Developments in industrial action 2005–2009

[or view as size 441 kb]

Introduction
Basic industrial action indicators
- Absolute industrial action levels
- Relative industrial action levels
Affected sectors
Reasons for industrial action
Strike threats
Government intervention
Annexes

This overview examines developments in industrial action across the European Union and Norway over 2005–2009. The most ‘strike-prone’ countries during the period were Denmark, France and Belgium, while Austria, Estonia and Latvia were essentially strike-free, and the level of industrial action in the new Member States was only about a quarter of that in the EU15. Manufacturing was the sector most prone to conflict, followed by the broad public sector and transport and communications. Pay disputes were the most common cause of industrial action.

Introduction

This report from the European Industrial Relations Observatory (EIRO), based on contributions from its national centres, seeks to provide a broad and general indication of trends in industrial action over the five-year period 2005–2009 in the European Union Member States and Norway.

Industrial action is an area where international comparisons are notoriously difficult. This is largely because the way in which statistics are produced differs greatly between countries, with the definition of recordable industrial action varying considerably, and different data being collected by a variety of official and other bodies. For example, despite international efforts to harmonise definitions, there are still many differing criteria for determining whether an incident is actually recorded, such as how long an industrial action must have lasted, the number of workers involved, the nature of the industrial action, and whether the action is official or unofficial. Furthermore some countries do not appear to produce statistics for some or all of the indicators of industrial action.

These issues should be borne in mind in reading this report and the notes accompanying tables should be read carefully. The report aims to present the national data available for recent developments with no attempt at harmonisation, while pointing out the pitfalls involved in comparisons.

Basic industrial action indicators

Absolute industrial action levels

Tables 1 to 3 below provide information on three basic indicators:

• the number of working days lost through industrial action;
• the number of workers involved in industrial action; and
• the number of disputes involving industrial action.

The report covers the five-year period from 2005 to 2009, though in some countries figures for 2009 are unavailable or only partially available at the time of writing. The data from some countries are very patchy and often not published for some considerable time after the year in question.

The statistics provided are from official public sources in most cases:

• national statistical offices in the cases of Denmark, Finland, Ireland, Italy, Lithuania, the Netherlands, Norway, Poland, Slovakia (partially) and the UK;
• labour ministries in the cases of Cyprus, France, Greece, Malta, Portugal, Romania and Spain;
• other relevant public bodies in Belgium (for two of the three indicators), Germany and Sweden.

Such official statistics seem to be completely or partially absent in the other countries, and we provide data from trade unions or related bodies for all or some indicators for Austria, Bulgaria, the Czech Republic and Slovenia. The figures for Estonia, Hungary and some for Slovakia are estimates based on reports from trade unions, public authorities, the media and similar sources, while, in the absence of national data, Eurostat figures are used for Belgium (for one indicator), Latvia and Luxembourg

As mentioned above, the definition of 'industrial action' recorded in the statistics provided varies considerably. In some countries – examples include Estonia, Greece, Lithuania, Malta, Poland, Portugal, Slovenia and Spain – the only form of industrial action recorded in the statistics is strikes. Other forms, notably lock-outs by employers, are explicitly included in cases such as Belgium, Cyprus, Finland, Germany, the Netherlands, Sweden and the UK. In several central and eastern European countries where official data are absent or sparse, notably the Czech Republic and Hungary, the data used from other sources includes action other than work stoppages, such as strike warnings, protest meetings or arbitration proceedings.

In countries such as Denmark, Sweden and the UK, both official/lawful and unofficial/unlawful action are clearly recorded in the statistics. By contrast, only official or authorised action is recorded in Austria, Ireland and Malta, while in Lithuania only strikes and warning strikes that comply with relevant legislation are included. In Romania only ‘conflicts of interest’, as defined in the relevant legislation - essentially disputes related to collective bargaining over new agreements - are counted.

While most countries do not appear to impose a minimum duration for industrial action to qualify for inclusion in the statistics, only stoppages lasting at least one day are recorded in Ireland (where there must also be a total loss of 10 or more working days), and in Norway and the UK. In Cyprus, only actions lasting more than two hours are recorded. Short warning strikes (as defined by law) are explicitly included in the data for Estonia, Lithuania and Poland.

While most do not exclude action involving fewer than a certain number of workers, a few do: action involving fewer than ten workers is not recorded in Germany or the UK, for instance, unless the total number of working days lost is 100 or more.

