Contracts that do not guarantee a minimum number of hours: March 2016

We produce biannual estimates of the number of contracts that do not guarantee a minimum number of hours (NGHC), based on a survey of businesses. These estimates were first produced on 30 April 2014. They complement the figures from the Labour Force Survey (LFS) which show the number of people who report that they are on a “zero-hours contract” in their main employment. This report includes the latest figures from the LFS for October to December 2015 as well as new estimates from the fourth and fifth survey of businesses for May and November 2015, respectively.

The latest estimate of the number of people who are employed on “zero-hours contracts” in their main employment, from the LFS, a survey of individuals in households, is 801,000 for October to December 2015, representing 2.5% of people in employment. It should be noted that responses to the LFS can be affected by respondents recognising the term “zero-hours contract”. This latest figure is higher than that for October to December 2014 (697,000 or 2.3 per cent of people in employment), but it is not possible to say how much of this increase is due to greater recognition of the term “zero-hours contracts” rather than additional contracts.

People on “zero-hours contracts” are more likely to be young, part time, women, or in full-time education when compared with other people in employment. On average, someone on a “zero-hours contract” usually works 26 hours a week. Around 1 in 3 people (37%) on a “zero-hours contract” want more hours, with most wanting them in their current job, as opposed to a different job which offers more hours. In comparison 10% of other people in employment wanted more hours.

The results from our latest survey of businesses indicates that there were around 1.7 million contracts that did not guarantee a minimum number of hours, measured with regard to work carried out during the fortnight beginning 9 November 2015. For the May 2015 reference period (the fortnight beginning 11 May), the equivalent estimate was 2.1 million. The estimates may be affected by seasonal factors, relating to when the data were collected. Therefore estimates presented for different time periods should not be compared directly. Also, estimates may fluctuate over time due to sampling variation (see Annex 1). The difference between estimates from the business survey and LFS are partly accounted for by people who have more than one “zero-hours contract” with different employers or who have a “zero-hours contract” to supplement their main employment.

This paper updates previous analysis on “zero-hours contracts” that we published on 2 September 2015. The report includes the latest figures from the LFS for the period October to December 2015 and the results from 2 surveys of businesses relating to May and November 2015, respectively. The results from the latest survey of businesses from November are not directly comparable with those from May 2015, and there was no equivalent survey in November 2014 to enable appropriate estimates of annual change.

What are “zero-hours contracts”?

There is no single agreed definition of what “zero-hours contracts” are. While some contracts are explicitly called zero-hours contracts, there are other definitions available and used in published statistics. The common element to the definitions is the lack of a guaranteed minimum number of hours.

When developing the survey of businesses, we consulted on the definition to be used and decided on the lack of any guaranteed hours. To provide clarity and prevent confusion with the other estimates of “zero-hours contracts” the rest of this article refers to estimates from the ONS business survey as no guaranteed hours contracts.

When comparing figures from the ONS business survey with the LFS estimates, a number of issues need to be considered:

• the LFS counts people who report that their main employment is a “zero-hours contract”

• the estimate from businesses is counting contracts, which will be greater than the number of people as people can have more than one contract

• estimates from businesses will include contracts that cover a variety of working arrangements, which will include instances where people in their main employment are working a regular number of hours a week (although these hours are not guaranteed by their contract) as well those who work on an irregular basis due to personal choice, availability of work or to fit in around their main employment

This section looks at the latest estimates from the LFS and the ONS survey of businesses. Both measures are estimates from surveys, which due to sampling error are subject to a degree of uncertainty. Where available, an indication of the level of uncertainty is provided in Annex 1.

Labour Force Survey

The LFS samples around 40,000 households a quarter and collects information about people’s employment status. One of the questions on the LFS, asked of people in employment, relates to special working arrangements that vary daily or weekly. Respondents can choose up to 3 different arrangements from a list of 8 options, one of which is “zero hours contracts” defined as “where a person is not contracted to work a set number of hours, and is only paid for the number of hours that they actually work”.

