A note from the ETUC regarding the meeting of Finance Ministers
who met on 9th March 2004

Regulation, product markets competition and the Lisbon innovation agenda.

Deregulated product markets intensify competition and are thereby supposed to lead to a more innovative economy. This idea seems to be confirmed by studies from the IMF and OECD that put a premium of 4% higher production and higher productivity on measures that would lift competitive pressures in the European product markets the level of the US. Other reports (the Sapir report for example) argue that more intense competition will push the production frontier outwards, thereby offering higher scope for accelerated productivity increases compared with the static advantage of economies of scale.

However, a recent study (European Commission 2003) questions this conventional wisdom :

Regression results in the Commission’s study conclude that R and D expenditures are not affected by regulation. Instead, market size ; education and government involvement are determining factors for investing in research.
Regressions are also run on total factor productivity (TFP), from which a similar message appears. Measures of regulation are ‘insignificant’ in explaining TFP growth. Instead, R and D expenditure as well as education are determining factors.
The fact that regulation does not show links with R and D, nor with TFP leads the Commission’s study to conclude that ‘the link between regulation and moving the production frontier outward is rather weak’.
The Commission’s study mentions an OECD study, which does conclude that deregulation would increase TFP growth. The Commission’s study immediately adds that the OECD work links (static) gains in productivity with privatisation, not with levels of regulation in general.
Finally, the Commission’s study also refers to a CEPR/IFS study, which reports that deregulation actually depresses TFP growth.

Policy simulations further support these general conclusions :

Moving to US levels of regulation would barely increase the long run rate of productivity growth by 0.15% per year.
However, an increase by 1% of the R and D share in GDP would result in extra long run productivity growth by 0.6% per year.
Likewise, an increase of 1 year in the average education level would boost productivity by 0.45% per year.
Finally, a permanent fall of 1 per cent point in hours worked increases the long run rate of productivity growth by 0.25 points, thereby illustrating the potential productivity effect of policies that limit competition on long working hours (important for the opting out in the working time directive).

Ultimately, the conclusion has to be that, in comparison with other policy measures, the possible benefits from regulatory reform seem rather limited. Moreover, the discussion so far leaves aside the question of comparability of levels of regulation. In the US, a vast body of jurisprudence compensates lower levels of state regulation. Companies in the US may face lower administrative (legal) burdens but do face substantial costs when working to avoid legal litigation procedures.

Labour market regulation, in particular employment protection legislation (EPL) and innovation

Evidence on negative impact of EPL on the general level of employment is mixed

From the outset, it should be stressed that the vast body of economic research that has been done is divided on the impact of loosening EPL on improved employment performance :

In a recent survey (CEPA/New York), 3 of out of a total of 6 studies only show a very limited impact of EPL on unemployment. Two other studies conclude on a major impact of EPL on unemployment and another one shows no effect at all.
Recent econometric work done by the IMF (WEO spring 2003) and CEPA also comes to mixed conclusions. The IMF shows that increased EPL leads to higher unemployment, but admits that the interaction with high union density reduces this effect significantly. The regression done by CEPA actually shows a negative relationship, with higher EPL leading to lower unemployment.

One reason for these mixed findings is that while EPL may deter firms to engage workers in the first phase of the economic upturn, it may also limit the impact on employment in the economic downturn, thereby making the economy more resilient against negative demand shocks. Moreover, higher EPL - costs may lead firms to weigh on the level of wages, thereby making workers finance (part of) EPL. The Commission’s Employment in Europe report (2003) provides some indications in that direction.

EPL and innovation incentives

Probably because of the ambiguous link between EPL and employment levels, the focus has shifted towards the impact of EPL on innovation incentives. It is argued that :

Strict EPL would result in workers capturing the rents from innovation and productivity improvements.
Strict EPL would limit the flows and turnovers on the labour market, thereby making it difficult for new (presumed to be innovating) firms to attract the workers they need.
EPL would have negative effects on market access of SME’s, which are again presumed to introduce higher innovative potential.