‘Political’ strikes not directly related to employment conditions are not included in the data in several countries, including Italy and the UK.

The figures do not always cover the whole economy. For instance the official statistics for France cover only firms with more than 10 employees in the market sector excluding agriculture, while Portugal’s official statistics do not include public administration. The trade union statistics used for Slovenia in the absence of official data refer only to strikes organised by affiliates of the Associations of Free Trade Unions of Slovenia (Zveza svobodnih sindikatov Slovenije, ZSSS), which means they exclude areas where ZSSS has few members, especially in the public sector, as well as wider strike action.

Finally, collection methods for official data differ. For example, the statistics are based on reports from employers in Germany, and from employer associations in Denmark and Finland. In the UK, information is gathered from the employer or trade union involved, regular centralised returns from certain major industries and public bodies, and articles in a selection of national and regional newspapers. In Italy, by contrast, data on conflicts are provided by local police offices.

Note: no data are available for Bulgaria, the Czech Republic and Greece; *Partial figures– see notes below.

Source: EIRO

The statistics in Table 1 should be read in conjunction with the following notes.

Austria: data from Austrian Trade Union Federation (Österreichischer Gewerkschaftsbund, ÖGB) Chamber of Labour (Arbeiterkammer, AK)

Belgium: figures are from the National Office for Social Security (Office National de la Sécurité sociale/Rijksdienst voor Sociale Zekerheid, ONSS/RSZ) and refer to days not worked because of strikes or lock-outs, as declared by employers to ONSS/RSZ; 2009 figure is for first half of year only.

Cyprus: figures are from the Department of Labour Relations at the Ministry of Labour and Social Insurance and cover all strikes and lock-outs lasting more than two hours .

Denmark: figures are from Statistics Denmark (Danmarks Statistik) and cover all official and unofficial strikes.

Estonia: figures are estimates by the EIRO national centre, based on data from the Confederation of Estonian Trade Unions (Eesti Ametiühingute Keskliit, EAKL), Estonian Transport and Road Workers’ Trade Union (Eesti Transpordi- ja Teetöötajate Ametiühing, ETTA), Estonian Seamen’s Independent Union (Eesti Meremeeste Sõltumatu Ametiühing, EMSA), public conciliator's office and media reports; figures relate to all strikes, including short warning strikes which may last up to one hour, and there is no information on lock-outs by employers.

Finland: figures, from Statistics Finland (Tilastokeskus), relate to industrial action organised by employees or employers.

France: figures are from the Ministry of Labour’s Directorate for Research, Studies and Statistics (Direction de l’animation de la recherche, des études et des statistiques, DARES), and are based on a representative survey of firms in the market sector (excluding agriculture) with more than 10 employees, and include all industrial action.

Germany: figures from the Federal Employment Agency (Bundesagentur für Arbeit, BA), based on reports by employers; strikes and lock-outs, are recorded in the official statistics only if they involve at least 10 employees from one workplace and last at least one day or – regardless of the number of workers involved or length of time – if they cause the loss of more than 100 working days per establishment; industrial relations researchers in Germany believe that there are major shortcomings in the official records of industrial action, with a substantial amount of strike activity not reported by employers to BA; the Institute for Economics and Social Science (Wirtschafts- und Sozialwissenschaftliches Institut, WSI), basing its conclusions primarily on strike data collected from trade unions (or, where trade unions could not provide data, referring to press reports), estimates the number of working days lost through industrial action at 175,000 in 2005, 1,607,000 in 2006, 725,000 in 2007 and 542,000 in 2008.

Hungary: figures are estimates based on monitoring of media reports by the EIRO national centre and include all reported actions related to terms and conditions of employment and other social issues; types of actions include strikes, warning strikes, other company-level actions (predominantly protest meetings), rallies, petitions, open letters, protests and mass resignations.

Ireland: figures are from the Central Statistical Office; disputes are included if they involve an official work stoppage lasting for at least one day and the total time lost is 10 or more working days.

Italy: figures are from National Institute of Statistics (Istituto Nazionale di Statistica, Istat); 2008 and 2009 figures are provisional; data refer only to work stoppages arising from employer-employee relationships and exclude those with different causes, such as political strikes.

Latvia: figures are from Eurostat and record strikes and lock-outs due to labour disputes.

Lithuania: figures are from Lithuanian Statistics (Lietuvos statistikos departamentas, STD) and cover only lawful strikes and warning strikes called in line with the Labour Code; 2009 figure for first three quarters only.

