Encouraging the unemployed into sustained work: experimental evidence from the UK and the US

In seeking to increase the employment rate in the UK, labour market policy has traditionally focused on encouraging employment entry. However, many people who leave benefit enter poorly-paid, insecure employment and before long have once again returned to welfare. For individuals stuck in such a 'low pay, no pay' cycle, labour market prospects are poor, the opportunity for progression is constrained and the risk of poverty is high. For policymakers, there are additional concerns to do with efficient functioning of the labour market and social mobility.

In view of this, there is a clear interest in encouraging those who enter work to remain in work. In both the UK and Texas, the possibility of using time-limited earnings supplements to achieve this has been trialled. The intuition behind offering support for a limited period of time rather than on an ongoing basis (as with tax credits) is that people face particular challenges as they move from benefit to employment but, if they can complete the transition successfully, their chances of remaining in work are greatly increased. Consequently, it is during the transition period that such support is most needed and this argues in favour of time-limited intervention.

While there is strong evidence that such supplements increase employment, the crucial question of how these increases arise is not well-understood. If they are due solely to more people entering work, this would indicate a failure of supplements to achieve their aim of improving retention and so the benefits that might flow from remaining longer in work - increased employment stability, skill acquisition, earnings growth, career advancement etc - are unlikely to be realised. If, on the other hand, the increases are largely attributable to more people remaining in work, this would point to the effectiveness of the supplements in supporting individuals in the early months of new employment (when the risk of job loss is highest) and suggest that such policies have the potential to address the 'low-pay, no-pay' cycle, thereby improving upward mobility in the labour market.

The first aim of the proposed study is to identify whether the positive employment effects found when trialling time-limited earnings supplements in the UK are due to increased employment retention or to increased employment entry. A similar investigation for the case of Texas will also be carried out. The results for the UK and Texas will be of interest in their own right since very little related evidence exists, but a comparison of the results may be particularly informative about whether the findings can generalise to other settings.

The second aim of the study is to understand the reasons for the differences between the UK and Texas in the effectiveness of time-limited earnings supplements for lone parents (most welfare recipients in Texas were lone parents). We will use an innovative technique to identify similar subgroups in the UK and Texas according to their estimated probability of employment in the absence of the earnings supplement. We will use impact estimates for these subgroups to explore the question of what impact the kind of supplements trialled in Texas would have had in the UK and, conversely, what impact the kind of supplements trialled in the UK would have had in Texas. This approach offers the potential to reduce the influence of observed and unobserved variation when making cross-country comparisons and so may be of use to other researchers.

The proposed study builds on large-scale evaluations that have been funded by the Department for Work and Pensions (DWP) in the UK and the Department of Health and Human Services (HHS) and the Department of Labor (DOL) in the US. Consequently, these departments have an immediate interest in the project. The results of the analysis will provide insights into the effect of time-limited earnings supplements on employment retention. Very little evidence exists on this internationally so policy makers in other countries may also find the results of the analysis informative, particularly if there is some evidence of common effects across the UK and Texas.

Equipping policy-makers with high-quality information increases the likelihood that labour market policy can be developed in such a way that the 'low pay, no pay' cycle is broken. If it can assist the formulation of appropriate policy in this way, the research may prove beneficial ultimately to those individuals and families with only a weak attachment to the labour market. If policy can be effectively developed in order to increase their employment retention and consequently their resilience in work, significant benefits will result, for example: improved employment prospects, greater earnings, more rewarding work, higher standard of living and improved social mobility. Benefits felt at the broader level would include a more efficient labour market and a more skilled workforce, as individuals increase their experience and knowledge.

The work has the potential to have an immediate and long-lasting impact on policy. The applicant and his US partners have a long history of working with the DWP, the HHS and the DOL. In order to raise the profile of the research with UK policy-makers, a seminar at DWP will be arranged, with H.M. Treasury and other interested departments invited. The research will also be presented at the Work and Pensions Labour Economics conference (which is sponsored by the DWP) in order to disseminate more broadly across interested government officials.