We’re still in the dark about insecure work

Before the coronavirus pandemic hit, it was evident that jobs were plentiful in much of the developed world. It was harder to know how secure they were. The economist Dani Rodrik diagnosed “a chronic state of shortage of good jobs.” Another economist, Guy Standing, described a growing “precariat” of insecure workers in many countries.

Others disagreed. In a paper last year, economists Alan Manning and Graham Mazeine concluded that “job security has remained remarkably constant” in the US, the UK and Germany since at least the turn of the 21st century. The Economist argued this month that pessimism about the world of work was fashionable but unjustified by the facts.

Job insecurity matters because the consequences reverberate in every area of life. It makes it harder to keep a roof over your head, raise a family, progress at work, stay out of debt and keep healthy. But the truth is, we don’t really know whether insecurity is on the rise because we have been very bad at measuring it.

Official statistics agencies in most countries count the numbers who are employed, self-employed, on temporary or agency contracts and so on. Some observers have conflated a rise in non-standard jobs with a rise in insecurity. But this is lazy and imprecise. A part-time worker might well enjoy all the same rights and protections as a full-time worker but simply not work on Fridays. An agency worker might be a locum doctor who has plenty of work and is free to turn down the shifts she doesn’t fancy.

There is some long-running survey data that focuses directly on job security, but it tends to use one narrow definition: how likely respondents think it is that they will lose their job. It is this data that Manning and Mazeine studied in the US, UK and Germany. It shows a lot of stability over the long-run (with understandable peaks at times of high unemployment).

Yet, as the authors acknowledge, fear of losing your job is only one facet of insecurity, and not always the most salient one. A care worker I interviewed in the UK, for example, didn’t worry that she would lose her job but still felt insecure because her employer changed her hours constantly with little notice.

She could be given 40 hours one week and 20 hours the next. She had to check her phone throughout the work day to see if she had been given more clients to visit. Sometimes she was overworked; other times she was underworked. “I just don’t know whether I can pay my mortgage at the end of each month. There’s no stability,” she said.

The UK’s Office for National Statistics measures the number of people on “zero-hours contracts”, which do not guarantee any hours of work each week. The latest data suggests that these account for about 3 per cent of the workforce, up from 0.6 per cent in 2007 before the 2008 global financial crisis. But this data set misses the extent of this type of work. The care worker I spoke to, for example, was on an “8-hour contract”. I once wrote about “one-hour contracts” advertised by the bank Santander.

In an attempt to fill in this gap in the official statistics, Britain’s Living Wage Foundation published the results of two surveys last week, based on 2,000 responses each, which found that almost two-fifths of UK workers were given less than a week’s notice of their shifts or work patterns. Among full-time low-paid workers, 55 per cent were given less than a week’s notice and 15 per cent less than 24 hours.

This wouldn’t be problematic for some, but it’s hard to imagine that it would be everyone’s preference. (Indeed, it would be useful the next time the survey is run to ask the respondents whether they are happy with the arrangement.)

In a similar vein, the Resolution Foundation analysed 7m anonymised bank accounts in 2018 and found that more than 80 per cent of low earners with steady jobs had volatile pay, compared with two-thirds of higher-earners.

These snapshots are not perfect and they can’t tell us whether the trends are waxing or waning. But the scale of the instability they suggest, particularly for low-paid workers, should serve as a jolt to official statistics agencies to capture this sort of data themselves at regular intervals. These new supplementary metrics should focus less on what people’s contracts say and more on the reality of their working lives.

There is plenty of debate about how to improve job insecurity as we emerge from the pandemic. We will arrive at better answers if we start by asking the right questions.

sarah.oconnor@ft.com