Luxembourg: figures are from Eurostat and refer to strikes and lock-outs due to labour disputes.

Malta: figures are from the government's Department of Industrial and Employment Relations (DIER) and exclude unofficial strikes and action short of a strike (go-slows, work-to-rules etc).

Netherlands: figures are from the Central Statistical Office (Centraal Bureau voor de Statistiek, CBS) and refer to all strikes, work stoppages and exclusions of workers from work by employers.

Norway: figures are from Statistics Norway (Statistisk sentralbyrå, SSB) and cover work stoppages due to labour disputes that last for at least one day.

Poland: figures are from the Central Statistical Office (Główny Urząd Statystyczny, GUS) and refer to strikes and warning strikes (which may last up to two hours).

Portugal: figures are from the Ministry of Labour and Social Solidarity and relate to strikes but exclude public administration.

Romania: figures from the Ministry of Labour, Family and Social Protection (Ministerul Muncii, Familiei şi Protecţiei Sociale, MMFPS) record temporary interruptions of work relating to ‘conflicts of interest’ between employers and employees.

Slovakia: figures are estimates by EIRO based on data from the Slovak Statistical Office (Štatistický úrad Slovenskej republiky, ŠÚ SR) and trade union sources about the numbers of disputes and workers involved, and the duration of any action.

Slovenia: figures from the Association of Free Trade Unions of Slovenia (Zveza svobodnih sindikatov Slovenije, ZSSS) cover only strikes organised by its affiliates in individual companies; large parts of the economy such as the public sector are excluded, as are strikes with a wider scope and those organised by other unions.

Spain: figures are from the Labour Statistics Bulletin (Boletín de Estadísticas Laborales) produced by the Ministry of Labour and Immigration (Ministerio de Trabajo e Inmigración, MTIN) ) and record working days lost due to strikes.

Sweden: figures are from the National Mediation Office (Medlingsinstitutet, MI) and cover all strikes, official and unofficial, and include short-term political demonstrations that affect work and lock-outs where work is suspended, irrespective of size of workforce.

UK: figures are from the Office for National Statistics (ONS) and cover all strikes, including ‘unlawful’ strikes, and lock-outs; however, stoppages involving fewer than 10 workers or lasting less than one day are not recorded unless the total number of working days lost in the dispute is 100 or more; also excluded are stoppages over issues not directly linked to employment terms and conditions; figures refer to disputes in progress in the year in question, rather than those beginning in that year; 2009 figures are for the first 11 months of that year.

Note: no data are available for Bulgaria and the Czech Republic; * Partial figures – see notes to Table 1.

Source: EIRO.

Sources and notes are as for Table 1, with the following differences or additional comments.

Belgium: no data available from national sources; figures are from Eurostat and refer to strikes and lock-outs due to labour disputes.

Germany: WSI estimates the number of workers involved in industrial action at 67,000 in 2005, 1,030,000 in 2006, 550,000 in 2007 and 1,550,000 in 2008.

Greece: these figures from the Ministry of Labour relate to strikes; the figures given are the aggregate of statistics for the private sector, public administration and ‘public enterprises of common interest’ and banks; the Ministry’s regional data collection is incomplete and the level of strike action is therefore underestimated.

Romania: 2009 figure refers to the first nine months of the year.

Slovakia: 2005 and 2006 figures come from ŠÚ SR; 2007–2009 figures are estimates based on information from trade union sources

Note: no data available for Belgium, Estonia, Germany, Greece and Luxembourg; *Partial figures – see notes to Table 1.

Source: EIRO.

Sources and notes are as for Table 1 ( with the following differences or additional comments:

Bulgaria: data from the strike committees of the Confederation of Independent Trade Unions in Bulgaria (CITUB) and Confederation of Labour Podkrepa (CL Podkrepa).

Cyprus: figures given are for strikes only.

Czech Republic: figures from Czech-Moravian Confederation of Trade Unions (Českomoravská konfederace odborových svazů, ČMKOS) refer to its affiliates only and covers not only strikes, but also strike alerts and mediation/arbitration proceedings.

Hungary: figures (estimates – see note to Table 1) include forms of action other than strikes; in 2009, there were 12 strikes or warning strikes.

Malta: figures given are for strikes only.

Romania: 2009 figure refers to first nine months of the year.

