{"id":40152,"date":"2018-08-20T18:43:55","date_gmt":"2018-08-20T08:43:55","guid":{"rendered":"https:\/\/billmitchell.org\/blog\/?p=40152"},"modified":"2018-08-20T18:43:55","modified_gmt":"2018-08-20T08:43:55","slug":"british-labour-market-low-unemployment-hiding-a-deeper-malaise","status":"publish","type":"post","link":"https:\/\/billmitchell.org\/blog\/?p=40152","title":{"rendered":"British labour market – low unemployment hiding a deeper malaise"},"content":{"rendered":"

\t\t\t\tThe British Office of National Statistics (ONS) published its latest labour market data last week (August 14, 2018) – Labour market economic commentary: August 2018<\/a>. The results are illuminating because they demonstrate how we must use broad concepts to appraise labour market health rather than just focusing on the official unemployment rate. A marked characteristic of the British labour market has been the near zero rate of growth in wages (and falling real wages) for some years, despite the claims that employment has reached record levels and the unemployment rate is now at levels unseen since the early 1970s. The research question is to dig deeper into the data to see what might be driving these relationships. The conclusion I reach is that the quality of employment has fallen somewhat over the last decade or so and the capacity of workers to successfully achieve wage settlements has fallen significantly as trade union membership has gone south. I am doing more formal research on this question to narrow down all the determinants and will report when I have more to say.
\n<\/p>\n

Employment – at record levels<\/h2>\n

The heading reflects the sort of statements that politicians make. In this case it is true. Total employment in the June-quarter was 32,386 thousand and it has never been higher.<\/p>\n

But then the total working age population (above 16 years) was 53,044 thousand and that has never been higher.<\/p>\n

So the statement “employment at record levels” is somewhat vacuous no matter how good it sounds.<\/p>\n

We are better off considering rates and the Employment to Population ratio shown in the next graph is at levels not witnessed since the June-quarter 1973, just before the OPEC oil crises hit.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Part-time, self-employed and temporary<\/h2>\n

The data also shows that the proportion of workers who are now in part-time positions has risen from 23.5 per cent in the June-quarter 1992 to 26.4 per cent in the June-quarter 2018 – not a significant shift.<\/p>\n

There has also not been much change in the proportion of workers working second jobs over that period and there has been a decline in the proportion of temporary employees as a percent of all employees.<\/p>\n

Further, contrary to popular perception, the proportion of total workers who are self-employed has barely shifted over this 26 year period – 13.6 per cent in the June-quarter 1992 to 14.7 per cent in the June-quarter 2018.<\/p>\n

But of those, the proportion of self-employed who are working part-time has risen from 17.6 per cent in the June-quarter 1992 to 30.1 per cent in the June-quarter 2018 – a large shift and likely to be hiding a significant shortfall in labour demand as people eke out a working life as best they can.<\/p>\n

Zero hour contracts<\/h2>\n

The ONS data – EMP17: People in employment on zero hours contracts<\/a> (Last updated: August 14, 2018) shows that the proportion of workers in this state has risen from 0.8 per cent of all workers in the December-quarter 2000 to a peak of 2.8 per cent in the December-quarter 2017.<\/p>\n

In the first six months of 2018, the proportion has dropped to 2.4 per cent.<\/p>\n

The following graph shows the history.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Unemployment rate down to early 1970 levels<\/h2>\n

The next graph shows the official unemployment rate from the first-quarter 1971 to the June-quarter 2018. It was 4 per cent in the most recent quarter.<\/p>\n

It was last at that level in the March-quarter 1975, just before the then Labour government declared Keynesian economics was dead!<\/p>\n

So on this aggregate Britain is performing well and there are no hints that the Brexit referendum has altered the steady fall in official unemployment since the most recent peak in the December-quarter 2011.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Hours worked – steady<\/h2>\n

The next graph shows the total weekly hours worked in Britain (in millions of hours) from the March-quarter 2005 to the June-quarter 2018.<\/p>\n

Since the pre-GFC peak, hours worked have only grown by a total of 8 per cent overall and only 1 per cent since the end of 2016.<\/p>\n

In the June-quarter 2018, there was a 0.1 per cent decline in hours worked recorded.
\nBritain<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Real earnings flat<\/h2>\n

The earnings statistics tells us that the labour market is not as strong as the narrow aggregates (such as the official unemployment rate) might suggest.<\/p>\n

The next graph shows real average weekly earnings (expressed in 2015 constant prices) from January 2005 to June 2018.<\/p>\n

This is not what a fully employed labour market looks like.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Underemployment<\/h2>\n

ONS also released their latest estimates of – EMP16: Underemployment and overemployment<\/a> – last week (August 14, 2018).<\/p>\n

Underemployment refers to part-time workers who desire more hours of work.<\/p>\n

