Last January we (and everybody else) reported the unemployment rate was 7.2%.
But some people have always been sceptical of the official definition of unemployment, and suggested the "true" rate of unemployment is always higher than the standard measure indicates. A new piece of work from two Central Bank economists suggests they may have a point.
And that is important for policy makers – not just in creating conditions in which those who want to work can work, but also in gaining a better understanding of the real amount of potential additional labour force that exists here.
That’s because a bigger potential labour force means employment can continue to grow strongly a little bit longer than previously thought, before it starts to put pressure on wage rate inflation. And that is something the European Central Bank is interested in.
Indeed, the ECB published a note on labour market slack in the Euro area earlier this year which compared with the narrower definition of unemployment, concluding that the higher level of under-utilised labour in the Euro area is likely to result in a continuation of the trend of low wage rises.
That has pretty big implications for the ECB’s inflation targeting strategy – bond buying, negative interest rates and all.
Using broader measures of non-employment (as opposed to unemployment), Central Bank economists have constructed a Non-Employment Index - and it shows a non-employment rate in December 2016 at 9.4% - just over two percentage points higher than the unemployment rate.
This higher rate also includes the category "part-time underemployed". These are people who are working part time but want extra hours.
Including that category increases the non-employment index significantly (especially considering their probability of transitioning to full-time employment appears to be quite low – at 3.7% - compared with say the short-term unemployed, with a transition probability of 16.3%).
The idea of using broader measures of non-employment to construct an index comes from the United States, where research two years ago led the San Francisco Fed to publish a non-employment index on a monthly basis.
Here, Stephen Byrne and Thomas Conefrey of the Central Bank followed the US methodology closely, applying it to the large amount of data gathered by the CSO’s quarterly national household survey between 1998 and 2016.
Because that data tracks various categories of employed and non-employed individuals, the researchers were able to construct a weighting system to gauge the probability of a person moving into employment from one of several categories of non-employment.
More importantly, it shows that the flow of people every quarter from inactivity to employment is higher than the flow from unemployed to employed.
The two measures are not directly comparable – NEI is calculated as a percentage of the working age population, while unemployment is a percentage of the labour force. But it does give us additional information about the large number of people who are not in work, but who do not meet the standard definition of being unemployed.
The labour force comprises the population aged 15-74 who are either employed or unemployed.
However, the unemployed are only a subset of the working age population that are not in work (that is, non- employed). Some of these people are "discouraged" job seekers, who have become dispirited about ever getting back to work; others have entered education or training. But the biggest group – in boom time or bust – are those listed as "Does not want a job" – a fairly broad category which includes retired as well as those who simply, as the description suggests, do not want a job.
The Irish research shows that the NEI rate was higher than the standard unemployment rate for all the ‘Celtic Tiger’ period (by about 2.5 percentage points), suggesting that the degree of underutilisation in the labour market was higher than suggested by the standard rate (remember, this was when there was very large scale inward migration, ostensibly to fill labour shortages).
Both measures moved up sharply in 2008, when the financial crisis became a great recession but interestingly, the standard unemployment rate went up higher than the NEI rate. The latter peaked at 12.1% in late 2009, while the standard unemployment rate peaked at 15.1% in Q4 2011.
This is mostly explained by the fact that the unemployment rate is measured against the labour force – which reduces in size in a downturn as people leave the labour force by for example entering education or training. This mechanically causes the unemployment rate to grow more rapidly in a downturn.
By contrast, the NEI uses the working age population as a denominator, which doesn’t shrink in the same rapid way that the labour force does, leaving a smaller impact on changes in the non-employment rate.
A second factor is to do with the rise of the number of long-term unemployed during the crisis years.
The Standard method accord long-term and short-term an equal weighting, but the NEI methodology down weights the long-term unemployed due to their lower observed job finding probabilities.
For this reason the rate of increase for NEI is lower than the rate of increase for standard unemployment.
Since 2012, the NEI had declined – but at a slower rate than standard unemployment. The NEI for Q4 2016 was 7.9% - only marginally higher than its average value in 2007 of 7.7%. Adding in part-time under employed people (only measured since 2008) gives a higher NEI in December 2016 of 9.4%.
The authors conclude that "Both estimates of the NEI imply there may be scope for further employment growth in the short run before significant labour shortages and associated wage pressure become apparent".
They say that there are very few examples of NEI’s for other countries, but one directly comparable example is the United States, because of that monthly data from the San Francisco Fed.
The authors do compare, and find that the Irish NEI was higher than the US NEI in 1999, but fell below the US rate from 2001 to 2008. It then moved ahead of the US rate during the crisis years.
Both rates declined, but in recent years the Irish rate has declined at a faster pace. In the final quarter of last year, the Irish Non Employment Index dipped below the US rate once again.