Slovakia: 2005 and 2006 figures from ŠÚ SR; 2007-2009 figures are estimates based on information from trade union sources

Relative industrial action levels

The figures in Tables 1 to 3 provide some indications of trends in individual countries, but they are of little use for purposes of international comparison. The very different sizes of the countries mean that the absolute figures give little indication of the extent to which countries are strike-prone when compared with others. The only measure that enables this comparison is the number of working days lost through industrial action per 1,000 employees. Table 4 provides data on this indicator (based on estimates in some cases) for 24 EU countries and Norway, excluding Bulgaria, the Czech Republic and Greece.

The data presented for a number of countries clearly underestimate the level of industrial action (see the notes to Table 1). For example, the Portuguese statistics exclude public administration workers; the Italian figures exclude political strikes which can add greatly to the amount of industrial action; and the Slovenian figures refer only to strikes organised in individual companies by the affiliates of one, largely private sector, trade union confederation – albeit the largest one, ZSSS. Furthermore, many industrial relations researchers in Germany argue that the country’s official statistics seriously underestimate the actual amount of industrial action because it is under-reported by employers. The WSI research institute puts the number of working days lost per 1,000 workers three times higher than the official figures between 2005–2008).

Averaged out across all the countries in these tables, the annual number of working days lost per 1,000 employees fell from 35.4 in 2005 to 22.3 in 2006 and 21.1 in 2007, before rising sharply to 50.5 in 2008 and falling back to 24.3 in 2009. However, the averages for 2008 and, particularly, 2009 are calculated using figures from a smaller number of companies. Within individual countries, consistent trends are hard to identify, with figures often rising and falling from year to year and following no discernable pattern. Exceptions include a sustained upward tendency between 2005 and 2009 in Spain; an upward trend until 2008, followed by a fall in 2009, in Denmark and Sweden; and a downward trend until 2008, followed by a rise in 2009, in Ireland. The annual variations can be very substantial, with countries experiencing major peaks in action before and after years of relatively low levels of industrial action, such as in Belgium and Finland in 2005, Norway in 2006 and Ireland in 2009.

Source: EIRO.

The figures provided in Table 4 for Belgium, Cyprus, Italy, Lithuania, the Netherlands, Poland, Slovakia and Slovenia are estimates calculated by dividing total working days lost by the number of employees as reported by Eurostat for each year. Otherwise, sources and notes are as for table 1, with the following additional comments.

Denmark: calculation by EIRO national centre based on national data on number of employees, except 2009 when Eurostat data used

Estonia: calculation by EIRO national centre based on national data on number of employees

Germany: WSI estimates the number of working days lost per 1,000 employees at 5.1 in 2005, 46.3 in 2006, 20.5 in 2007 and 15.1 in 2008.

Ireland: calculation by EIRO national centre based on national data on number of employees

Malta: calculation by EIRO national centre based on national data on number of employees

Norway: 2009 figure is based on Eurostat data on number of employees.

Portugal: 2007 figure is based on Eurostat data on number of employees.

Spain: 2009 figure is based on Eurostat data on number of employees.

Sweden: 2009 figure is calculated by EIRO national centre based on national data on number of employees.

UK: 2009 figure is an estimate by EIRO, using Eurostat data on number of employees.

Figure 1 gives the annual average figure for working days lost through industrial action per 1,000 employees over the five-year period 2005–2009. It should be noted that the average is calculated over four or three years in a number of cases (see notes below Figure 1) for 25 countries where the relevant information is available. The various caveats mentioned above about incomplete data for some countries should be borne in mind.

On average, between 2005–2009, the highest levels of industrial action were found in Denmark (159.4 days lost per 1,000 workers – however, see further information on these figures below) and France (132 days lost). The lowest levels were in Austria (no days lost) and Estonia (0.1 days lost). The average for all 25 countries was 30.6 days lost. The average in the new Member States (11 days lost) was only about a quarter of that in the EU 15 plus Norway (43.6 days).

A rough comparison with the previous decade can be made on the basis of international comparative statistics, based on data from Eurostat, the International Labour Organisation and the Organisation for Economic Cooperation and Development (OECD), published by the UK Office for National Statistics (ONS) (see Economic and Labour Market Review, April 2008). According to this source, the average number of days lost over the period 1997–2006 was 39 for the whole of the EU, based on data for 22 countries, and 42 for the EU15 minus Greece. The EIRO average for all countries of 30.6 days over the five years between 2005 and 2009 suggests an overall fall since the previous decade, although the figure for the EU15 has changed little.