The following graph shows the evolution of the underemployment rate from the March-quarter 2002 to the June-quarter 2018.<\/p>\n

While its trajectory is downwards, the current rate of 7.4 per cent remains above the pre-GFC level of 7 per cent (June-quarter 2008).<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

Trade Unions in decline<\/h2>\n

One of the key shifts in the British labour market has been the decline in trade union membership after Margaret Thatcher’s elevation to government.<\/p>\n

The – data<\/a> – shown in the following graph reveals the decline.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

What does this look like in terms of coverage ratios – that is, membership as a proportion of all employees?<\/p>\n

The data shows that the proportion has dropped from 38.6 per cent in 1989 to 22.9 per cent in 2017 (for Great Britain).<\/p>\n

In 1995, the coverage ratio was 32.4 per cent overall (all employees), 21.4 per cent for the private sector, and 61.3 per cent for the public sector.<\/p>\n

By 2017, the ratio had dropped to 23.3 for all employees (in the UK), 13.5 per cent in the private sector and 51.8 per cent in the public sector.<\/p>\n

This has translated into a significant decline in the proportion of employees who have their wages and conditions covered by a collective agreement.<\/p>\n

In 1996, this proportion was 36 per cent (overall), 23.2 per cent (private) and 74.4 per cent (public).<\/p>\n

By 2017, it was 26 per cent (overall), 15.2 per cent (private) and 57.6 per cent (public).<\/p>\n

So the significant shifts have been in the way the successive governments have sought to hollow out the protections for workers employed in the public sector.<\/p>\n

Phillips Curve variations<\/h2>\n

What is very surprising from my perspective is the way the indicators of labour underutilisation interact with the British inflation rate and the wage inflation rate.<\/p>\n

The following two graphs tell the story and tell me that there has been a dislocation between the drivers on wage inflation (which has stalled around zero) and the rate of change in the Consumer Price Index (general price inflation).<\/p>\n

The first graph shows the relationship between the two labour underutilisation indicators (unemployment and underemployment) pre-GFC and post-GFC and the annual inflation rate (CPI-based).<\/p>\n

The straight dotted lines are simple linear regression trends.<\/p>\n

While the relationships have shifted out somewhat post-GFC, the degree of labour underutilisation does not appear to be a disciplining factor on the inflation rate in Britain over this 15-year period (the period where data was comparable).<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

The next graph shows the relationship between the two labour underutilisation indicators (unemployment and underemployment) pre-GFC and post-GFC and the annual growth in private average weekly earnings (wage inflation).<\/p>\n

Now we see that post-GFC, the degree of labour underutilisation both official unemployment and underemployment has shifted and is now exerting a downward pressure on wage inflation.<\/p>\n

\"\"<\/a><\/p>\n

<\/div>\n

These summary results have raised a number of research questions which I will investigate when I get time.<\/p>\n

Gross Flows and Transition Probabilities<\/h2>\n

Instead of just looking at the raw employment and unemployment data, I find it useful to examine the gross flows data – which shows the flows between labour force categories.<\/p>\n

What you realise is that underpinning the quarterly movements in the aggregates are massive flows of workers between the various labour force categories.<\/p>\n

The latest gross flows data from ONS was published last week (August 14, 2018) – Labour market flows: August 2018<\/a>.<\/p>\n

ONS summarise the data in this way:<\/p>\n

\n– The gross flow to unemployment has decreased to its lowest since October to December 2001.<\/p>\n

– The gross flow to employment has decreased to its lowest since July to September 2011.<\/p>\n

– The gross flow from inactivity has decreased to its lowest since April to June 2009.\n<\/p><\/blockquote>\n

I updated my British gross flows dataset for the UK today (which takes us to the June-quarter 2018) and computed updated Transition Probabilities<\/strong>, which are derived from the Gross Flows labour market data.<\/p>\n

To fully understand the way gross flows are assembled and the transition probabilities calculated you might like to read these blogs – What can the gross flows tell us?<\/a> and More calls for job creation – but then<\/a>. For earlier US analysis see this blog – Jobs are needed in the US but that would require leadership<\/a><\/p>\n

Gross flows analysis allows us to trace flows of workers between different labour market states (employment; unemployment; and non-participation) between months. So we can see the size of the flows in and out of the labour force more easily and into the respective labour force states (employment and unemployment).<\/p>\n

Each period there are a large number of workers that flow between the labour market states – employment (E), unemployment (U) and not in the labour force (N). The stock measure of each state indicates the level at some point in time, while the flows measure the transitions between the states over two periods (for example, between two months).<\/p>\n

The net changes each month – between the stock measures – are small relative to the absolute flows into and out of the labour market states.<\/p>\n

The various stocks and flows are denoted as follows (single letters denote stocks, dual letters are flows between the stocks):<\/p>\n