Recruiting and retaining employees in low-paid labour markets

Donna Brown, Richard Dickens, Paul Gregg, Stephen Machin and Alan Manning

This study explores recruitment and retention in firms offering 'entry-level' jobs (those requiring no specialist education or skill), whose wages are in the bottom fifth of the national earnings distribution.

Based on five large firms operating in up to eight local labour markets, it includes findings and conclusions on:

  • recruitment procedures How firms recruit, who they recruit and how sensitive recruitment is to wages offered and local unemployment levels.
  • turnover rates Who leaves the firm, how quickly, and how this varies by wages offered and local labour market conditions.
  • turnover costs The costs involved in staff turnover, and what proportion of total labour costs for entry-level staff are accounted for by these costs.
  • staff make-up Characteristics of the people employed by the case-study firms.
  • staff satisfaction What levels of staff satisfaction are, and whether dissatisfied staff disproportionately leave the firms.

Part of the Work and Opportunity series, the report not only throws light on how these firms and their workers behave, but also explores how their behaviour relates to economic theory.

Summary

Summary

A recent study has found that companies which are competing in local labour markets for entry-level employees, that is people without specialist education or skills, pursue different approaches to recruitment and wages. The study, by Donna Brown, Richard Dickens, Paul Gregg, Stephen Machin and Alan Manning of the Centre for Economic Performance at the London School of Economics, found:

  • The wages in almost all entry-level jobs in large service sector companies were already higher than the national minimum wage when it was introduced in April 1999. Thus the minimum wage was unlikely to have had much direct impact on these companies.
  • It typically takes a month to fill a vacancy. Existing employees and Jobcentres are the most common methods of attracting job applicants.
  • On average each vacancy attracted just under three applicants. The number of applicants was sensitive to the wage of the job on offer relative to the local area and local unemployment. However, nine out of ten vacancies were filled. Attracting additional applicants offers employers greater choice, but does not generally influence whether or not the job is filled at all. 
  • Firms find it easier to recruit for town-centre sites than for those on the outskirts. The availability of transport is part of the explanation for this. 
  • Firms selected around three-quarters of applicants to attend an interview. Those without relevant experience, especially those not in work, were less likely to be interviewed. Two-thirds of those interviewed were offered jobs. The decision to make an offer was based on impressions made in the interview or references rather than employment histories.
  • Turnover costs account for between 5 and 10 per cent of wage costs.
  • Turnover is sensitive to relative wages. Higher wages reduce the propensity of staff to leave and lead to higher levels of employment on average. However, the reduced turnover costs would not be sufficient to offset completely the higher wage bill.
  • Leavers are more likely to be young, well-qualified and single, but their job satisfaction levels are similar to those of people who stay.

The companies in the study

The study covered five companies - a restaurant chain, a leisure company, a supermarket chain, a hotel group and a food manufacturing company - which were operating in seven different local labour markets. Each company was represented in at least five of the seven areas.

The jobs in the study were of a kind that requires no particular qualifications or previous experience - sales assistants, waiting or kitchen staff, routine production work, bar work, cleaning and preparing hotel rooms. However, they were not unskilled in the sense that they often required the use of initiative, involved interaction with customers, the processing of orders and financial transactions.

Workforce profile

Most of the workers in the study firms were women, the exception being the food manufacturer which employed 80 per cent men. They were mainly young (many were students) and a third had dependent children. In general, each company was operating in a slightly different segment of the labour force. The restaurant employed almost exclusively young people (80 per cent of staff were under 25 and 60 per cent were students). The supermarket employed mainly mothers (80 per cent of staff were female and two-thirds had dependent children) as did the leisure company. The hotel had the most mixed workforce - a third students, half mothers, half under 25.

Pay

In all cases the wages for entry-level jobs were quite modest - they were typically in the bottom fifth of the overall distribution of earnings. Nevertheless, they were above the level at which the national minimum wage was introduced in April 1999. In some of the companies, wages were centrally determined, but in others site managers had a good deal of discretion. Thus, the relative position of each company in the local earnings league varied in the different local labour markets. Only one of the five companies - the restaurant chain - had more than a handful of employees who were paid below the minimum wage level, and even there, almost nine out of ten were paid at higher rates.

Recruitment

Each job vacancy had an average of around three applicants. Those in areas of higher unemployment had more applicants than those situated in tighter labour markets. The number of applicants was higher where the wage offer was higher relative to the local labour market. Jobs located in the centre of towns or other areas with good public transport tended to have more applicants than those in suburban or out-of-town areas. Advertising in newspapers also increased the number of applicants, but the costs are considerably higher than those for other recruitment methods.

Employers were generally able to fill all their vacancies from the pool of applicants available to them. Measures which increased the number of applicants widened the choice available to them, but increased costs as well. In general, the employers in the study used the lowest cost methods of recruitment - recommendations from existing employees (60 per cent of vacancies), notification to Jobcentres (60 per cent of vacancies), notices in the window (40 per cent of vacancies). In a third of cases employers advertised in newspapers (typical cost around £200) or made approaches to people known to the company.

Even in those companies with relatively straightforward recruitment processes, each vacancy took around a month to fill. Typically around a fortnight was allowed between the vacancy being opened and the closing date for applications. The process of arranging and holding interviews took another week, and there was a final week between the job offer being made and the new recruit starting work. This period inevitably means the company either is unable to operate at full capacity, or incurs additional costs by having to pay other staff to cover the additional work.

