While the evidence is mounting that zero-hour contracts are here to stay, we shouldn’t overplay their impact on the labour market: they still affect only a very small minority of workers and some value the flexibility they bring, writes Laura Gardiner.
For many, zero-hours contracts (ZHCs) are emblematic of the UK’s labour market experience during the financial crash, contributing to stronger-than-expected employment figures but also symbolising rising job insecurity. But with the economic recovery now gaining ground, the key question has been whether they’d fade away or remain as a permanent feature of the labour market. New ONS figures suggest that zero-hours contracts are here to stay.
While ZHCs have been captured in the ONS survey of employees for many years, for the first time we have comparable data from a survey of businesses. Estimates from both surveys are shown on the chart below, with year-on-year comparisons connected by the lines.
With an array of comparable data now available, a definitive view on the level of zero-hours contract working – which has proven elusive up to now – might be expected. But the chart above shows that the true extent of ZHCs remains far from clear. So what are and aren’t we able to take away from yesterday’s release?
- First, there are entirely plausible reasons why the level of ZHC working appears quite different in the survey of employees and the business survey. Principally, the LFS captures workers in their main job only, while the business survey captures contracts, and workers can hold more than one contract at a time. In addition, the ONS highlights that there may be some undercounting in the LFS, because businesses tend to be more aware of their employees’ contractual arrangements than employees themselves.
- Second, and stemming from the above, increasing recognition of ZHCs following extensive media coverage may have driven some of the previous increases observed in the employee survey (but is less likely to have affected the business survey results). In particular, the steep increase during 2013 tallies with the height of ZHC hype. While the ONS argues that rising recognition may still be driving the upward trend in this data, we would suggest this effect is likely to have faded as the concept has normalised.
- Third, the sharp increase in contracts with no guaranteed hours where no work was carried out is a bit of a conundrum. The ONS has previously conducted some follow-up work on what these entail: this proved inconclusive but its judgement is that they capture a mix of employers not offering any work and employees choosing not to accept work that was offered. Given that these two explanations reflect a completely different balance of power between workers and businesses, we suggest that understanding their relative prevalence should be a central focus of future analysis of ZHCs.
- Finally, taking the above points and caveats together, we can be (relatively) confident in concluding that the prevalence of ZHCs rose through 2014 and early 2015, or was at the very least maintained. While all the lines on the chart are rising, the increase in the central measure (contracts with no guaranteed hours where work was carried out) is well within the margin of error. A degree of caution is therefore necessary in interpreting the stats, but it’s clear that ZHCs aren’t disappearing.
While the evidence is mounting that ZHCs are here to stay we shouldn’t overplay their impact on the labour market: they still affect only a very small minority of workers and some value the flexibility they bring. However, for many they represent an undesired and unstable employment offer, and the argument that this is a recession-related problem that will take care of itself no longer holds true.
Therefore, as we’ve recently set out, the case for targeted policy action to ensure that the benefits of labour market flexibility are balanced appropriately against protecting workers’ rights is increasingly compelling.
Laura Gardiner is Senior Research and Policy Analyst at Resolution Foundation.
Cheap labour