As the LFS is based on respondents’ views about their working arrangements, and counts people rather than contracts, it is likely that any estimate of “zero-hours contracts” from the LFS will be less than an estimate obtained from businesses. The number of people the LFS classes as being on a “zero-hours contract” will be those who:

1. are employed (have done at least 1 hour of paid work in the week before they were interviewed or reported that they were temporarily away from their job)

2. report that their working arrangements in their main employment include some form of flexibility

3. recognise that the flexibility of their working arrangements is a result of being on a “zero-hours contract”

Therefore, the people identified by the LFS as being on a “zero-hours contract” will be those in employment who are aware that their contract allows for them to be offered no hours. This might exclude some people who select another option, such as on-call working, although they have the opportunity to report a “zero-hours contract” as well.

The latest estimate from the LFS shows that 801,000 people reported that they were on a “zero-hours contract” in the period October to December 2015, representing 2.5% of people in employment. This is 15% higher than the reported figure from the same period in 2014 (697,000 or 2.3% of people in employment). It is possible that this increase may be affected by increased recognition of the term “zero-hours contracts” rather than additional contracts themselves. However, it is not possible to estimate the scale of any such increased recognition.

When looking at the length of time in current job, over half of the increase in “zero-hours contracts” is from people in their job for more than a year, that is, they were already with their current employer in October to December 2014. This could reflect either increased recognition or people moving on to a “zero-hours contract” with the same employer. The number of people on a "zero-hours contract" who had been in their current job less than a year also increased. This could have been due either to a rise in the prevalence of “zero-hours contracts” or to increased awareness of the terms of the contract when people start work (see Figure 1).

Comparisons with 2012 and earlier years are complicated by a large increase between 2012 and 2013, which appeared to be mainly due to increased recognition of “zero-hours contracts”. This change was covered in a previous report we published on 30 April 2014.

ONS business survey

In November 2015 the ONS business survey asked a sample of 5,000 businesses how many people were employed on contracts that do not guarantee a minimum number of hours (NGHCs) and received around 2,300 responses. The latest estimates from the employer survey indicate that there were 1.7 million contracts that did not guarantee a minimum number of hours where work was carried out in the fortnight beginning 9 November 2015 (around 6% of all contracts). This total excludes contracts that do not guarantee a minimum number of hours where work was not carried out in the reference period.

Table 1 provides a summary of the main results of the five business surveys that have been carried out so far.

All of the estimates presented in Table 1 are subject to a degree of uncertainty so should be interpreted with caution (see Annex 1).

The differences between the estimates shown in Table 1 should be treated with caution because the surveys are from different times of the year and so may be affected by seasonal factors. Currently we do not have a sufficient time series to produce seasonally adjusted estimates. The only change not affected by seasonality is the annual change between January 2014 and January 2015, where the increase is not statistically significant. The survey reference periods have been changed in the last year to bring them more into line with the zero hours estimates derived from the Labour Force Survey, which relate either to April to June or October to December each year.

As well as the number of contracts, the ONS business survey also estimated that 10% of businesses make some use of NGHCs. However, the proportion of businesses using NGHCs differs when industry or size of business are considered. Figure 2 shows the proportion of businesses using NGHCs by size of business. It shows that over 40% of businesses with employment of 250 and over make some use of NGHCs, compared with around 10% of businesses with employment of less than 10.

Looking at industry, the proportion of businesses using NGHCs varies considerably (Figure 3). In Accommodation and Food Services more than 1 in 4 companies (26%) made some use of NGHCs in November 2015, compared with construction where around 1 in 20 companies made some use of NGHCs (5%). In both these sectors, use of NGHCs was lower than in May 2015, suggesting they may be affected by seasonal factors.

In addition to the 1.7 million contracts where work was carried out, in November 2015 there were also 2.0 million contracts where no work was carried out. In May 2015 there were 1.8 million contracts where no work was carried out. Following the business survey in January 2014, ONS carried out some follow-up work with businesses to gather some more information about these contracts. The follow-up work did not highlight any single reason why these contracts did not provide any work. A number of reasons were given including:

• Employees not accepting work offered to them for personal reasons. Employers may not know the exact reasons but it could include people not currently in a position to work (eg due to sickness, holiday etc) or those not available to the employer in the reference period because they had a permanent or no guaranteed hours job with a different employer

• Openings for work that employers had not been able to fill because the flexibility, or other aspects of the contract, meant that no-one had accepted the work

• Openings for work the employer might have at other times but, due to fluctuating demand, were not available in the reference period. It is not possible to say whether employers would have been able to fill these positions had they been available

• Some employers having some out-of-date contracts on their list, although businesses said they generally kept their lists up to date.

