A Guide to using Statistics in Employment and Equality Litigation
Numbers can be anathema to many lawyers. Yet statistics are a useful weapon in the litigation armoury. This week the Government released its Race Disparity Audit which provides a wealth of such statistics and is a timely reminder of the role that they can play in litigation. Tom Gillie discusses three recent examples of how statistics can be used to advance successful arguments in employment litigation and broader equality context, for example, in relation to the provision of goods, facilities and services.
Government Race Disparity Audit
All of the data collated as part of the Government’s Race Disparity Audit is available at www.ethnicity-facts-figures.service.gov.uk, including information related to work pay and benefits and health issues. Notably:
in 2016, every ethnic minority (other than White ethnic minorities) had lower rates of employment than White British people as follows:
in the last 3 months of 2016, the average (mean) hourly pay for all employees was £13.69
in the last 3 months of 2016, the average hourly pay for White employees was £13.75, while the average hourly pay for employees from other ethnic groups was £13.18
Indian employees earned the highest average hourly pay (£15.81), while Pakistani/Bangladeshi employees earned the lowest average hourly pay (£11.42)
on average, employees from Mixed and Indian backgrounds earned more than White employees, whereas Pakistani/Bangladeshi, Black, and Other ethnicity employees earned less
In respect to the civil service workforce:
the percentage of Asian and Black staff at more senior grades was lower than at junior grades - 3.4% and 1.3% respectively at Senior Civil Service grade (the highest grade), compared to 6.5% and 3.3% respectively of administrative officers and assistants (the most junior grade).
the executive officer grade (the second most junior grade) had the highest percentage of both Asian staff (6.8%) and Black staff (3.8%).
‘Common Mental Disorders’ were more prevalent in White British women than in non-British White women, at 21% and 16% respectively.
just under three-quarters of patients (72.7%) had a positive experience when making a GP appointment in 2016/2017.
Irish patients were the most likely to have had a positive experience of making a GP appointment, and Pakistani patients the least likely to.
in each year from 2011/12 to 2016/2017, patients from an Irish, African, White British and any other Black background were more likely than those from other groups to report a positive experience.
Asian patients (Pakistani, Bangladeshi, Indian and Chinese) were the least likely to report a positive experience each year from 2011/12 to 2016/2017.
patients in the Gypsy or Irish Traveller group also had rates of satisfaction below the national average from 2011/12 to 2016/2017, although the findings are less reliable for this group because of the small numbers of respondents.
So how can the data from the Government’s Race Disparity Audit be used in litigation?
Indirect discrimination cases
These national statistics should be born in mind as a useful tool when establishing group disadvantage in indirect discrimination cases, in addition to the Office for National Statistics (ONS) Labour Market Statistics and the ‘Workplace Employment Relations Study’ (2011). Particularly useful figures in the Race Audit’s ‘Ethnicity Facts and Figures’ as they currently stand are those relating to pay and public sector job grades mentioned above.
When deploying these (and similar statistics) in litigation, it should be remembered that there is no one-size-fits-all approach to assessing disparate impact. The guidance set out by the EAT in Harvest Town Circle Ltd v Rutherford remains the starting point when deploying these statistics. In particular:
proportions and numbers, both in disadvantaged and non-disadvantaged groups should be considered, as well as the respective proportion in the disadvantaged groups expressed as a ratio of each other;
it will never be wrong for a tribunal to look at more than one form of comparison. Moreover, if there is any doubt as to the obviousness of the case, the tendency should always be to look at a second or further form of comparison;
No distinction is to be drawn between a considerable and a substantial disparity. That being so, it would be a mistake to conclude that anything that was merely not trivial or de minimis sufficed; and
after looking in detail at such figures, the tribunal must stand back and judge whether the apparently neutral provision criterion or practice in issue has a disparate impact that could fairly be described as considerable or substantial in light of all the figures.
Direct discrimination
The data from the Government Race Audit may also be useful evidence in direct discrimination cases. Specifically, the data may be most helpful for Claimants who have brought cases against the civil service and public bodies from which the government’s statistics have been collected. Where, for example, black men are being paid less on average than other workers in other groups, this may be a fact from which a prima facie case of direct discrimination could be inferred. More generally, the data may be useful for Claimants pursuing employers in other sectors, but care should be taken to make sure the relevance of such data is clearly set out before a tribunal.
