2 1 7Q4 DEVELOPMENT BRIEF Number 63 The World Bank November 1995 likely to adopt productivity-raising Unions can be bad innovations relating to technologi- cal change, changing product mix, or good and reorganization of work. That is, good for efficiency and equity-if product Increased equality and reduced dis- crimination. Trade unions tend to in- markets are competitive, bargaining is in firms or crease wage disparities between plants, and laborers are free to choose their union union and nonunion workers, but or to join none at all they usually push for greater wage equality for their own members. S upporters of labor unions will be more willing to share this When union members are women see them as giving workers a information if they are confident of or belong to ethnic minorities, collective voice and enhanc- benefiting from any resulting unions also fight against discrimi- ing productivity and equality. Op- change in organization. The pres- nation. Although sometimes desir- ponents see them as monopolists, ence of an agent on the workers' be- able, the compression of wage limiting employment to raise mem- half, the union, may make them less differences due to union activity bers' wages. Do economic analyses suspicious that any information they may reduce efficiency, by sending and country experiences allow such reveal will benefit only management. wrong signals to workers about generalizations? No. Unions can If the union involves workers in ac- which skills are most needed and have positive or negative effects, tivities that improve efficiency, which industries and occupations depending on the incentives they unionism can be associated with a have the highest productivity. face and the regulatory environ- more productive organization. The reduction in wage dispersion ment in which they operate.* There are very few studies of the in unionized firms is well docu- relationship between trade unions mented for industrial countries, and Positive effects and productivity in low- and there are indications that the same middle-income countries. But a re- effect occurs in low- and middle- Improved productivity. Trade union cent analysis of Malaysian data pro- income countries as well. A study activities can be conducive to vides some support for the view in the Republic of Korea in 1988-90 higher efficiency and productivity. that unions can enhance productiv- found that unions there placed By balancing the power relationship ity and efficiency (table 1). Union- great value on wage equalization between workers and managers, ized Malaysian firms tended to and that the degree of wage dis- unions limit employer behavior that train their workers more and to use persion in the unionized sector is arbitrary, exploitative, or retalia- job rotation to enhance flexibility was 5.2% lower than in the non- tory. By establishing grievance and and efficiency. They were also more unionized sector. In Mexico union arbitration procedures, unions re- duce turnover and promote stability Table 1 Effect of unionization on productivity-enhancing initiatives in the workforce-conditions that, by firms in Malaysia when combined with an overall im- (percentage of firms taking action indicated) provement in industrial relations, en- Implementation hance workers' productivity. of job rotation Reorganization Upgrading of Extension of In many jobs workers are better Type of firm policya Of work technology product range informed than management about Plant union 30 29 40 20 how to improve productivity. They Industrial union 31 32 37 26 Nonunion 22 18 26 20 'For more details, see World Bank, World Development Report 1995. Workers in an Integrating World. New York. Oxford a. Firms with more than 1,000 workers only. Uoversity Press, 1955 Source: World Bank, World Development Report 1995, New York: Oxford University Press, 1995. action appears to have helped re- developing economies only a small Opposition to reform. Trade unions duce discrimination. A study using fraction of the working population have sometimes wielded their wage data for 1989 concluded that, belong to trade unions. For ex- political power against structural ad- in the nonunion sector, men en- ample, union coverage is less than justment. India's unions continue to joyed a 17.5% wage advantage over 4% of the labor force in Pakistan, criticize the government's recent lib- women with identical skills and ex- 5% in Kenya, and 10% in Malaysia. eralization efforts, despite the perience, but the study found no In such settings unions can play an apparent success of many of these significant wage differential be- important role in determining pay initiatives since 1991. The unions tween men and women in the differentials between workers in have organized nationwide general unionized sector. Similarly, the the small formal economy and the strikes to oppose incipient industrial study found a significant wage dis- vast informal and rural working restructuring measures aimed in advantage for indigenous people in population. If the formal economy part at increasing India's outward the nonunionized sector but no dis- is unionized, the distributional out- orientation and the reform of state crimination in firms covered by come is likely to be regressive. enterprises. India's unions are prop- trade unions. Where formal sector production ping up the part of the economy also enjoys trade protection, most in need of reform. Latin Negative effects unions are likely to share the spoils America also has a long history of with capital holders at the expense union opposition to adjustment. In Monopolistic behavior. Unions do of consumers and the masses of the 1980s organized labor opposed often act as monopolists, improving workers in the nonunionized such comprehensive reform pro- wages and working conditions for sectors. grams as the Cruzado Plan in Brazil their members at the expense of But the union wage premium is and the Austral Plan in Argentina. capital holders, consumers, and not always proof of negative distri- Unions are likely to have positive nonunion (unorganized) labor. The butional effects, even in develop- effects on efficiency and equity, and higher wages unions win for their ing countries. The actions of black their potential negative effects are members either reduce business unions in South Africa implied a likely to be minimized, when they profits or get passed on to consum- union wage premium of 10-24% in operate in an environment in which ers in the form of higher prices. Ei- 1985. But by raising the wages of product markets are competitive, ther result leads unionized firms to unionized black workers and collective bargaining occurs at the hire fewer workers, increasing the bringing them closer to those of enterprise or the plant level, and la- supply of labor to the unorganized white workers, union action may bor laws protect the right of indi- sector and depressing wages there. have helped improve the distribu- vidual workers to join the union of The size of the union wage effect- tion of income. their choosing, or none at all. the difference in compensation be- tween otherwise similar workers Table 2 Union wage premiums in selected countries that is attributed to union member- ship-has been studied in several Estimated difference between countries. Results indicate that it union and nonunion wages Country Year (percent) can reach up to 31% in developing C - countries, 10% in Europe, and South Africa, 1985 10-24 around 20% in North America Mexico 1989 10 Malaysia 1988 15-20 (table 2). Ghana 1992-93 31 Where wages for the relatively United States 1985-87 20 few workers who are unionized United Kingdom 1985-87 10 are pushed up, the actions of Germany 1985-87 5 unions can adversely affect the a. Black unions only. distribution of income. In most Source: World Bank, World Development Report 1995, New York: Oxford University Press, 1995. Developnrst Briefs are issued by the World Bank to inform the media, business, academic, and government policy communities about development policy analyses and results from the Bank's research activites. They are drawn from the work of individual Bank researchers and do not necessarily represent the views of the World Bank and its member countries-and should not therefore be attributed to the World Bank or its affiliates. Briefs are issued periodically by the Research Advisory Staff, Development Economics Vice Presidency, The World Bank, 1818 H Street, NW, Washington, DC 20433. Tel: (202)473-3984, Fax: (202)477-0955. Briefs are not copyrighted and may be reproduced with the appropriate attribution.