Three-quarters of applicants were selected for interview. The evidence on the recruitment process suggests that neither age nor gender were important in explaining which applicants were offered an interview. However, previous and current work experience did increase a person's chances. The study found some evidence that job seekers who were not in employment at the time of their application were less likely than employed applicants to be invited for an interview. However, those who got through to the interview stage were as likely to be offered a job as employed applicants. Overall around a quarter of applicants were not selected for interview, and around a tenth of those invited did not attend. Two-thirds of those interviewed were offered jobs, and nine out of ten accepted the offer. The most common reasons for not offering jobs were that the applicant had poor references or made a poor impression at interview, or that it had not proved possible to agree on suitable hours.

Costs of turnover

Labour turnover adds to employers' costs in four ways:

  • Direct costs, for example of newspaper advertisements;
  • Management time;
  • Lower productivity both by a new employee while he or she learns how to do the job and by supervisors and experienced workers who are involved in their training;
  • Loss of capacity while a vacancy is unfilled.

In the firms in this study direct costs were low because most companies used only 'free' recruitment methods. Lower productivity by the new employee meant that on average two weeks' output per recruit was lost. It took between six and ten weeks for a new recruit to be able to handle the same workload as an experienced worker. The cost of managers', supervisors' and colleagues' time in shortlisting, interviewing and training a new recruit was equivalent to around a week of the recruit's wages. On average it costs a total of around four weeks' wages to recruit a new employee. 

In all the firms in the study the level of turnover was high, ranging from 40 per cent per year in the food manufacturer to 100 per cent in the restaurant. This compares with around 20 per cent in the economy as a whole. Existing staff had been with the firm on average for between just under two years in the restaurant to over six years in food manufacturing. However, this average concealed the fact that all of the companies had a small number of long-serving staff and a large number who stayed for only very short periods. 

In the supermarket four out of five recruits stayed for more than thirteen weeks; in food manufacturing and leisure three out of four did. However, in the restaurant a third of recruits had left within the first three months, while in the hotel 40 per cent had. In three of the companies total turnover costs were around 5 per cent of the wage bill, while in the other two it was around 10 per cent. However, to some extent this figure masks the true costs, because it is kept down by including in the denominator long-standing employees who have higher wages and zero turnover costs. The turnover costs of each new recruit in the two companies with the highest turnover add around a third to their wage costs over the time that they spend with the firm. In the other three companies this figure is much lower - around 10 per cent. This is mainly because the new recruits stay for longer periods so the time over which their recruitment costs have to be spread is longer.

Would higher wages reduce turnover?

Given the large costs of turnover in these companies, and the sensitivity of recruitment to wage levels, the study examined whether higher wage rates could be expected to reduce turnover. It then considered whether the higher costs incurred by paying higher wages could be offset by lower turnover costs. In other words could the companies reduce their total labour costs by paying higher wages. 

The workplaces in the study that had the highest earnings relative to their local labour market had the lowest turnover rates (around 43 per cent) while those which had the lowest pay levels relative to the area had the highest turnover rates (around 95 per cent). Turnover rates are also influenced by local unemployment rates (workers are less likely to leave when there is more competition for alternative job opportunities).

The study suggests that on average increasing wages by 10 per cent reduces leaving rates by between 15 and 23 per cent. As each vacancy is typically unfilled for a month, this illustrates how higher wages (for example, following the introduction of the national minimum wage) can be associated with higher levels of employment. The lower staff leaving rate in turn reduces turnover costs by 11/2 to 21/2 per cent of the wage bill. Thus in most companies lower turnover would not be sufficient to offset the effect on the wage bill of higher earnings. 

Job satisfaction

In all five companies workers liked their jobs on balance. Relationships with colleagues and their supervisor and ease of travel to work were the most highly valued in all cases. However, pay was the least satisfactory aspect of their jobs in all cases other than retailing, where the pay was higher relative to the local labour market than in the other four companies. Women employees were more satisfied than men about every aspect of their jobs except promotion prospects. Women were particularly satisfied with their hours of work. Young people and those with higher levels of education were less satisfied in these low wage and relatively low skilled jobs than older workers and those with low or no qualifications.

However, there was no relationship between job satisfaction and turnover. Leavers were no more likely than those who stayed to be dissatisfied with their jobs. Rather, other features associated with alternative job prospects such as qualifications, absence of past experience of unemployment and being single without dependent children were associated with higher leaving rates. 

Labour supply and demand

The researchers conclude that the firms in the study are maintaining a balance between their wage bills and their turnover rates. They are living with the inconvenience of higher rates of turnover than they would like in order to keep their overall labour costs down. Their wage rates are such that within their local areas they are able to attract sufficient recruits to continue to operate the business. Higher wages reduce turnover costs mainly by reducing the rate at which workers leave the firm. These lower turnover costs don't fully offset the increased wage costs. However, as the labour market tightens the problems of holding staff and replacing leavers gets harder and firms are likely to face pressure to raise wages.

About the study

The research is based on data collected in 1996 and 1997 within 5 companies operating across 7 regions: North West London, Metropolitan West Midlands, West of England (Cheltenham and Gloucester), West Yorkshire (Leeds and Bradford), Welsh Borders (Shropshire and Cheshire), Merseyside and the North West (Blackpool and Preston). The companies in the study were a restaurant chain, a hotel chain, leisure company, a supermarket chain and a food manufacturing firm. The study analysed administrative records including payroll data and financial performance records. A series of questionnaires was distributed, designed to obtain personal information from appointees, current staff, leavers, site information from managers, and details on the vacancy process from personnel or line managers. In addition to newly collected data comparisons were facilitated by use of the Quarterly Labour Force Survey (QLFS) and the New Earnings Survey (NES).

Downloads

Downloads