising employment and by increasing the "employability" of everybody.

It too seems to divide the poor into the deserving-undeserving-transgressing categories. It advocates "welfare-to-work" schemes to integrate the victims of economic dynamism and labour market flexibility, tends to support wage and employment subsidies, tax credits, such as the EITC, and "welfare pluralism", in which social protection and social services are converted into "multi-tierism". Ultimately, Third Wayism is suffused with the �new paternalism� that guided US welfare reform in the mid-1990s, or the desire to have the state intervene to reduce "dependency", in deciding what is proper behaviour and what is improper.

In the light of these conflicting yet overlapping models, and in the context of globalisation, perhaps we should reconsider what socio-economic security should entail in a Good Society of the 21^st century. Of the seven forms of labour security, with more open economies and flexible labour markets (like them or not), labour market security is unlikely to return. In any case, one can argue that this particular form of security and employment security should be regarded as primarily instrumental, and by the nature of desired economic policy and labour market structure cannot be offered to all. While work, job and skill reproduction security are derived needs, for a society in which freedom to choose and to act and economic security are equally respected, the crucial forms are income security and representation security. Without basic income security, freedom to make choices and develop skills and a niche in society will be impossible. Without effective Voice, those on the margins of society, those vulnerable to impoverishing detachment, will be unable to overcome their marginalisation, and will be ignored in local and national policymaking, and in the expanding sphere of supra-national policymaking. Fragmentation and social dumping will become pervasive.

The challenge before us is to find ways of giving Voice to all interest groups and to find ways of giving all such groups that minimal income security. The good news here is that there is growing intellectual and political unease at the morality, legitimacy and sustainability of the individualistic, disembedded winner-takes-all economy. Perhaps the ILO can reposition itself to turn that good news into a strategy for promoting new forms of security and distribution appropriate for the emerging global economic system. It might be worth the effort to try to do so. In the new century, Richard Tawney�s aphorism cited at the outset of this paper seems peculiarly apt.

I would like to conclude with a statement of sentiment that seems even more apt for a conference with a strong gender orientation. It is taken from a dull paper that I have recently written about modes of control and the ILO�s "decent work" agenda. The theme is developed at length in a recently completed book on modern state paternalism.

"Paternalistic control must be overcome as part of an agenda for occupational security. In doing so, how can a balance be achieved between freedom as autonomy and freedom as responsibility? The anti-paternalist believes in freedom from controls and freedom to be able to make rational choices � "the wish to be self-directed and not to be directed by others". But freedom is not unbridled individualism. Any individual needs some constraints � boundaries or pressures to direct him or her away from pure egotism. The desirable constraint is some form of collective that limits opportunism while facilitating the freedom to develop. This is the Voice security that we should be seeking.

The trouble is that any collective by itself will become oppressive unless checked by some other form of collective. Thus, the family will be oppressive unless its members can draw strength from belonging to a wider community; the union will be oppressive unless civic associations can give strength to individuals; the civil society organisation will become oppressive and opportunistic unless its members can identify with a balancing group, and so on. We need a set of collectives. As noted earlier, G.D.H.Cole put it well in 1920, at a time of ferment as trade unions and cooperatives struggled for identity, "A person requires as many forms of representation as he has distinct organisable interests or points of view." In short, freedom requires a system of cooperative individualism in order to restrain moral and immoral hazards.

This should lead us to consider current populist imagery. The notion of "empowerment" should be disquieting. We should feel uneasy about the language of battle. Of course, social relations are about adversarial bargaining and "struggle". It is intellectually reprehensible to talk or write as if there were no conflicts of interest; this leads to flabby thinking by bureaucratic minds unwilling to take intellectual risks. However, a danger of the current discourse around "development as freedom" is that it depicts freedom as competitive individualism, consumerism, possession, aggrandisement, maximisation of short-term profits and individual advantage. It fosters a Hobbesian mentality, which turns all social relations into "winners" and "losers". This freedom to be endlessly at war with our fellow beings, with nature and ourselves, is driving us into a frenzy of "competitiveness", egotism, stress, "labouraholism", "presenteeism", karoshi and other social sicknesses.

Of course we need production, which requires incentives. However, we must reflect more. We need a softer tone, a less abrasive way of living, in which self-control does not mean merely freedom to compete opportunistically and frenetically with others more "equally". The stress that is the modern illness of the labour ethic will not be addressed by this route. We must reject the language of empowerment. It is "dis-empowerment" that is required; it is the negation of those controls, in order to liberate our enthusiasms, our creativity, and most of all, our capacity for contemplation and reflection. That is real security. The greatest freedom of all is to be still. Dignified work can only evolve if ordinary people have the capacity to say "No". This is a disturbing message for those who want to see the extension of markets to every crevice of human existence and who see the multiplication of jobs as the answer to the human condition. Dignified work needs basic security, or real freedom is denied. The ultimate paradox is that it requires the freedom to do no work at all. Dignified work can only exist when it is done for intrinsic reasons, not because a landlord, a boss or the State says it shall be so."