For the five-year period 2005–2009, the 25 European countries can be divided into three groups:

• countries where industrial action was at low levels, with an average of fewer than 20 working days lost per year for every 1,000 employees. This group includes all the central and eastern European countries for which data are available – Estonia, Latvia, Lithuania, Hungary, Poland, Romania, Slovakia and Slovenia, along with Austria, Germany, Luxembourg, the Netherlands, Portugal and Sweden,
• countries where industrial action was at moderate levels, with an average of 20–60 working days lost per year for every 1,000 workers. This group is made up of Cyprus, Malta, Italy, Ireland, Norway and the UK;
• countries where industrial action was at relatively high levels, with an average of more than 60 working days lost per year for every 1,000 workers. This group comprises Belgium, Denmark, Finland, France and Spain.

However, the average figures for such a short period cannot give a proper picture of national trends since there are individual years when highly unusual levels of industrial action affect the figures disproportionately. For example, Denmark had an annual average of 23.7 working days lost if only 2005, 2006, 2007 and 2009 are considered. In 2008, however, a two-month strike over the renewal of pay agreements by nurses, child and youth educators and home carers employed by municipal and regional authorities accounted for 98% of the working days lost during that year, and the 2008 figures were the highest for a decade (DK0907019I). As a result, in 2008 Denmark recorded the highest annual figure (701.9) for working days lost per 1,000 workers of any country during the period 2005–2009, inflating its annual average for the five-year period to more than 159 days lost per 1,000 workers. In the same way, Ireland’s average between 2005 and 2009 was pushed up to 38 days lost per thousand workers by a one-day national public sector strike in 2009 in protest at pay and job cuts (IE0912019I), accounting for more than 70% of the working days lost that year. Excluding the 2009 figures, the average was just 5.6 working days. Similar situations can be seen in Belgium and Finland in their 2005 figures.

Affected sectors

Industrial action is rarely spread evenly throughout the economy, often being concentrated in particular sectors; levels may also rise or fall over time Annex 1 at the end of this report shows the three sectors most affected by industrial action each year between 2005 and 2009, although data are not available for all years for every sector in each of the 26 countries for which information is available. There is no information available for Luxembourg or Latvia. Annex 1 also shows the percentage of days lost through industrial action by each sector where this information is available. In some cases, as indicated, percentages of the number of disputes, or of the number of workers involved in industrial action, are used instead.

Differing definitions of sectors in different countries preclude a detailed analysis of the data presented in Annex 1. However, it can be noted that the sectors most affected by industrial action between 2005 and 2009 were manufacturing (with metalworking particularly prominent in this category), followed closely by the broad public sector (notably education, healthcare/social work and public administration) and transport and communications. Private-sector services were some way behind, but were notably conflict-prone in some or all years in countries such as Cyprus (in hotels and commerce), Finland, Germany (retail), Italy (hotels and commerce) and Norway (finance). Construction was also sometimes a key area of industrial action in countries such as Cyprus, Ireland, Norway, Romania, Spain and Sweden. Utilities featured strongly in some years in Bulgaria, Ireland, Romania and Sweden. In some generally quiet sectors, levels of industrial action rose sharply in certain years – for instance, oil in Norway, mining in Poland and agriculture in Spain.

The nature of the data available for 2005–2009 does not always make it possible to differentiate between private and public sectors. France’s transport sector, for instance, reported to be the one most affected by industrial action, includes both private and publicly run organisations. Where data are available, it appears that the distribution of industrial action between the two varies considerably between countries. For example, Lithuania, Malta, Poland, Slovakia and the UK are countries where a relatively high proportion of the most strike-prone industries are in the public sector, while the private sector seems to have higher levels of industrial action in countries such as Belgium, Cyprus, the Czech Republic, Finland, Germany, the Netherlands and Romania.

Annex 1 shows the extent to which a particular sector may dominate the industrial action statistics of a particular year. For example, a single sector accounted for half or more of all working days lost in the following countries:

• Belgium in 2005–2007 (manufacturing);
• Denmark in 2006 and 2008 (public sector);
• Ireland in 2005 (transport, storage and communication) and 2006 (construction);
• Italy in 2005 (metalworking);
• the Netherlands in 2006 and 2008 (transport and communication);
• Norway in 2005 (construction), 2007 (health and social work) and 2008 (education);
• Romania in 2005 (transport, storage and communication);
• Sweden in 2007 (transport and storage), 2008 (health and social work) and 2009 (transport and storage);
• the UK in 2006 (public administration), 2007 (transport and communication) and 2008 (public administration).

