The stubborn persistence of unemployment in industrialized countries, both in the aggregate and – for many individuals – as a long-term phenomenon, has generated a range of new – and increasingly radical – policy proposals. The role of schemes directed to reduce long-term unemployment, for example, has received renewed attention in the light of the new Labour Government’s proposals, announced by Chancellor of the Exchequer, Gordon Brown, in his first budget in July 1997. Similarly, there is growing recognition that economic growth alone will not be sufficient to resolve the overall scale of unemployment and that other supplementary measures, or ‘active labour market policies’ (ALMPs) are needed.

The merits and limitations of a number of these new ideas were discussed at a seminar jointly organized by CEPR and NERA on 10 July 1997. First, Jon Stern, a NERA Senior Consultant, discussed some of the ALMP proposals directed at the UK’s long-term unemployment problem. Second, Gerry Holtham, Director of the Institute for Public Policy Research, explored a number of the more general labour-market policy options for dealing with unemployment. And, finally, Dennis Snower (Birkbeck College, London, and CEPR) explained the radical notion of ‘unemployment and training accounts’ (UTAs).

Stern noted that the unemployment rate had been falling rapidly in the United Kingdom in recent years, from 10.1% of the labour force in 1992 to 7% in January 1997. The rate was now substantially lower than in France (12.5%), Germany (9%) and the EU average (10.8%). Long-term unemployment remained high, however, with 36% of benefit claimants having been unemployed for spells of 12 months or more. Therefore, the priority – as recognised in the budget – was to increase the

employability of the long-term unemployed and others on the margins of the labour force. This would help maintain their attachment to the labour force, and allow the economy to operate at a higher level of output and employment without jeopardising inflation targets.

There are various types of ALMPs, falling broadly within two categories: wage-subsidy schemes to encourage employers to hire the long-term unemployed; and public-sector employment schemes. In discussing these, Stern drew on recent reports prepared by NERA, which included a survey of the impact of wage-subsidy schemes in OECD economies, and a NERA-designed proposal called ‘Opportunity to Work’. Wage subsidies for hiring the long-term unemployed have the potential to reduce the sustainable level of unemployment, as well as to raise long-run employment and output levels. It is also possible that they will be self-financing, although this outcome is less certain and might take some years to be realized. Wage-subsidy schemes, however, tend to have low levels of take-up. Moreover, they have the disadvantage that employers can receive a subsidy for recruits whom they would have been willing to hire in any event. These ‘dead-weight’ levels can be as high as 50% or more, so that the number of people for whom subsidies are paid can be twice as many (or more) as the reduction in long-term unemployment. Thus the effectiveness of wage-subsidy schemes may be more limited in practice than their advocates might claim.

NERA’s ‘Opportunity to Work’ proposal was intended to operate as a community programme for the 1990s with the emphasis on maximizing the employability benefits for the long-term unemployed. The scheme would use the ‘challenge principle’ – which had also been employed in the Urban Programme – to invite bids from work-scheme providers for the organization of specified and relevant training, including day-release and job-search assistance. The schemes would offer supported work programmes, typically of up to 12 months’ duration, for the long-term unemployed, but the payments received by sponsors would be higher, the greater the specific employability benefits offered by their schemes.

Stern argued that, since the scheme was designed to use economic incentive effects to ‘leverage in’ employability benefits, this should help to maximize the subsequent employment prospects of scheme leavers, who could also be expected to secure better-quality jobs. It was acknowledged, however, that experience in practice would be needed to establish whether the number of scheme sponsors applying for funding would be sufficient for the economic incentives to work as intended. The take-up rate would therefore be a key issue, and it would be crucial for the success of the scheme that it be run by a small and effective agency. Stern nonetheless acknowledged that both UK and international experience suggested that the impact of ALMPs of all kinds on long-term unemployment had been limited and that too much should not be expected from such schemes.

Turning to the more general policy options for dealing with unemployment, Gerry Holtham noted that, while a period of sustained growth at 2.5% per annum would do a great deal to alleviate the problems of unemployment, given the scale of inactivity in the United Kingdom, specific measures would still be needed to supplement the effects of economic growth. Where pockets of inactivity are close to centres of economic activity, the government’s present proposals might offer a good start: Although wage subsidies would result in substitution, the resultant ‘churning’ of the pool of unemployed may be all that is needed to remove such pockets of unemployment in a context of general growth.

There was plenty of evidence, however, showing that people on ‘sink estates’ could be left untouched by economic growth in ‘their’ town or locality. In these and other black spots, therefore, it was unlikely that currently announced measures would prove adequate. There were several reasons for this. Take-up rates of labour subsidies may be too low; all past evidence indicated that training was a remarkably inefficient antidote to unemployment; schemes that did not result in worthwhile work would become discredited; and there are many inactive people outside the scope of the measures. Moreover, for people with dependants, it was essential – as the government was aware – to tackle the severe unemployment traps created by the benefit system.

Present proposals should be supplemented, therefore, by other measures. First, there was a need for job-creation schemes in low-activity areas, managed with the collaboration of local government. These would cost perhaps a billion pounds a year. Second, although it would also be expensive, the different benefit schemes – notably income support and family credit – should be integrated, and ‘forgiveness’ and ‘tapering’ should be modified to reduce work disincentives. But Holtham also advocated a number of pilot projects for more radical measures. In particular, poor areas could be singled out as ‘social enterprise zones’ for experiments in integrating the taxation and benefit systems – e.g. offering tax credits (or negative income tax) for the low paid – and abolishing all benefit rules for a five-
year period, thus allowing people to retain benefits to
which they acquire an initial entitlement, irrespective of the work they find during this period.

