Europeanisation of collective bargaining? The case of Belgian metalworking

The "Europeanisation" of collective bargaining is a key emerging issue in the field of industrial relations, and the way that the European Metalworkers' Federation (EMF) fosters coordination is often put forward as a significant step in exploring the possibilities of a more integrated bargaining policy. Here we analyse the outcomes of the latest bargaining round (covering 1999-2000) in the Belgian metalworking sector, in the light of the collective bargaining guidelines and minimum standards formulated by EMF.

In December 1998, the European Metalworkers' Federation (EMF) adopted a "European coordination rule" for national collective bargaining (DE9812283F), which is by far the most developed concept in the pursuit of a "Europeanisation" of bargaining (TN9907201S). EMF's model of fostering coordination, which has also involved the conclusion of a Charter on Working Time ,is often cited as a significant step in exploring the possibilities of a more integrated Europe-wide bargaining policy. This article investigates the extent to which the Belgian metalworking industry - known as the FABRIMETAL sector, after the industry's employers' association - complies with the core elements of the EMF coordination approach, examining (1) whether the outcomes in terms of pay increases are in keeping with the EMF wage policy and (2) the relationship of the EMF Charter on Working Time with the Belgian situation.

Wage policy

EMF European coordination rule

The core of the EMF guidelines on wage policy lies in the "European coordination rule", which was adopted during the organisation's third collective bargaining conference held in Frankfurt in December 1998. The resolution states that: "the key point of reference and criterion for trade union wage policy in all countries must be to offset the rate of inflation and to ensure that workers' income retain a balanced participation in productivity gains."

However, this demand for a return to a productivity-oriented bargaining policy - in order to avoid a possible downward spiral off wage and working conditions - does not imply that all metalworking trade unions have to follow the same bargaining pattern. The unions keep full autonomy on how they use the total "envelope" for improvement of wages, employment-related measures relevant to training or the reduction of working hours, modernisation of work organisation and special benefits such as early retirement.

Wage policy of Belgian metalworking unions

Belgium's national intersectoral agreement for the period 1999-2000 (BE9811252F) provides for a wage norm that links wage movements in Belgium to those of its three main trading partners (France, Germany and the Netherlands). As a consequence, the maximum wage increase for 1999-2000, as calculated by the bipartite Central Economic Council (Conseil Central de l'Economie/Centrale Raad voor het Bedrijfsleven), may not exceed 5.9%. All pay negotiations must take place within the framework of this wage norm.

As opposed to the bargaining tradition in neighbouring countries such as the Netherlands and Germany, the (joint) wage demands of the Belgian metalworking unions are not exclusively based on macroeconomic indicators. Instead, the actual power relations between trade unions and employers' organisations can influence quantitative and qualitative (eg training) aspects of bargaining. This specific feature of the bargaining culture may at times lead to unexpected results (such as wage increases during periods of economic slump).

Bargaining outcomes

The bargaining outcomes in 1999-2000 lead to the conclusion that recent agreements and developments in the Belgian metalworking sector largely comply with the EMF coordination rule. Moreover, with an (estimated) labour cost progression of 5.89% over the two years, the metalworking sector has clearly tended to stick to the negotiation margin of 5.9% as defined in the wage norm. These figures comprise seniority increases, inflation, nominal wage increases and the cost of qualitative elements.

The preservation of purchasing power through compensation for price increases is commonly achieved in the whole of Belgian industry by means of an indexation mechanism - ie the automatic linkage of wages to the price of goods and services (embedded in national agreements). Under this system of indexation, a pay rise of 2%, which also applies to most welfare benefits, is triggered each time the cost of living rises by 2%. In a number of sectors, there is a recent tendency to carry out the indexation at fixed points in time. This is, amongst others, the case in the Belgian metalworking industry. During the last bargaining round, it was concluded that indexation would be effected twice during the period covered by the agreement, each time on 1 July. As regards the first indexation (on 1 July 1999), the unions obtained a minimum indexation of 1.2%. Since real inflation turned out to be 0.9%, purchasing power rose by an additional 0.3%. The second indexation (on 1 July 2000) will tally with the current inflation rate on that date.

The total wage increase in metalworking amounts to 2.49% for the entire period (ie 2.19% as agreed plus an additional 0.3% due to overrunning the inflation rate in 1999). This percentage is an estimate derived from calculations based on minimum wages - as a result, it underestimates the real cost of the agreement.

Finally, several elements of a more qualitative kind need to be taken in to account. First, the social partners agreed upon the creation of a sectoral pension fund, the cost of which is borne by the employers. The whole set-up of this supplementary pension on top of the legal provisions is illustrative of the unorthodox turn of pay negotiations in the metalworking sector. The creation of the fund was a way out for both social partners during bargaining. While it seemed highly unlikely that employers would fully meet the unions' demand for a further; unconditional pay rise of 1%, the creation of a sectoral pension fund at a cost of 1% of the gross wages was more realistic. This has everything to do with the fact that the contributions to the fund are not subject to employers' social security contributions and are thus "cheaper" than a 1% pay rise (on which social security contributions are due). Aside from the 1% cost of the creation of this fund, an agreed prolongation of of notice periods, at a cost of 0.2% of paybill, also needs to be included.