None of the businesses contacted in the follow-up used exclusivity clauses. While the work has proved inconclusive, it indicates that for most of these contracts the reason that work was not offered was due to the employees not accepting work or employers not having suitable work available. Therefore we will continue to lead on the estimate for contracts where work was carried out.

Comparison of LFS and ONS business survey

The number of NGHC contracts from the business survey is higher than the number of workers reporting “zero-hours contracts” in the LFS. The results of the business survey will differ from the LFS for a number of reasons:

• the employer survey will count contracts not people and will provide higher estimates (as one person can have more than one contract)

• employers are likely to be more aware of their employees’ formal contractual arrangements and this may differ from the perception of employees if their normal working hours are relatively stable or if changes in hours are mainly as a result of personal choice

• there may be multiple contracts for each job in the business survey

Figure 4 shows the distribution by industry of NGHCs from the ONS business survey and “zero-hours contracts” from the LFS. Where there are differences in the distributions, this will be partly due to how people are classified in the 2 surveys. On the LFS, people are self-classified to an industry. Businesses are allocated to the industry where most of their employees work. For example, many local authorities are classified to Education (Section P), while their employees will cover other areas such as social work (Section Q), public administration (Section O) and recreation (Section R). Similarly, employment agencies are classified to Administration & Support Services (Section N) on the business survey, while people employed by them but placed at another employer may give a different answer on the LFS. The distribution may also be affected by the business survey including second and subsequent jobs, whereas the LFS covers main employment.

While the business survey provides a measure of the number of contracts that do not guarantee hours, the LFS can provide additional information about the type of people who report that their main employment is on a “zero-hours contract”.

Who are they?

Looking at the type of people who report that they are employed on a “zero-hours contract” compared with other people in employment shows that there are differences in the type of people on “zero-hours contracts” (Figures 5 and 6). For October to December 2015:

• women make up a bigger proportion of those reporting working on “zero-hours contracts” (53%) compared with women in employment not on “zero-hours contracts” (47%)

• people who report being on a “zero-hours contract” are more likely to be at the youngest end of the age range, 38% of people on “zero-hours contracts” are aged 16 to 24 (compared with 12% for all people in employment not on a zero-hours contract)

• 23% of people on “zero-hours contracts” are in full-time education compared with 3% of other people in employment

These patterns may partly reflect the groups most likely to find the flexibility of “zero-hours contracts” an advantage, for example, young people who combine flexible working with their studies.

Hours worked and flexibility

The majority of people on “zero-hours contracts” (63%) reported that they worked part time, compared with 26% of other workers. This means that the average actual weekly hours worked in their main job by someone on a “zero-hours contract” is lower at 21.3 per week compared with the average actual weekly hours for all workers at 34.4. This shows a similar pattern to usual hours worked which were 25.8 and 36.7 respectively.

In October to December 2015, 17% of people on “zero-hours contracts” worked no hours in the week before their LFS interview compared with 12% of other workers.

Comparing usual and actual hours, Figure 7 shows the differences between actual and usual hours worked for people on “zero-hours contracts” and other workers. For October to December 2015:

• 44% of people on “zero-hours contracts” worked their usual hours compared with 57% of other workers

• 40% of people on “zero-hours contracts” worked less than their usual hours compared with 33% of other workers

• 16% of people on “zero-hours contracts” worked more than their usual hours compared with 10% of other workers

Around a third (37%) of people on “zero-hours contracts” want more hours compared with 10% of people in employment not on a zero-hours contract, though this could be linked to a higher proportion of “zero-hours contract” jobs being part-time. Looking in more detail, 15% of people on “zero-hours contracts” would like a different job with more hours compared with 5% for other people in employment (the remainder would like more hours in their current job or an additional job) (Figure 8).

Labour Force Survey

The estimate of 801,000 people employed on “zero-hour contracts” has a 95% confidence interval of ±64,000, which means the true figure is likely to lie between 737,000 and 865,000.

ONS business survey

The latest estimate of 1.7 million contracts that do not guarantee hours and where work was carried out has a 95% confidence interval of ±425,000, which means the true figure is likely to lie between 1.3 million and 2.2 million.

As part of the movement from experimental to official statistics, every aspect of the methodology for business survey is currently being evaluated. The sample design will be revised for the next round of data collection; this is likely to impact on estimates for the number of NGHCs worked in a particular fortnight.