In charts: UK self-employment

Britain’s flexible labour market has worked relatively well since the financial crisis. Predictions of mass unemployment were largely confounded and instead the country has seen declining labour productivity and wages and growing self-employment and part-time work.

A report out today from the Resolution Foundation asks whether or not the growth in self-employment is a permanent structural change in the economy or a temporary stop-gap that will disappear when the economy recovers.

This is one of the questions the Bank of England’s Monetary Policy Committee will be considering this week as they debate the possibility of interest rate rises and the amount of slack in Britain’s labour market.

Providing evidence for a structural change, self-employment has been increasing since long before the crisis. The number of self-employed has risen by just over a third, from 3.3m in 2003 to 4.5m at the latest count.

But bolstering the cyclical side is the rapid increase in self-employment over the last few years — about half a million since 2011 — and even excluding this increase Britain’s post-crisis experience was an outlier among the OECD.

Only London has seen an increase in employee jobs, all other job growth has come from self-employment.

This growth has come at the expense of income. The Resolution Foundation estimates the self-employed earn about 40% less than an employed worker. This is after a 20% fall between 2007 and 2012 compared to only a 6% fall in the wages of the employed. Although much of the decline in the average is driven by composition effects as well, part-time self-employment is increasing and more women — who tend to be paid less — are increasingly likely to be self-employed.

Before the crisis the self-employed tended to be more over-worked than the employed, since the two groups have converged and underemployment is slightly more common among the self-employed.

Despite the lower wages the majority of those who have become self-employed within the last five years would still chose to be self-employed, although this is lower than those who have been self-employed for more than five years.

And the same pattern is seen for their choices of self-employment over being an employee.

But the difference in the feelings of the two groups is not necessarily any evidence of a change in who is becoming self-employed. It’s likely that those who became self-employed before 2009 and would rather have been an employee are a lot less likely to still be self-employed in 2014 when the interviews took place. Perhaps if Ipsos MORI had taken the same survey five years ago they would have still seen this difference in the two groups of the self-employed.

Ultimately there probably isn’t yet enough data about the experience of the self-employed to be certain about whether the growth is structural or cyclical — the self-employed aren’t included in many national statistics, including official data on wages — and anecdotes about entrepreneurial Brits taking advantage of the rise of the internet compete against stories of only nominally self-employed freelancers looking to save on tax. While both theories have more than an element of truth they won’t provide a great deal of help the MPC.