However, none of these three arguments stands up to criticism and empirical facts :

To the extent that firms can resort to internal labour markets, costs of EPL can be minimal zed. There is an important link here with coordinated wage bargaining. Coordinated wage bargaining compresses wage structures, thereby preventing other firms from ‘poaching’ on each other’s pool of skilled labour. At the same time, corporations profit from the difference between the earnings and the productivity of skilled workers. Both mechanisms induce firms to invest in training of workers and build an internal labour market.
According to draft World Bank research, annual job creation and destruction flows in manufacturing is much higher in European continental countries, compared with the US. In fact the US, has almost the lowest ‘flow’ of jobs in a set of 20 countries. Apparently, this didn’t stop the US from having the innovation and productivity boom at the end of the nineties.
The contribution of new entry of firms to labour productivity growth is not so convincing. According to the OECD Economics Outlook, the entry of new firms in manufacturing industries lowered productivity substantially in the US between ’92 and ‘97. France and the UK also saw a negative impact from the entry of new firms on manufacturing labour productivity.

Finally, EPL may on the contrary have strong positive links with innovation and productivity increases :

Workers that have their job protected will be more willing to engage and invest in training and lifelong learning. In the era of globalisation where corporate investment is demanding ever-increasing security for its investments, it is strange to see that investment from workers in their human capital base is being neglected and even attacked.
Similarly, workers that feel protected will be more willing to share their knowledge with colleague workers. They will not do so when they feel experience their colleagues as competitors that could take up their job at any moment. The knowledge society needs social capital and adequate EPL is a part of this agenda.
EPL may also contribute to regional mobility. Unemployed will only be willing to invest in resettlement if the returns on this are not completely uncertain.
Finally, if the innovation process depends on ‘incremental’ research on the basis of a routinised regime (OECD working paper), the loss of a few workers may set the innovation process significantly back. In this case, stronger EPL will increase innovative performance. The OECD estimates that 40% of business performed Rand D in manufacturing can indeed be described as a routinised technological regime.

A more promising way to innovation : Coordinated wage bargaining and the ‘no-bail out’ option

Besides providing incentives for training of workers, coordinated bargaining systems have additional important effects on innovation incentives :

In contrast with enterprise level bargaining, coordinated (sectoral) bargaining will most probably not try to exploit the full benefit of the productivity improvement in the most innovative firms. In doing so, wage premiums on innovation are avoided. Instead, incentives for firms to invest in innovation are provided.
At the same time, coordinated bargaining forces all firms to keep up wit the best performing ones. With wage rate increases in line with average productivity developments, all firms are poised to try and follow those that lead in productivity.
In contrast, widening wage differentials between firms result in innovation rents being ‘captured’ by other parties. If firms that do not invest in innovation succeed in compensating their competitive disadvantage with innovating firms by keeping wages depressed, then innovation does not pay off and firms that do innovate will see their potential competitive disadvantage quickly disappear. In general, incentives for investing in innovation will be undermined when corporations come to realise that they can depend on workers ‘bailing them out’ in case of any problems.

Forgotten lessons from the US model

How to explain the spur in productivity growth in the US in the late nineties ?

Certainly not by further deregulation of the labour market. Instead, legislation was enacted enhancing individual employee rights and minimum wages were raised twice. The Fair Labour Standards Act remained in place, requiring time and a half overtime and safeguarding the trade union role in employee involvement committees (Freeman).
The acceleration of labour productivity in the US has much to do with the process of capital deepening. Productivity gains from capital deepening have doubled in the US (but more than halved in Europe) from the first to the second half of the nineties (Commission 2003). Changes in relative prices between labour and capital account for this. The combination of overstretched labour markets in the US with perfect mobility of workers pushed wages systematically higher, forcing each individual US corporations to increase productivity standards. The policy conclusion is that macro economic policies that boost the economy and reduce unemployment significantly may trigger a self-sustaining spiral of investment, higher productivity and hence also higher potential output growth.
Ultimately, the acceleration of productivity growth in the US was limited to two or three sectors (wholesale, retail and financial services) (Commission 2003). This should raise some suspicion about the validity of policy recommendations that link productivity improvements with economy wide measures. Alternative explanations point to the geography of the US and the combination of informatics with the element of vast spaces, enabling productivity gains that are geographically not possible in Europe.

Selected references

European Commission. ‘Drivers of productivity growth’ in European Economy 2003 Review
Freeman, Richard, Ways and means of increasing employment. International labour market conference, Hannover 2000.
IMF. Unemployment and Labour Market Institutions in WEO 2003.
Baker,Schmitt,Glyn,Howell. Labour Market Institutions and Unemployment. A critical assessment of the cross country evidence, CEPA, New York, 2002
OECD. Labour Market Institutions, product market regulation and innovation. Economics Department Working Papers nr 316, 2002
OECD Productivity and innovation : The impact of product and labour market policies. OECD Economic Outlook, chapter VII.

RJ/08/03/2004