Reasons for industrial action

Industrial action statistics and other sources from 22 countries allow the main causes of industrial action to be identified. This is not possible for Belgium, Denmark, Ireland, Latvia, Luxembourg and Slovenia. Annex 2 at the end of this report shows the three main causes of industrial action between 2005 and 2009, although data are not available for all years or all causes. The percentage of all days lost through industrial action is given for each cause where this information is available. Otherwise, where indicated, the percentages refer to the numbers of disputes or numbers of workers involved in industrial action.

Widely differing ways of categorising the reasons for industrial action make comparisons difficult, but it can be said that pay is clearly the most important cause. Only in Cyprus, the Netherlands and Spain is pay not among the leading causes of industrial disputes. Pay is consistently the single most important issue in industrial action in France, Lithuania, Portugal, Romania and the UK, and in most years in Estonia, Hungary and Poland. While demands for wage increases are the most frequent cause of pay disputes, wage arrears are also important triggers of disputes in countries such as Romania.

Pay can also be assumed to play a part in many collective bargaining- or collective agreement- related disputes which are the prominent causes of action in countries such as Cyprus, the Czech Republic, Finland, Italy, Malta, the Netherlands, Spain and Sweden.

After pay and collective bargaining, other frequent causes of industrial action are:

• the broad theme of working conditions (prominent to varying degrees in Bulgaria, Finland, France, the Netherlands, Poland, Portugal, Romania and the UK) and working hours (which features on occasion in the Czech Republic, the Netherlands, Romania and the UK);
• job losses, dismissals and redundancies (often prominent in Finland, France, Italy and the UK and in some years in the Czech Republic and Sweden), and restructuring disputes (as in countries such as Hungary, Poland and Romania) that probably also often involve concerns about job losses;
• broadly ‘political’ issues, with general or specific government policies (such as social security issues, labour law reforms, public sector pay and jobs, privatisation and sector restructuring) being an important cause of action in Bulgaria, Greece, Hungary, Malta, Poland and Spain and occasionally in Estonia and Sweden. Italy too sees substantial political action in some years, but the data used here exclude such action.

There are also a number of national specificities, such as the prominence in some years of trade union/labour rights disputes in Poland, or issues of health and safety in Malta.

Strike threats

While the focus of this report is on industrial action that results in the loss of working days, strikes may also be called or threatened by trade unions but called off before they can happen (this is a separate practice from holding short ‘warning strikes’, as occurs in countries such as Estonia, Germany, Hungary, Lithuania and Poland). This is often a tactic to increase pressure in negotiations, and in some countries giving notice of strikes is part of the ‘normal’ bargaining process.

It appears that systematic data on strikes called but not held are very rarely collected, making it difficult to assess the scale of the practice in individual countries or to compare countries. The only cases where any relevant statistics are available are Italy and the UK.

In Italy, under legislation restricting the right to strike in essential public services, strike calls in these services are monitored by an independent Guarantee Authority (Commissione di Garanzia). In 2008, the most recent year for which information is available, trade unions at national and local level issued 2,195 notifications or threats of strike action, of which 856 (39%) were called off before they were due to start.

An estimate of the extent of the phenomenon in the UK can be obtained by comparing the number of disputes involving industrial action with the number of successful ballots for strike action as reported by Electoral Reform Services. A ballot of the union members involved is obligatory before taking industrial action. There were 663 successful ballots and 116 disputes in 2005 (in other words, 17% of the planned strikes took place); 1,094 successful ballots and 158 disputes in 2006 (14%); 697 successful ballots and 142 disputes in 2007 (20%); and 658 successful ballots and 144 disputes in 2008 (22%).

While no statistical data are available, unfulfilled strike threats also seem relatively common and part of standard industrial relations tactics in countries such as Ireland, Norway and Sweden. In Sweden in 2009, for example, there were two cases of national strikes called but cancelled, while the affiliates of Swedish Trade Union Confederation (Landsorganisationen, LO) gave notice of 16 regional strikes but called them all off. In Norway, notice of a strike will almost always be given as part of the normal bargaining process and this serves the purpose of giving the parties a deadline for concluding the negotiations. This rarely leads to a strike. When industrial action is announced, indicating that a new collective agreement cannot be reached, the state mediator will normally impose compulsory mediation which will usually lead to a new agreement.