Working Time

EMF Charter on Working Time

The core of the EMF policy on working time is to be found in its Charter on Working Time, which was endorsed in July 1998. The key elements of this charter are:

• the confirmation of the policy of a 35-hour working week with no loss of pay (an EMF objective that dates back to the early 1980s);

• rejecting all demands for an extension of (contractual) working time;

• the introduction of a European minimum standard of 1,750 hours maximum contractual working time per year; and

• restriction of overtime which can be compensated by money alone to 100 hours per year.

The EMF minimum standard of 1,750 hours, which was the result of two years of intensive discussion, corresponds to a 38-hour week. According to the results of a survey carried out by EMF amongst its affiliates in 1997, working time in more than half the EU Member States was still above the EMF minimum standard.

Working time in Belgian metalworking

Following the national intersectoral agreement of 1987, weekly working time in the FABRIMETAL sector was reduced from 40 to 38 hours. During the 1999-2000 bargaining round, the Belgian trade unions refrained from putting working time reduction on the agenda. From the very first contact with the employers' organisation it was clear that demands for further hours reductions were deemed unacceptable. However, whereas the collectively agreed maximum working time in metalworking is 38 hours, the actual average working time is about 36.5 to 37 hours, pursuant to company agreements. If we take this sectoral mean as the basis for calculation, it can be stated that the EMF working time standard of 1,750 hours is easily met. With an average working time that ranges from 1,679 to 1,702 hours per year, Belgian metalworking is situated well below the EMF's maximum level. Company size is an important factor in this area: large companies (eg automotive assembly lines) are characterised by lower average working hours, whereas small and medium-sized companies tend to come closer to the collectively agreed 38 hour-week.

With regard to possible demands for an extension of contractual working time, the situation in Belgian metalworking is clear-cut. All existing schemes for flexible working arrangements are enshrined in a national collective agreement for the entire metalworking industry. However, the concrete implementation of (any one of) these flexible working arrangements requires an additional collective agreement at company level. During the last sectoral bargaining round, all demands, on behalf of employers, for increased flexibility were turned down by the unions.

Overtime

As regards the issue of overtime (ie extension of production hours) very few statistics are available for the Belgian metalworking sector. Along with overtime premia, which vary from 50% to 100% of the hourly wage, Belgian labour legislation provides that overtime should, under normal conditions, be compensated by time off (the so-called "recuperation days"). The overtime is restricted to a quarterly amount of 65 hours (per person), which must be fully compensated in time off during the course of the following quarter. The restrictions on overtime were somewhat relaxed with the introduction of a Royal Decree on "minor flexibility", which allowed sectoral-level agreements to deviate from this regulation. The amount of overtime can be extended to a maximum of 130 hours over a period of six months accompanied by a full compensation in time off within the six following months. In this way, the demands of some companies that experience relatively long production peaks were met by providing leeway for certain derogations from normal regulations. The metalworking industry has systematically seized this opportunity and includes the "minor flexibility" regulations in its national agreements.

Despite the general philosophy of Belgian labour legislation, characterised by the desire to avoid "real overtime" (ie compensated by money alone) as much as possible, an exception was enacted in 1993. Provided that such an exception is included in a sectoral collective agreement, up to 65 hours overtime per year and per worker can be performed and compensated by money alone. The FABRIMETAL sector has adopted but also tightened the provisions of this exception. Deviations from the rule on "compensation in time" off for overtime are possible to the extent determined by law, but can be granted only through a collective agreement concluded at company level. Thus, theoretically, it was made possible for any company in the Belgian metalworking industry to effect a maximum of 65 hours overtime per year and per worker that are solely compensated by money. However, to date no such agreement has been concluded in any companies in the FABRIMETAL sector. In any case, the legal maximum of 65 hours is far below the EMF maximum of 100 hours.

In certain sectors of the metalworking industry, non-declared "informal" or "clandestine" overtime is widespread. Due to the highly sensitive nature of this issue, it is very difficult - if possible at all - to give a more or less accurate estimate of the magnitude of the phenomenon.

Commentary

The overall cost of the agreement covering 1999-2000 in the metalworking industry amounts to a 5.89% increase. With respect to the first provision of the EMF coordination rule, offsetting the inflation rate, the case is clear. The preservation of purchasing power in Belgian metalworking is - like in all Belgian sectors - achieved de facto through the indexation mechanism. The wording of the second provision - to ensure that workers' income retains a balanced participation in productivity gains- is unambiguous but lacks accuracy. What should be understood by balanced participation is subject to heated discussions. However, with a mean yearly productivity growth of 3% in the metalworking industry in the 1990s and a labour cost progression of 3.69% (indexation not included), our conclusion is rather positive. With respect to the EMF Charter on Working Time, the above brief overview of the situation in the Belgian metalworking industry clearly shows that the sector easily falls within the standards set out by the Charter.

This leads us to the observation that, precisely because the Belgian metalworking sector is complying with the EMF standards, no explicit attention is paid to these standards as far as concrete negotiating practice is concerned. (Jürgen Oste, Jacques Vilrokx, TESA - Vrije Universiteit Brussel)