According to the Labour Force Survey, this was 3.3 per 100 employees in 2004.²⁷ Applied to the WERS 2004 population, this equates to around 740,000 voluntary exits per annum, or just over one voluntary exit per workplace (1.06). WERS estimates that around 322,000 workplaces are covered by some form of employee representation. To illustrate costs, if representatives in these workplaces
contributed to reducing the number of voluntary exits by 5-10 per cent, then the benefits translate to around 17,000-34,000 fewer exits.²⁸

Avoiding such exits will carry significant financial benefits. As above, recruitment costs for the employer could amount to around £72 - £143 million. Furthermore, there are substantial psychic costs for employees, including anxiety, stress and inactivity as workers become discouraged from participating in the labour force.

Since 2000, workers have had the right to be accompanied to disciplinary and grievance hearings. Employee representatives, who bring support and valuable experience to the process, can contribute to the equitable resolution of disputes. Antcliffe and Saundry (2006) using data from WERS 2004 found that establishments that fulfilled the requirement of accompaniment at grievance meetings reduced the Employment Tribunal (ET) application rate by 19 per cent.²⁹

The argument is that employee representation contributes to reducing the tribunal rate (number of tribunal applications per 100 employees) for covered employees. According to WERS 2004, the tribunal rate per 100 employees was 0.24 for employees in workplaces with 5+ employees.³⁰  If we assume that the representatives in covered workplaces (around 321,000 covering around 15.3 million employees) contributed to reducing the mean dismissal rate by 10-20 per cent (e.g. in their absence the rate would have been 0.26-0.29 per 100 employees), this translates as 3,600-7,300 fewer ET applications. If we assume that the average cost of an ET case to business is £4,900 and £990 to the Exchequer,³¹ this equates to annual savings in the region of £22m-£43m as a result of fewer ET applications.

Worker safety
Kaufman and Levine (2000) discuss how the public good nature of improvement in working conditions often discourages individual employees from contributing to a safer, efficient workplace. No single employee has the incentive to investigate and tackle hazardous or impeding conditions. Representation solves this problem by 'pooling' worker interests and encouraging workers to participate. This leads to more creativity and efficiency in devising and implementing health and safety measures in the workplace. The reasons worker involvement through representatives tends to be effective could include better quality of information available for decision-making, and where representatives have been involved in drawing up procedures, they help to ensure that they are practical and so reduce the number of safety rules broken.

The WERS 2004 survey captures the variety of ways that employers consult with their workforce on health and safety issues. Managers were asked if there was a committee present that dealt with issues of health and safety, either multi-issue or dedicated to H&S issues only. The findings are described in Tables 5A and 5B above, and show that around 27 per cent of workplaces with 5+ employees had some form of dedicated health and safety representation, covering 65 per cent of employees.

There is good evidence to suggest that H&S representatives make a valuable contribution to the workplace. One research study, which included a review of the empirical literature as well as original analysis of WIRS1990 and WERS 1998, concluded that their representation and consultation has a significant role to play in improving health and safety at work.³²  A number of papers have used previous WERS surveys to find the impact of representation, one using WERS 98 data found that when there was a union presence in the workplace, the injury rate is 24 per cent lower than in workplaces where there is no union presence.³³  Another study by Paci, Reilly and Holl from 1994 based on WIRS1990³⁴, is also commonly quoted as saying that safety reps roughly halve injury rates. Walters et al tried to reproduce their analysis and could not establish evidence for the same scale of effects. We therefore need to be cautious about ascribing a particular numerical value to the effectiveness of safety reps.

According to WERS 2004, the average injury incidence rate (number of employees per 100 sustaining listed injuries in previous 12 months) was 0.57 for all workplaces with 5+ employees³⁵ . To illustrate potential benefits, we assume that H&S reps contribute to reducing the injury rate in workplaces by 10-15 per cent. A mean ratio of 0.57 per 100 employees estimates around 128,250 injuries for the WERS population per annum. For workplaces where there is indirect H&S representation this ratio implies around 83,400 injuries in aggregate. If in the absence of representatives, the injury rate would have been 10-15 per cent higher (e.g. 0.63-0.66), then these benefits translate to around 8,000-13,000 fewer injuries in workplaces where there is indirect H&S representation.

This constitutes significant financial and personal benefits to the employer, employee and society, including reduced lost earnings, avoidance of sick pay and compensation payments, as well as increased output. The HSE estimate that the full costs to society as a result of injury in the workplace to be £5.9-£10.7bn in 2001/02³⁶. There were 7.3 million working days lost in 2001/02, therefore the estimated cost to society per working day lost through injury in 2001/02 is £810-£1,470³⁷.  Further work to refine this calculation would require taking into account that the costs per day could be dependent on the period of absence, and potential benefits would depend on the type of injury prevented.

The mean number of working days lost per reportable injury in 2004 was 19.2³⁸. Using this assumption, we estimate that the avoidance of 8,000-13,000 injuries is equivalent to around 161,000-241,000 fewer working days lost. Uprating the 2001/02 cost estimates to 2004 prices, this brings the potential range of benefits in this instance to £136m-£371m.

In a similar manner, we can calculate the range of benefits associated with reduced illness³⁹ rates due to the presence of H&S reps. According to WERS2004, the average illness incidence rate was 1.76 for all workplaces with 5+ employees. This implies around 257,000 cases in workplaces with indirect H&S representation.