Unfulfilled strike threats were also reported in 2009 from countries such as Austria, Belgium, Bulgaria, the Czech Republic, Estonia, Finland, France, Germany, Hungary, Latvia, Malta, Portugal and Romania. It is not known how frequently this happens in most of these countries, but there were only a handful of cases reported in Austria, Belgium, Bulgaria, Estonia, Hungary, Latvia, Malta, Portugal and Romania in 2009. No notable cases at all are reported in 2009 in Cyprus, Denmark, Lithuania, Luxembourg, and Slovakia, and it is not clear if the practice ever occurs in these cases. Finally, no data at all are available on the issue from Greece, the Netherlands, Slovenia and Spain.

Table 5 gives a number of examples of unfulfilled strike threats in 2009. In most cases, the threatened action was in support of collective bargaining demands, especially for pay increases, or in protest at planned job losses or other employers’ actions. A few cases involved political demands, with unions seeking to influence government policy on the public sector, or more generally. In almost all cases, the action was called off because the unions achieved their demands or a satisfactory compromise through further talks, in some cases involving the government, or reached agreement on a framework for renewed negotiations.

Source: EIRO.

Government intervention

Almost all countries examined here have mechanisms for resolving industrial disputes through arbitration, mediation and/or conciliation, provided by the state or the social partners themselves. In this section, another aspect of the prevention of industrial action will be examined – the extent to which the government or other public authorities (other than specialist dispute-resolution bodies) intervene in order to prevent planned strikes, or to stop strikes that are already under way.

The evidence from 2009 indicates that it is rare for governments and public authorities, other than dispute-resolution bodies, to intervene explicitly in disputes in organisations and sectors where they are not the employer. No clear-cut cases are reported. A somewhat indirect intervention occurred in the French hotels and restaurant sector in November 2009. Here, collective bargaining had broken down over unions’ wage demands, especially in relation to low-paid employees, and unions threatened a strike. The deadlock was broken by a government announcement that it would maintain indefinitely a greatly reduced rate of VAT for the sector (introduced earlier in the year), which changed the employers’ position and enabled an agreement to be signed.

Explicit government intervention in disputes in areas where it is not the employer is rare (though behind-the-scenes intervention or indirect pressure may be hard to gauge). In the public sector, though, the boundaries can be blurred. In the public sector the government, whether at national or at lower levels, is essentially the employer party in industrial disputes. However, employer responsibilities are often devolved to the managements of public administrations or state-owned organisations, and the extent to which governments take an active role in disputes in such bodies varies considerably. Cases reported in 2009 include the following.

• In Bulgaria, the Minister of Transport, Communications and Information Technology assisted in reaching the settlement that prevented a strike at the National Railway Infrastructure Company (see Table 5 and BG0911019I).
• In Poland, the government made concessions that prevented further strike action opposing the privatisation of the majority state-owned Copper Mining and Smelting Industrial Complex (Koncern Górniczo Hutniczo Miedziowy, KGHM) (the government agreed that the state would retain a majority stake in the concern).
• The Portuguese government intervened and signed an agreement in August on employment guarantees with management and unions at the state-owned TAP airline, halting a strike begun by ground staff.
• In Romania, the Ministry of Transport and Infrastructure (Ministerul Transporturilor şi Infrastructurii, MTI) successfully brought a court case in November that ruled a planned strike at METROREX, the Bucharest underground railway system, illegal (RO1002039I);
• The Spanish national government intervened to prevent a strike at the publicly-owned Renfe rail company in Catalonia over the devolution to regional level of responsibility for rail services, reaching agreement with the unions and the Catalan regional government to set up a body to negotiate about how to ensure the social sustainability of rail services following the transfer.

Another form of government intervention in the public sector is to make budgetary decisions that enable settlements to be reached. For example, in Estonia a dispute over plans to make 110 workers redundant in the public rescue service because of budget cuts (see Table 5), was resolved (temporarily at least) when the Ministry of the Interior found extra resources to reduce the number of redundancies to 40. Workers at Hungarian Public Television (Magyar Televízió, MTV) set up a strike committee to oppose substantial budget and job cuts, but the prospect of industrial action was averted when the Ministry of Finance earmarked extra funding for MTV from budgetary reserves.

