LSE - Small Logo
LSE - Small Logo

Blog Admin

May 29th, 2012

It is time to move away from policy witchcraft and into an era where evidence is taken seriously

1 comment

Estimated reading time: 5 minutes

Blog Admin

May 29th, 2012

It is time to move away from policy witchcraft and into an era where evidence is taken seriously

1 comment

Estimated reading time: 5 minutes

Adrian Beecroft’s report on employment law has angered many who feel his recommendations are partisan and seek to erode a number of worker’s rights. John Van Reenan argues that although the report is both timely and addresses key issues in regulation, Beecroft appears to have ignored the wealth of research into the economic effects of employment laws. This evidence goes against his view that relaxing employment laws will generate large economic improvements in the UK.

Adrian Beecroft’s report on employment law makes some radical recommendations to deregulate UK employment laws, for example making it much easier to dismiss workers. When reluctantly released in mid-May, the report generated a political row as the author is a major donor to the Conservative party and provoked Vince Cable, UK Secretary of State for Business and Innovation, into dismissing the “bonkers” proposals. In reply Beecroft labelled Cable an anti-business socialist.I am sympathetic with the need to reform needless regulations and there are sensible proposals in Beecroft to make it easier to employ foreign workers through faster work permit checks and simplifying the immigration system. At a time when the UK is being perceived to be an unwelcome home to global talent this is to be welcomed.

But the most depressing thing about the report and the rancorous public debate around it is the total lack of any evidence on the likely impact of the proposals. Beecroft claimed that GDP would increase by 5 per cent through employment law reforms – a huge increase which would restore UK output from its current position of under 4 per cent below 2008 output, back to pre-crisis levels.  Unfortunately, this 5 per cent appears to be plucked from thin air, as if by witchcraft. There is no evidence presented at all in the Beecroft report to gauge such benefits. All I could find was a breezy sentence or two like “Quantifying the loss of jobs arising from the burden of regulation is an impossible task.” (p.6).

Putting aside the fact that this statement contradicts the claim of an additional 5 per cent growth, there is in fact a substantial body of evidence on the economic effects of employment laws. Indeed, it has been one of the most studied areas in economics over the last decade or two. It is unclear whether Beecroft is ignorant of this work or thinks it is entirely wrong or irrelevant. But as I discussed on the BBC’s “More or Less” this weekend here is a more detailed summary of the empirical evidence which, in short, does not support Beecroft’s assertions. For the more wonkish readers I offer some of the theory later after first discussing the empirical evidence

 Empirical Evidence of the Effects of Employment Protection Laws (EPL)

The main empirical evidence on the Effects of Employment Protection Laws (EPLs) comes from using indicators of the “toughness” of these laws and regulations. These are quantified by various organizations such as the OECD and the World Bank. The OECD’s latest indicators are in Figure 1 below, for example. They show that the UK had the lowest degree of protection in Europe and third lowest of all the countries examined by the OECD (only the US and Canada had weaker protection). This suggests that reforming EPL is hardly the Number 1 priority for raising UK growth rates.

Figure 1: OECD Index of the strength of Employment Laws. UK third weakest protection of workers, latest data (compiled in 2010)

Notes: Data are for 2009 for France and Portugal and 2008 for other countries. OECD average is the unweighted average for the 30 countries that were members of the OECD in 2008. Source is available online

Nevertheless, maybe some relaxing of EPL could have some more modest positive effects. The typical approach to examine this question is to analyze what happens when employment laws are weakened and see whether this is associated with any increase in unemployment, controlling for other factors that could influence joblessness such as the overall state of demand and the generosity of unemployment benefits, etc. Early approaches looked at a point of time, but more recent approaches look at changes over several years (both upwards and downwards) which is better as it controls for many country-level factors (e.g. culture).

The bottom line? The conclusion of the vast majority of studies is that there is no significant effect, i.e. tougher EPL does not increase unemployment (but neither do they reduce them). Don’t just take my word for it – this was also the conclusion of the OECD’s voluminous 2012 summary on growth policies (e.g. p.173).

This approach was pioneered by Professor Stephen Nickell, now at the Office of Budget Responsibility, when he was at the Centre for Economic Performance. For example see his summary in the magisterial Unemployment book or classic 1997 Journal of Economic Perspectives article.  His most recent paper on this looked at 20 countries (including the UK) for over three decades (1961-1995) finding essentially zero effects of EPL on unemployment. By contrast, other labour market institutions such as the duration and level of benefit durations, the effectiveness of the Employment Service, training polices and unions were important in affecting unemployment.  It is here, rather than EPL, which should be the main targets for reform.

Griffith et al (2007) took a similar approach to Nickell using more recent data between 1986 and 2000. Their main results (e.g. Table 3 columns 1-4) actually found that tougher EPL reduced unemployment, but this effect was statistically insignificant (i.e. they could not reject a zero effect). By contrast, toughening competition in the product market had a robust effect on reducing unemployment. It would be very surprising if Beecroft and his team were unaware of this paper as one of the co-authors, Rupert Harrison, is currently the Chancellor of the Exchequer’s special economic advisor (in the interests of full disclosure I should mention that I was also Rupert’s PhD supervisor). Perhaps communications between No 10 and No 11 are not so much better than they were in the Brown-Blair years after all.

