The Disability Discrimination Act in the UK: Helping or hindering employment among the disabled?

Abstract

The enactment of the Americans with Disabilities Act (ADA) in 1990 triggered a substantial academic debate about its consequences on employment rates of disabled people. In contrast, the employment provision of the 1996 Disability Discrimination Act (DDA) in Britain has received little attention. Exploiting both pooled and longitudinal data, this paper provides robust evidence that, similar to the ADA in the USA, the DDA has had no impact on the employment rate of disabled people or possibly worsened it. Possible reasons for this are higher uncertainty around litigation costs, low levels of general awareness about the Act among disabled people and employers, and a lack of financial support.

Introduction

In 1995 the UK Government passed the Disability Discrimination Act (DDA). Its intention is to end discrimination against disabled people. It aims to protect people in the areas of employment, access to goods, facilities and services, the management, buying or renting of land or property and education. Under Part II of the Act, which came into force in December 1996, it is unlawful for employers covered by the Act to discriminate against disabled employees. Similar to the Americans with Disabilities Act (ADA) which was introduced in 1990, the employment provision of the DDA aims to overcome barriers to employment for disabled people. It was hoped that the extended rights under the DDA would significantly increase the chances of disabled people to obtain and remain in employment. Critics pointed out that additional costs imposed by the legislation (hiring and firing, possible litigation and adjustments costs, and higher discount rates through more uncertainty) might lower employment rather than raise it. Thus the UK has experienced a similar debate about the merits of legislation intended to improve disabled peoples’ labour market outcomes. However, its effects and particularly its impact on employment have not been subject to the same rigorous analysis as has occurred in the USA. This paper seeks to provide such evidence.

The theoretical impact of anti-discrimination legislation is ambiguous. By giving them additional rights, disabled people are more likely to supply labour and existing employees may become more productive. But higher expected hiring and firing costs may dissuade employers from hiring disabled employees in the first place.¹ In the absence of efficient enforcement mechanisms, employers will seek to avoid such extra costs. Such enforcement can either be formal (through tribunals and courts) or informal (name and shame).

The ADA has sparked a large body of economic studies on its effects. DeLeire (2000) claims that it decreased employment rates for disabled men on average by 7.2% compared to the pre-ADA period. Yet, there has been no change to male wages post-legislation. Acemoglu and Angrist (2001) find a strong decline in hours worked shortly after the introduction of the ADA for men of all working ages and women under 40. They argue that only part of this can be explained by an increase in Social Security Disability Insurance (SSDI). Houtenville and Burkhauser (2004) replicate Acemoglu and Angrist (2001), but they use a more long-term definition of disability that utilises the matched CPS two-period work limitation definition of disability rather than the one-period used by Acemoglu and Angrist. With this longer-term definition of disability, they do not find evidence of negative employment effects from the ADA. Estimates of policy effects may be sensitive to choice of data. Houtenville and Burkhauser (2004) argue that the decline in employment among the disabled began at the time of the expansion of SSDI eligibility. Autor and Duggan (2006) also highlight the adverse effects on employment of increased SSDI take-up.

Measured effects on employment are larger in medium-sized firms possibly because small firms were exempt. Jolls and Prescott (2004) exploit inter-state variation in pre-ADA disability legislation to show that accommodation costs for disabled people account for much of the decline in employment rates observed in the CPS. However, they also argue that factors other than the ADA itself contributed to the observed decline. Finally, Hotchkiss (2004) finds little effect on employment rates of disabled people post-ADA modelling participation and employment effects separately and even higher employment rates for certain disability types.

Results in these studies are not without controversy. Disability measures based on self-reported work limiting health problems may suffer from several shortcomings (Kruse and Schur, 2003). If for example people with disabilities get access to better jobs they might no longer declare themselves as having a work limiting disability. Furthermore, it has been argued that a substantial proportion of those reporting a work disability may not be covered by the ADA disability definition. Finally, empowering disabled people with more rights may remove the stigma attached to disability and increase the likelihood that it is reported. Each of these issues will impact on the measure of the employment rate of disabled people. Nevertheless, Burkhauser et al. (2002) argue that differences in trends in self-reported work limitation and in health impairments data from US employment surveys are insignificant. Hence they argue that data on self-reported health contains useful information which can be used to estimate the population of individuals with disabilities.

As mentioned above, other economic and policy developments affect the employment prospects of disabled people and may have biased estimates of the effects of the ADA. For example, the ADA was introduced while the US economy was slowing down. Employment growth averaged 0.3% between 1989 and 1992, compared with 1.7% over the entire period between 1980 and 2000. Economic slowdown may have differentially affected disabled workers and consequently led to misinterpretation of the effects of the ADA.

Aware of these difficulties, we use both pooled and longitudinal data to study the effect of the DDA on employment rates of disabled people in England. The paper contributes to the literature in at least two ways. First, to our knowledge this is the first formal economic pre- and post-evaluation of the DDA though Jones et al. (2006) study the impact of the legislation since its introduction. Second, the paper also aims to overcome some of the conceptual problems around the disability definition (mainly work limiting) and transition in and out of disability inherent in the US literature by using a range of disability definitions and by using longitudinal data.