In a number of countries, dispute-resolution bodies play a relatively active role in intervening in cases of actual or potential industrial action and this, in some cases, might arguably be seen as a form of state intervention. For example, Ireland’s National Implementation Body (NIB) plays an important role in intervening in order to prevent or stop strikes. In 2009 it intervened in a dispute at a Lufthansa Airmotive aviation plant, brokering proposals to secure the future of the site that were approved by the workforce in April (IE0905019I). The NIB also intervened to recommend a mediation process that led to an agreement in November on redundancy terms at Coca-Cola HBC, which ended two months of strike action.

Similarly, in Italy, the Guarantee Authority that oversees industrial action in essential public services may intervene in disputes. It also has the power to impose guaranteed minimum services during strikes on sectors that have not reached an adequate agreement on this issue. It can refer cases of failure to observe the rules on calling strikes or on minimum services to the public authorities, which may order workers to return to work where strikes pose a threat to public safety or constitutional rights. In 2009, the Authority intervened in 549 cases. In 361 cases (66%) the intervention was successful, leading to the industrial action being called off in 230 cases and postponed in 131.

Malta’s statutory Director of Industrial and Employment Relations (IER) regularly and successfully intervenes through mediation to pre-empt or halt strikes. For example, in 2009 the Director intervened in disputes at organisations such as Air Malta, Go (telecommunications) and health centres, leading to the prevention or cessation of industrial action.

Norway has a system of compulsory arbitration by the National Wages Board that may be used by the government to end industrial action that threatens life or health, or has other seriously damaging effects on society. In 2009, there was one case of compulsory arbitration, preventing a strike at the Norwegian Air Ambulance.

Overall, however, the picture is of governments not intervening directly or overtly in industrial disputes. Many apparently take the line of the UK government which is generally concerned to limit its involvement in industrial action, stating that this is a matter for employers and employees to resolve.

Mark Carley, IRRU/SPIRE Associates

Annexes

Notes: see Table 1 above; data for some countries based on information from trade unions etc or EIRO national centre estimates

Source: EIRO.

Notes: see Table 1; data for some countries based on information from trade unions etc or EIRO national centre estimates.

Source: EIRO

Annex 3 – European Company Survey 2009: Findings on industrial action

The findings of the 2009 edition of the European Company Survey (ECS 2009) give an interesting snapshot of industrial action at company level across Europe. The survey shows that about 20% of employee representatives in the EU27 report that some form of industrial action took place at their establishment during 2008, including strikes, short warning strikes, refusal to do overtime and other forms of action.

In the EU27, 9% of the representatives say that the industrial action took the form of a short work stoppage of less than one day, and 7% say that there was a strike of one day or more at their establishment in the course of 2008. Hence, while industrial action is not uncommon in establishments with employee representation, major conflicts such as strikes are rather exceptional.

Confining an analysis of industrial disputes to strike action, a higher incidence can be detected in some southern countries. Greece is well ahead of the other countries, with 45% of the interviewed employee representatives reporting some form of strike at their establishment during 2008. In 26% of establishments with employee representatives in Portugal; the equivalent figures for France were 25% and for Italy, 16%. Above-average levels of strike activity in the same year were also reported for Germany (14%), the Czech Republic (12%) and Belgium (12%).

When asked for the reasons for various forms of industrial action, employee representatives report that pay is by far the most significant cause, particularly for strike action: 74% of employee representatives give pay-related concerns as the reason for strikes. Other important causes include changes in the organisation of work (37%) and restructuring, mergers or relocations (28%).

The ECS 2009 looked into flexibility practices at the establishment level (such as flexibility with regard to working time, pay and employment contracts) and workplace social dialogue in 30 European countries. The ECS interviewed HR managers and, where possible, employee representatives (who may be works council members or trade union representatives, depending on the national system of employee representation).

Data on industrial action stems from the employee representative interview. The incidence and coverage of employee representation differs greatly by country. In some countries, such as Sweden, employee representation is almost universal across the whole economy. In other countries, companies with employee representatives are in the minority as, for instance, in Greece. This ‘bias’ has to be taken into account when interpreting the data. The figures for industrial action pertain to ‘establishments with an employee representation’ and not to ‘all’ establishments.

More information on the survey is available at http://www.eurofound.europa.eu/surveys/companysurvey/index.htm

Mark Carley, IRRU/SPIRE Associates