Justification of unfavourable treatment
Statistics are also useful in litigation when deciding questions of justification, because of their assistance in making a practical assessment of the impact of unfavourable treatment. In August 2017, the EAT set out in British Airways v Pinaud that statistical evidence produced by the parties was relevant to the question of justification of unfavourable treatment under the Part-time Workers (Prevention of Less Favourable Treatment) Regulations 2000. Readers will remember the Supreme Court’s comments in Ministry of Justice v O’Brien: it will be more difficult for an employer to justify the proportionality of means chosen to carry out its aims if it did not conduct the exercise of examining alternatives, or gathering the necessary evidence to inform the choice of the time.
The EAT decided in Pinaud that, having accepted that the unfavourable treatment was in pursuance of a legitimate aim, the ET was bound to make a practical assessment of the impact of the unfavourable treatment when deciding whether the treatment was appropriate and necessary for achieving the objective pursued. The mere fact that the claimant was subjected to unfavourable treatment (in this case, having to be available to work on proportionately more days than a full time worker) did not mean that statistics were irrelevant. The unfavourable treatment was required to be justified. Its existence did not rule out an enquiry into the extent to which it impacted on the claimant so that a conclusion could be reached about whether the measure was proportionate.
The approach in Pinaud may also provide a useful avenue of response to Claimants’ arguments about justification in Equality Act claims. Claimants often seek to argue, with reference to the Supreme Court decision in Akerman-Livingstone v Aster Communities Limited, that impugned measures were disproportionate because alternative measures could have met the legitimate aim without such a discriminatory effect. It is anticipated that data gathered as part of the Government’s Race Audit could assist here too.
Mitigation
The Race Audit statistics should also be considered when preparing arguments about mitigation in remedy hearings. The data shows that:
amongst unemployed people in 2016, a higher percentage of people from all Other ethnic groups (35%) were unemployed for 3 to 12 months than White people (31%) and for 1 year or more (28% compared to 25% respectively); and
unemployment rates were higher for people from ethnic minorities (other than White ethnic minorities) than for White people across the country; these differences were largest in London (9% for ethnic minorities and 4% for White), the West Midlands (11% for ethnic minorities and 5% for White) and the North West (9% for ethnic minorities and 5% for White).
Claimants from ethnic minority backgrounds may wish to rely on these figures to support arguments that it will be harder for them to obtain new employment and to support arguments of longer- term financial loss flowing from unemployment as a result of dismissal.
Statistics in access to justice and services: Joseph Rowntree Foundation data
There are other sources of data beyond official government statistics.
The successful challenge to Employment Tribunal fees in the Supreme Court in R (on the application of UNISON) (Appellant) v Lord Chancellor (Respondent) demonstrates how data about social trends can be an important tool in discrimination and public law litigation alike. In particular, the Joseph Rowntree Foundation (JRF) has a wealth of data about social issues, including about average household income, poverty and disability status and hours worked by age and gender. That data was used to great effect by the Appellant to tie the deleterious effect of tribunal fees on a number of hypothetical claimants in low and middle income households to likely outcomes in the ‘real’ world. After considering the JRF statistics, the Supreme Court set out:
“Fundamentally, the question arises whether the sacrifice of ordinary and reasonable expenditure can properly be the price of access to one’s rights (paragraph 55)…The question whether fees effectively prevent access to justice must be decided according to the likely impact of the fees on behaviour in the real world (paragraph 93)…It is common ground that payment of the fees would result in the hypothetical households having less income than is estimated by the Joseph Rowntree Foundation as being necessary to meet acceptable living standards. … the fundamental problem is the assumption that the right of access to courts and tribunals can lawfully be made subject to impositions which low to middle income households can only meet by sacrificing ordinary and reasonable expenditure for substantial periods of time. (paragraph 94)”
In the continuing absence of the socio-economic duties envisaged by the Equality Act 2010, evidence of national trends in household income and expenditure may be especially helpful when challenging government decisions relating to access to justice and public services. Such statistics are also useful more widely when persuading courts of the likely hypothetical impact of government, and employers’, decisions.
Conclusion
Employment lawyers should remember the wealth of data from respectable sources about social trends and problems. Statistics are not merely cold, boring numbers to be trawled through in court. They can be the foundation of the most creative and successful lines of argument and can make or break a case. And traditional lawyers among us can relax in the knowledge that, in the majority of cases, no mathematical dexterity is required!
12 October 2017