آیا عدم تشکیل اتحادیه،منجر به برون سپاری بین المللی می شود؟

We analyze unionized firms' incentives to outsource intermediate goods production to foreign (low-cost) subcontractors. Such outsourcing leads to increased wages for the remaining in-house production. We find that stronger unions, which imply higher domestic wages, reduce incentives for international outsourcing. Though somewhat surprising, this result provides a theoretical reconciliation of the empirically observed trends of deunionization and increased international outsourcing in many countries. We further show that globalization – interpreted as either market integration or increased product market competition – will increase incentives for international outsourcing.

Many fear the consequences of globalization for ‘ordinary workers’ in the developed world. Will their jobs disappear to countries where labor costs are only a fraction of what they are in Western Europe and the US? Perhaps the rich world is left with ‘the new enterprise’ where highly skilled workers perform a firm's core activities — and where everything that can be outsourced to low-income countries, is in fact outsourced. What will then happen to the less skilled? An interesting question concerns the role of trade unions in such a situation. Are they the cause behind job losses in rich countries? Could it be that weaker unions would lead to more flexible wage setting, so that job losses could have been prevented — albeit at the price of higher wage dispersion among skilled and unskilled workers? The role of trade unionism has evolved dramatically differently in different countries over the recent years.1 The perhaps most drastic example of deunionization is the UK, where the percentage of workers covered by collective bargaining has fallen sharply over the last 15 years. The US always had weaker unions than Europe, but also there union coverage has been falling, albeit from a level that was low to begin with. In Continental Europe and Scandinavia union coverage is almost unchanged. Many of these countries are characterized by more centralized bargaining systems than in the UK and the US, and union membership rates remain at a high level. There are also a couple of countries, notably France and Portugal, where membership rates have fallen to quite low levels, but where union coverage – the percentage of the workforce that is covered by collective agreements – is still very high. If trade unions and a lack of downwards wage-flexibility were important factors behind firms' rush to outsource tasks to low-income countries, one would expect that outsourcing was more prevalent in countries with strong unions than in countries with weak unions. The facts do not seem to support this notion. Although it is not easy to find good data on country-wide outsourcing, one possible measure that may capture international outsourcing activities is the share of parts and components (input factors) in total imports. In Fig. 1, we use data on this share found in Yeats (2001), and plot them against bargaining coverage levels – which is arguably the most relevant measure of the degree of unionization in a country – from OECD (1997), augmented by data from Dell'Aringa et al. (2004) for the case of Ireland.

By way of conclusion, we would like briefly to mention one important underlying assumption that has not yet been discussed. A main result of our analysis is that deunionization can trigger international outsourcing. In its literal version, this result builds on the assumption that it is not economically possible to outsource the last task performed by unionized workers. So whenever a task is outsourced, there are always some workers that remain and experience that the demand for their labor becomes less elastic, which makes them push their wages up. What if this was not the case? The organizational structure decision of the firm would then be a two-tier one in the following sense: First, the firm would have to decide how many tasks that should be outsourced and how many should remain at home, given that the firm should keep some domestic presence. Then it would have to decide whether to go for this solution of partial outsourcing — or choose to offshore the firm in its entirety to a foreign location. Weaker unions would then presumably imply that given that the firm retains its domestic presence, the level of outsourcing goes up — but the likelihood that the whole firm is offshored goes down. Finally, let us briefly make some comments about the efficiency properties of the model. A central result in our analysis is that strong unions can protect an economy from outsourcing of jobs, and even more so in high-tech industries. This is seemingly at odds with the idea that trade unions constitute a departure from free market competition, but only seemingly so. Grout (1984) and Manning (1987) pointed out that strong unions could deter investments prior to a unionized wage-employment settlement. The same type of inefficiency is present here. The firms invest in setting up import channels for inputs, but if unions are too strong, the firm owners capture too little of the total gains from this investment. In consequence, domestic employment is inefficiently high. Of course, one could speculate that there are positive exernalities from the saved jobs, which would overturn the inefficiency result, but this discussion is outside the scope of this paper.