Firm-level Evidence
Beecroft could argue that all this macro evidence is flawed and could cite reasons such as (1) measuring the laws is so hard and (2) there are so many other things happening at the country-level it is impossible to adequately control for them. Of course, these problems could bias the effect upwards to over-estimate the negative effect of EPL as well as to underestimate them. Nevertheless, it is useful to look at the growing micro-economic literature which examines firms and plants rather than entire economies.

This micro evidence does not give much comfort to Beecroft either. There do appear to be affects of EPL on firm behaviour, but it is hard to find that these are sufficiently large in magnitude to believe that his recommendations would have any major positive effect on GDP.

Let me take one study I have been involved in which looks at the case of France where EPL increases very substantially when firms reach 50 employees. French firms respond to the increase in costs by choosing to remain small – there is a big spike of firms who are just below the threshold at 48-49 employees to avoid the regulation.  This reduces total output as many of these firms would like to grow and employ more people but are deterred from doing so by the regulation, just as Beecroft alleges. But this effect is not huge – very efficient firms will still choose to pay the regulatory “tax” and become large. For example, a 1,000 employee firm is not likely to shed 950 of its workers simply to avoid the regulation.

There is a similar story in other countries. Mario Monti has (correctly in my view) been involved in a drive to reduce the heavy firing costs in Italian firms who are larger than 15 employees. As in my French study, Torrini and Schivaldi found that firms just below the 15 cut-off are much less likely to grow. But again, in aggregate the effects are relatively modest and cannot cause large changes in employment or output.

France and Italy are two countries where labour regulation is very tough (see Figure 1) so there is a big room for improvement. For the UK with a very flexible regime already the supposed benefits of loosening EPL are likely to be even more minor.

The studies also highlight a further problem with creating more EPL exemptions for small firms as Beecroft recommends. As the regulations kick in when firms get large they will be reluctant to grow. These size-contingent regulations can reduce the incentives of small firms to become larger, effectively subsidising companies to remain inefficiently small.

Theory: What are the pros and cons of Employment Protection Laws?

The downsides of EPL are well rehearsed in Beecroft. Larger firing costs increases labour costs and so makes employers reluctant to hire more workers which will lead to unemployment and loss of output. There could be more subtle effects as well. Because workers know that they are less likely to be fired they may slack off and managers will try less hard to identify and remove underperforming employees. Similarly, the protected “insider” workers may drive up wages, safe in the knowledge they won’t be fired. Higher average wages will mean even higher employer costs and so even fewer jobs. It might be particularly hard for individuals more on the fringes of the labour market like the young, women and long-term unemployed.

On the other side of the ledger, firing costs mean that fewer workers will be involuntarily dismissed. This reduces the level of unemployment in a downturn, at least temporarily. Obviously, EPL is often desired by workers as it gives them greater safety and security (as long as they have jobs). More subtly, this benefit may lead them to accept lower wages and this can offset the increase in labour costs as Edward Lazear, former head of George Bush’s Council for Economic Advisors has argued. In some cases this can mean that the entire additional cost of the employment regulation is born by workers in the form of lower wages, leaving unemployment completely unaffected. Of course, this effect will be limited in the presence of binding minimum wages or high unemployment benefits.

The lower turnover from reduced hiring and firing that result from EPL may mean that there is more professionalisation of careers. For example, firms and workers may invest more in training as jobs are longer and more stable. Finally, since the state pays unemployment benefits (and this social cost is ignored by firms when making) there may be an inefficiently high level of layoffs as argued by IMF Chief Economist, Olivier Blanchard. In these circumstances EPL can sometimes act as an efficient “layoff tax”.

The upshot of this is that the theoretical effects of EPL on jobs and output is an empirical issue and cannot be decided one way or the other based on theory or “common sense”.

Some of the Beecroft report is sensible such as the need to make the hiring of foreign workers easier for employers. But policy needs to be based on evidence not just on casual anecdote and assertion. Too many of the government’s policies seem to be based on ideology rather than pragmatism and if Beecroft is an example of policy formation we can see why. There is not even an attempt to engage in the wide body of evidence by academics and international bodies like the OECD, IMF, ILO and World Bank. At the LSE Growth Commission we are amassing more systematic evidence of what policies are good for long-term UK growth.

It’s time to move away from policy witchcraft and into an era where evidence is taken seriously.

Note: This article gives the views of the author(s), and not the position of the Impact of Social Sciences blog, nor of the London School of Economics

This blog post was originally published on the LSE’s British Politics and Policy blog.  


Print Friendly, PDF & Email

About the author

Blog Admin

Posted In: Evidence-based policy | Impact