This paper proposes a conceptual framework that makes it possible to investigate the effects of central bank independence, the degree of centralization of wage bargaining and the interaction between those institutional variables on the real wage, unemployment and inflation. This is done by considering a two-stage strategic interaction between a central bank (CB) with a given degree of conservativeness and a number of unions each of which sets its own nominal wage taking the nominal wages of other unions and the reaction-function of the CB as given. In the second stage the CB picks inflation so as to minimize the combined costs of inflation and unemployment, taking union's wage rates as given. Since unions are averse to inflation they partly moderate their wage demands in order to induce the CB to inflate at a lower rate. An increase in the degree of centralization of wage bargaining (a decrease in the number of unions) triggers two opposite effects on real wages, unemployment and inflation. The decrease in the number of unions reduces the substitutability between the workers of different unions and therefore the degree of effective competition between them. This "reduced competition effect" raises real wages, unemployment and inflation. But the decrease in the number of unions also strengthens the moderating effect of inflationary fears on the real wage demands of each union. This "strategic effect" lowers real wages, unemployment and inflation. The interaction between those two effects produces a Calmfors- Driffill type relation between real wages and centralization. The paper analyzes the effects of centralization and independence on the position and the shape of this Calmfors-Driffill relation as well as on inflation and unemployment. Some of the resulting implications are tested empirically using data from nineteen developed economies. Implications for the optimal degree of conservativeness and for EMU are also discussed.