Zero-hours contracts: The trouble with the facts

By Ian Brinkley, Chief Economist, CIPD

The latest estimates around zero-hours contracts confirm how difficult it has been to measure the number of people on these or similar contracts, let alone how much that volume has grown in recent years. The reality is that we cannot pin down these numbers with any confident degree of certainty.

In attempting to piece the picture together, we have two measures. The first is from the Labour Force Survey (LFS) which tracks individual responses. In the period Dec 2017–Feb 2018, just over 900,000 people or about 2.8% of the workforce said they were on a zero-hours contract. The second, more recent measure is the Business Survey (BS), which showed that as at Nov 2017, employers had issued 1.8 million ‘non-guaranteed hours contracts’ or NGHCs to use the Office for National Statistics (ONS) label.

There are good reasons to suppose that the number of contracts issued would be greater than the number of people reporting to be on them. For example, some people will hold more than one zero-hours job. Moreover, these are different surveys using different methods and asking different sources, so some differences would have to be expected. Yet the gap is so large that it’s highly likely that the LFS has underestimated the total number on zero-hours contracts. This may not be a surprise as ‘zero-hours’ has no legal status and seldom appears as a phrase in employment contracts.

What we know

The LFS showed that the number of people reporting zero-hours did not change very much between 2000 and 2012, but then exploded from 2012 onwards, increasing two and half times between 2012 and 2016, from 0.8% of the workforce to 2.8%. Recent figures show that both the share and level have now stabilised.

Meanwhile, the BS showed a slower but still significant increase in contracts issued – 22% between 2014 and 2017 – with the share of all contracts issued edging up slightly from 5% to 6%. The BS also showed that the share of businesses using such contracts declined significantly over the same period – from 13% to 6%. This would be consistent with fewer businesses using them but with those that did intensifying their use.

The table below sets out these recent figures.

Zero-hours working 2012–17

Year of survey






Labour Force Survey (LFS) 000s people






Business Survey (BS) 000s contracts












LFS – share of workforce






BS – share of contracts












BS – share of businesses






Note: LFS is Q4 all years. BS is Jan 2014 then November 2015 to 2017. All figures seasonally unadjusted. Source: Office for National Statistics

What we don’t know

It is unfortunate however, that neither survey provides an accurate guide to recent changes.

Taking the LFS first, it is very unusual to see such rapid and large increases in labour market figures outside of recessions, and it would be stretching credibility to imply that UK employers suddenly and spontaneously turned to this form of contracting almost overnight. The ONS says that much of the apparent increase was due to more accurate reporting in response to extensive media coverage and campaigns run by trade unions and other organisations.

In the meantime, the BS has been beset by problems of consistency and methodology. For example, it would not be safe to compare the most recent figures as the 2016 survey was voluntary and the 2017 survey compulsory, so the widely reported increase is in fact, inaccurate. Other apparent trends such as the concentration of zero-hours contracts in fewer businesses may also be suspect. The ONS is discontinuing the BS but says it cannot afford to replace it with a more soundly based one.

The CIPD and others have noted that the share of atypical working – self-employment, temporary working, family workers, and government employment schemes – in the economy has barely changed over the last twenty years at just over 20%. The latest employment figures confirm that all the employment growth over the past 12 months has come from permanent employees, mostly working full time. This is shown in the chart below.

Change in employment over previous 12 months

Note: all figures 12 months to Dec-Feb 2018 except zero hours which are 12 months to Oct-Dec 2017. UK, seasonally adjusted. “Other” are family workers and those on government employment schemes. Source: Office for National Statistics Labour Market Bulletin, April 2018

Zero-hours contracts still remain controversial, though they seem to be relinquishing top spot to the gig economy as a symbol of everything wrong in modern labour markets. The loss of the Business Survey will diminish our knowledge a little. Yet it is hard for the ONS to justify spending large sums on surveying a form of employment which covers less than 3% of the workforce and which as things stand, looks unlikely to become more important in the future.

Thank you for your comments. There may be a short delay in this going live on the blog page as we moderate the comments added to our blogs.