There is very limited empirical evidence on the impact of H&S reps on the incidence of work-related illness. A majority of studies (e.g. Eaton and Nocerino, 2000)⁴⁰ generally support the idea that the presence of some form of H&S representation can reduce the incidence of work-related illness. However, in some cases, such as in analysis by Fenn and Ashby (2001)⁴¹, essentially the opposite effects are found, in essence, workplaces with some form of H&S representation (in this case, trade unions) reported a higher number of cases of work-related illness. One possible explanation for such results was that in workplaces where there was some form of H&S rep, employees were made more aware of their rights and therefore more likely to report work related illnesses. We should therefore make the cautious assumption that H&S reps contribute to reducing the illness rate in workplaces by 1-3 per cent. This translates to around 3,000 to 8,000 fewer cases of illness in these workplaces.

The HSE estimate that the full costs to society as a result of ill health in the workplace to be £11.3-£17.3bn in 2001/02. There were 32.9 million working days lost in 2001/02 through illness caused or made worse by work : therefore the estimated cost to society per working day lost through illness in 2001/02 is £340-£530.

The mean number of working days lost per new case of work-related ill health in 2004 was 48.6⁴². Therefore, the avoidance of 3,000-8,000 cases of work-related illness is equivalent to 125,000-375,000 fewer working days lost. Uprating the 2001/02 cost estimates to 2004 prices brings the potential range of benefits from avoiding work-related illness through the presence of H&S reps to £45m to £207m.

_________________________

²⁷ D. Heap, ONS 'Job Separations in the UK' ONS Labour Market Trends, June 2005.
²⁸ Calculation - Assume 1.06 exits per workplace = 342,000 exits in workplaces with representation.
²⁹ In their analysis, accompaniment could be from either a fellow colleague or formal workplace representative.
³⁰ DTI estimates. 31 DTI estimates based on The Survey of Employment Tribunal Applications, (SETA) 2003. Rounded to nearest £1m.
³¹ DTI estimates based on The Survey of Employment Tribunal Applications, (SETA) 2003. Rounded to nearest £1m.
³² Walters, D. et al., The role and effectiveness of safety representatives in influencing workplace health and safety, HSE Contract Research Report 363, 2005, http://www.hse.gov.uk/research/rrpdf/rr363.pdf. A comprehensive review of the literature is beyond the scope of this paper; see Walters et al for a fuller discussion.
³³ Litwin, A. S., Trade Unions and Industrial Injury in Great Britain, LSE, 2000.
³⁴ City University Dept of Economics Discussion Paper 31.
³⁵ The WERS data on injuries arises from questions IINJURY and INJUMINJ on p.112 of the WERS2004 Management Questionnaire. The HSE estimate an injury incidence rate of 1.33 per 100 employees from LFS data for 2003/04. This figure is based on 'reportable' injuries (i.e. the most serious injuries that have led to, or are likely to lead to, time off work, disablement or death). 
     In trying to explain the difference between the two figures, the two following factors could be influential: Firstly, one might expect some under-reporting in the WERS sample compared to the LFS: an employee is more likely to remember an injury that happened to them, whereas a manager might forget some incidents (and perhaps might also wish to limit the number they report so as not to appear to be a 'dangerous' employer). Secondly, sampling error should be considered. Only 8% of all workplaces reported any of the injuries listed in WERS, therefore the 95% confidence interval will be rather large (the LFS estimate is also likely to be subject to some sampling error, but due to a larger sample size, the confidence interval will be measurably smaller).
³⁶ Interim update of the 'Costs to Britain of Workplace Accidents and Work Related Ill Health', 2004,
http://www.hse.gov.uk/statistics/pdf/costs.pdf.  The full breakdown of costs include costs to employees and dependants in the form of lost earnings, human costs (e.g. psychic costs due to pain and suffering); sick pay paid by employers (in addition to Statutory Sick Pay) and non-injury accident costs; and losses to society in the form of a loss of output, and medical treatment.
³⁷ HSE, Health and Safety Statistics Highlights 2001/02.  http://www.hse.gov.uk/statistics/overall/hssh0102.pdf
³⁸ ONS, Health and Safety Statistics 2004/05. Using data from LFS, there were an estimated 363,000 reportable injuries, and 7 million working days lost through injury. 7m/363,000= average 19.2 days.
³⁹ By 'illness' we refer to the list of illnesses, disabilities and other physical problems under the ILLNESS question on p.112 of the WERS2004 Management Questionnaire. According to the Health and Safety Executive's Statistics Highlights 2001/02 (see footnote 26), the most common types of illness were musculo-skeletal disorders, occupational stress, depression or anxiety, followed by breathing, lung and hearing problems.
⁴⁰ Eaton, A. E., Nocerino, T., (2000), The Effectiveness of Health and Safety Committees: Results of a Survey of Public-Sector Workplaces, Industrial Relations, vol. 39, Issue. 2, p. 265-290.
⁴¹ Fenn, P., Ashby, S., (2001), Workplace Risk, Establishment Size, and Union Density: New Evidence, Centre for Risk & Insurance Studies, University of Nottingham.
⁴² There were and estimated 576,000 new cases of work related illness and 28 million working days
lost through such illness in 2004 (28m/576,000=48.6).

_________________________

A copy of this HandS article, as a PDF zipped file, is available here

The whole of the Consultation paper, in Adobe PDF format,  is available here:
Workplace representatives: a review of their facilities and facility time - January 2007 (405kb) PDF