A radical proposal of a different kind was outlined by Dennis Snower. This was the notion of
‘unemployment and training accounts’ (UTAs).

Snower pointed out that, since unemployment falls much more heavily on the unskilled than the skilled, the government’s employment policies cannot be seen in isolation from its policies to promote skills. The UK government currently spends a very large sum of money on unemployment support, further education and training. The question that Snower asked was whether these funds could be redirected to create more incentives for people to become employed and acquire skills.

Referring to research that he had recently undertaken with J Michael Orszag, Snower argued that replacing the current unemployment benefit and public-sector training systems by UTAs could fulfil
this prerequisite. In essence, the proposal was that every employable person should be assigned two


accounts: an unemployment account, to provide support against job loss; and a training account, to provide funding to acquire new skills. Instead of paying taxes to finance general government spending on unemployment support, further education and training, all employed people would be required to make regular contributions to their UTAs. The size of these mandatory contributions would rise with their incomes. To maintain the living standards of the poor, the government would pay the contributions of the lowest income groups, and would tax the contributions of the higher income groups. People could also make voluntary contributions in excess of the mandatory amounts.

When people became unemployed, therefore, they
could make limited withdrawals from their unemployment accounts instead of receiving unemployment benefits. If they also wished to acquire skills, they could draw on their training accounts, instead of receiving government grants, subsidies and loans. If their UTA balances fell below a specified limit, they would receive public assistance on the same basis as under the current system. If their UTA balances became sufficiently high, they could use the surplus funds for other purposes. At the end of their working lives, their remaining UTA balances could be used to top up their pensions.

Withdrawals could be made from training accounts at any point in people’s working lifetimes. Those who identified their preferred careers early in their working lives could draw substantially on their accounts soon after leaving secondary school. Those who took longer to find their niche in the labour market, or who required retraining upon changing occupations, would make significant withdrawals much later in their careers. In this way, training accounts would enable people to remain employable and adaptable throughout their working lives.

Further ideas included the possibility of people being able to borrow money on favourable terms for their training accounts, thus enabling them to finance their acquisition of skills through their future incomes. Unemployed people who developed promising job market strategies at their ‘Restart’ interviews could receive government loan guarantees when borrowing their training account money. And employers’ contributions to employees’ training accounts should receive favourable tax treatment.

Initially, the UTAs would be managed largely on a pay-as-you-go basis (similar to saving accounts, from which people can make withdrawals even though the banks use most of the money for other purposes). With the passage of time, however, they could eventually be turned into a fully-funded system, in which individuals would have discretion over who could manage their UTAs. To guard against bankruptcy, the financial activities of private-sector UTA fund managers would be regulated, along lines similar to the regulation of commercial banks.

Adoption of the UTA system, so Snower maintained, could substantially reduce the level of long-term unemployment and promote the acquisition of skills. In particular, moving from unemployment benefits to unemployment accounts would give individuals greater incentives to avoid long periods of unemployment. For the longer people remain unemployed, the lower their unemployment account balances would become and, consequently, the smaller would be the funds available to them later on. Moreover, since the unemployment accounts would generate more employment than unemployment benefits, the unemployment account contributions necessary to finance a given level of employment support would be lower than the taxes necessary to finance the same level of unemployment benefits.

Similarly, the training accounts would be better suited than current education and training programmes to ensuring people’s lifetime employability, since the funds could be accessed whenever employees (and their employers) found it worthwhile. In this way, both employers and employees would stand to gain from a switch to UTAs. Retired people would also gain, through their ability to use their UTA balances to augment their pensions. And the government would gain, since the removal of the distortions from the unemployment benefit system would promote new economic activity and thereby generate increased tax revenue. Beyond that, the UTAs would be more efficient than the current system at redistributing income from rich to poor, since unemployment benefits and training schemes are not targeted exclusively at the poor, whereas government contributions to UTAs would be.

To provide additional incentives to find work and acquire relevant skills, the government would subsidize long-term unemployed people who draw on their UTAs to provide recruitment or training vouchers for firms that hire them. The size of each individual’s voucher would depend on his or her wages earned over the subsequent two years of employment. The recruitment vouchers would reduce the cost to firms of training the recruits; and the subsidies would be set so that they could be financed through the tax revenues from the recruits’ first two years of subsequent employment and through the abolition of in-work benefits.

Finally, Snower argued, replacing the current system by UTAs would not only reduce unemployment but would simultaneously promote equality. While people are generally resentful of their tax burden and often feel demeaned by the existing unemployment benefits and training programmes, they would be more willing to contribute to personalized accounts for their own purposes. UTAs would afford people more freedom to use employment support and training funds to meet their diverse individual needs, and greater latitude to respond to changing job opportunities, finance periods of job search, acquire skills and provide for retirement. Furthermore, all this could be done without creating greater inequality or increasing government expenditure.

G de la Dehesa and D J Snower (eds.), Unemployment Policy: Government Options for the Labour Market, Cambridge University Press for CEPR, 1997.