Was it just pure tactics?  According to the Economist (and many other commentators) Sarkozy’s successful battle to remove from the stated EU objectives “free and undistorted competition” had a simple goal: buy popular support at home for the new Treaty.  But Sarkozy is unlikely to repeat the same mistakes of Chirac and De Villepin: rather than killing himself with a risky referendum on the Treaty, he will ask the Parliament, where he holds a comfortable majority, to ratify the Treaty.  There is yet another interpretation of his editing of the Treaty: he believes that protectionism is indeed the way out of the unskilled workers problem.  France is spending some €21 billion  every year in social security deductions for low-skilled workers, plus €4 billion for "primes pour l'emploi"  targeting the same group.  Roughly another 2% of GDP goes to negative income taxes and various types of transfers to the poor, including many working poor.  In the Presidential campaign, Sarkozy promised to cut down significantly these rapidly increasing expenditures.  But the reason why these costs increased so much is not simply poor enforcement and abuse.  The issue is that wages of unskilled workers are declining, sometimes not only in relative terms.  This is happening not only in France, but also in most OECD countries: the labour share of low-skilled sectors in OECD countries declined by 10 percentage points since the beginning of the 1980s, just while the labour share of skilled sectors was increasing.

While the decline of unskilled labour in rich countries is now undisputed, its main determinants are still to be identified.  Is it trade, technological change or weaker wage compressing institutions (e..g., de-unionisation) or combinations of all these factors? Sarkozy seems to believe that it is trade and that some sort of factor price equalisation is at work: allowing free trade, wages of the unskilled workers are unavoidably bound to converge to those of Bangladeshi workers. Thus, we should protect our products rather than subsidising low-skilled workers.

There are three fundamental problems with this way of reading the unskilled labour share problem.  First, it does not consider the counterfactual. Second, it is not enforceable.  Third, it does not take into account that factor price equalisation is not at all unavoidable under free trade.

What is the counterfactual?

The new international division of labour is mainly the byproduct of falling communication and coordination costs.  These technological improvements allow not only manufacturing, but also many service-sector jobs to be offshored .  This means that there is much less which is non-tradeable than just a few years ago.  True, protectionism can prevent trade from occurring in these activities now subject to international competition. But forcing a tradeable good or service to be non-traded is not just like having a non-tradeable good, a good which is not at all internationally traded.  We simply do not know what could happen in this case.

Enforcement

This brings us to the second issue.  It seems to me that changes occurring in the division of labour are unpredictable.  And they take place within industries, more than across them.  In Italy, for instance, in the period 1998-2005 roughly 120,000 manufacturing jobs were destroyed in traditional low-skilled sectors, such as textiles, clothing and shoe-making, but another 220,000 were created over the same period in distribution and marketing within these sectors.  The issue is: would it be possible to isolate manufacturing workers from competition of Chinese workers without damaging employment opportunities of workers involved in the retail activities? And do we know what is going to happen next – i.e., which jobs, which tasks will be decentralised in the next ten years?  I believe that old-style protectionism cannot cope with the unbundling of goods.  It may out of date.

The diversification cone

What is certainly dated is my knowledge of trade theory.  But I remember that there are many conditions under which factor price equalisation does not occur.  This is the case when the endowments of trading partners are very different from each other.  Large differences in technologies and factor intensities also prevent factor price equalisation.  Data on sectoral K/L ratios suggest that the cross-country variation in factor endowments and capital/labour ratios is often much more marked than the within-country variation.  This means that countries may just not belong to the same diversification cone.  Under these conditions, even unrestricted trade does not necessarily lead to factor price equalisation. 

What can be done then for  unskilled workers?

This does not mean that the issue of unskilled labour is unimportant.  Quite the opposite, it is the major issue facing social policies in the next decade. But it cannot be addressed by protectionist trade policies.  True, the French way is very expensive and may soon become limitlessly expensive. But there is not only the French way.  Canada's Self-Sufficiency Project and the US Temporary Assistance for Needy Families are examples of how social assistance can be reorganised to become more cost-effective and avoid poverty traps.

More imaginative thinking is also required.  Wage insurance, supplementing earnings of workers displaced by globalisation in order to preserve their pre-displacement incomes, is a very fascinating idea, but also faces a number of dauntingly difficult enforcement issues.  Can pre-displacement earnings be monitored? Isn't there a risk of collusion between employers and displaced workers in raising earnings just prior to the dismissal? And how about artificial dismissals aimed at reducing labour costs of employers?

But the idea of insuring wages is there and is being taken seriously.  Experiments should and will be made.  Meanwhile, it is better not to use trade policies to protect low-skilled workers.  Such a policy agenda would be a major disappointment for the consumers and for the unskilled workers themselves.