A lump of fallacy, or two plus two does not equal five

This is a joint post from Christopher Brooks, Employment and Skills Policy Adviser, and Jose Iparraguirre, Chief Economist.

A worrying trend has recently emerged among the policy commentariat, which is taking aim at older workers and laying the blame for the high rates of youth unemployment at their door. Yes, the debate around ‘job blocking’ has emerged once again, in spite of academic and mainstream policy thinking having discredited this theory many, many years ago.

Every now and again the ‘job blocking’ argument manifests itself in policy. Most notably for our purposes, in 1970s Britain there was a predominant view that older workers directly ‘blocked’ younger workers from entering the workforce. The Government of the day introduced the Job Release Scheme in response, which incentivised early retirement on the basis that the unemployed youth would then move into the vacancies. The scheme ran for around a decade before ending as a total and spectacular failure – not only had the scheme failed to improve youth unemployment, but it had proven that if anything there was a direct correlation between high employment rates among older workers  and high employment rates among younger workers.

Since then, thankfully, thinking among governments and policy makers has moved on, and the real relationship and causes – that the labour market is highly flexible, and that there is precious little competition for jobs between young and older seekers – have been recognised.

But every now and again the original theory rears its ugly head. Indeed, it is so well-known that economists have even coined a term for it: the ‘lump-of-labour’ fallacy. For some background, this theory has been around since the 19th century, and it seems to die hard (according to Nobel-prize winning economist Paul Krugman, it tends to bounce back whenever the economy contracts).

Firstly, what it is all about? During this current contraction of economic activity employment levels among older people have been going up whereas the number of younger people in employment has been falling. This is true both for men and women (especially above state pension age).

Although, if we look at particular quarters other age groups do better, the charts above show the overall picture over the last three years.

So what’s wrong with the ‘job blocking’ argument in the current context? The implication is that as older employment has grown while youth employment has stalled or declined, older workers must have crowded their younger counterparts out of the labour market. We have seen in recent weeks continued growth in 65+ employment and relatively good figures for the 50-65 age group, which when combined to falling rates of youth employment shows a direct crowding out younger workers by their older ‘colleagues’. Right?

Wrong. This is the latest incarnation of the ‘lump-of-labour’ fallacy. In the words of Banks et al, from the Institute for Fiscal Studies:

Taken to the extreme, this view would support that the idea that a high level of employment of one group of individuals can only be at the expense of another group: if, for instance, were the population of a country to increase, younger individuals would be unemployed as older individuals would not “release” enough jobs for the new entrants. The absurdity of this view in the long term is simply seen by considering the fact that the size of a country does not bear any relation to the share of population unemployed.

Needless to say, we should look beyond theoretical battlefields and try and find whether any ‘hard’ evidence may tip the balance towards one side or the other.

A book recently published by the US National Bureau of Economic Research collates academic papers from different countries investigating whether older workers substitute or crowd out younger workers. Not a shred of evidence is found.

Not in the United States:

We develop a measure of the variation over time in the incentives for retirement of the elderly and relate that to the labor supply of both the elderly and younger workers. Overall, our data suggest little substitution across these groups.

Not in Canada:

We find little visual evidence that trends in elderly labor force participation have had an impact on the labor markets of the younger. Finally, our regression evidence has shown that—if anything—the employment of the young and prime aged tends to move in the same direction as the elderly.

Not in France:

Evidence of the correlation between youth labor market outcomes and old workers’ labor force participation plead more in favor of a positive association between young and old workers in the labor market. An increase in the old workers’ participation is indeed correlated with an increase in the employment rate of young workers and a decrease in their unemployment rate.

Not in Germany; not in the Netherlands; not in Sweden. The list could go on – and all of it points to one conclusion. ‘Job blocking’ is a myth; the ‘lump-of-labour’ is indeed a fallacy. Employment of older workers has increased. True. Employment of younger people has come down. True. The second phenomenon is a direct consequence of the first. False.

Some commentators, however, are putting two and two together and making five.

3 thoughts on “A lump of fallacy, or two plus two does not equal five”

  1. This fallacy is entirely consistent with the late 20th century tendency towards ‘victimhood’ – if someone or some group is unemployed then ‘it is not their fault’, ergo it has to be someone else’s fault. It is seen always as a zero sum game. However, I recall the words of Henry Mencken -‘For every complex problem there is a simple, neat solution…and it’s wrong!’

Leave a Reply to Tony Armstrong Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s