Mixed: good for some union members in the short run, often costly for non-members, firms, consumers, and overall economic efficiency.
Unions are labor cartels. They raise wages and benefits for covered workers by restricting labor supply and exercising monopoly power in bargaining, but this comes with trade-offs in employment, flexibility, productivity, and prices. The net effects depend on market structure (competitive vs. monopsonistic employers), private vs. public sector, industry, and time horizon. Economics does not support a blanket "unions are good" or "unions are bad."
Benefits for Covered Workers
•Wage and benefit premium: Union members typically earn 10-20% more in wages plus better health, pension, and leave benefits than comparable non-union workers (see reviews by Freeman/Medoff, Card, Hirsch). This is the core attraction.
•Working conditions and voice: Collective bargaining can improve safety rules, reduce arbitrary treatment, and provide grievance processes. Historically, unions pushed norms around hours and child labor that later became law.
•Insurance against shocks: seniority systems and just-cause protections buffer some workers during downturns.
These gains are real for insiders. They are largest where employers have monopsony power (few alternative jobs) or where unions capture regulated/oligopolistic rents (old auto, steel, airlines, government).
Costs and Downsides
•Employment and exclusion: The wage premium prices some workers out. Standard labor demand elasticities imply lower employment in unionized firms/sectors—especially for younger, less-skilled, or minority workers who are last hired/first fired under seniority. Non-union workers and the unemployed bear costs. "What Do Unions Do?" literature finds monopoly effects often dominate voice effects.
•Rigidities and misallocation: Work rules, featherbedding, resistance to technology/performance pay, and strike threats reduce productivity and firm adaptability. Classic examples include legacy airlines, autos (pre-bankruptcy), and docks. Seniority over merit can protect low performers.
•Dues, agency, and internal problems: Members pay dues; unions can prioritize political spending, leadership perks, or preferred groups over rank-and-file. Corruption scandals (historical Teamsters, some construction trades) and violence in organizing/strikes are recurring, not mythical.
•Public-sector difference: Teachers, government employee, and transit unions bargain against taxpayers with weak market discipline. This frequently yields unsustainable pensions, resistance to evaluation/accountability, and service disruptions. Evidence on student outcomes and fiscal stress is not flattering.
•Consumers and capital: Higher labor costs pass into prices or reduced investment/entry. Capital flees high-union jurisdictions over time (Rust Belt patterns, manufacturing shifts).
Right-to-work laws and declining private-sector density (U.S. private unionization ~6%) correlate with faster job growth in many studies; the wage hit is smaller than monopoly models predict because of competition.
Economy-Wide Effects
•Efficiency: Unions redistribute more than they create. Deadweight loss arises from reduced employment and output in covered sectors. Cross-country patterns (high-density Europe vs. U.S./East Asia) show higher structural unemployment and slower reallocation where unions + employment protection are strong, though Nordic "flexicurity" variants mitigate this.
•Inequality and demand: They compress wages within firms and can reduce overall wage inequality, but they also create insider-outsider gaps and do not reliably raise aggregate labor share long-term (automation, trade, and skills dominate). Boosting member income can support demand locally; pricing people out does the opposite.
•Growth and dynamism: Evidence leans negative for rigid union environments (see manufacturing decline, construction costs in heavy-union cities). Competitive labor markets + general skills raise living standards more reliably—real wages and consumption rose substantially as private union density fell post-1950s/60s. Innovation clusters (tech) have low unionization.
•Empirical bottom line: Meta-analyses and firm-level studies (Hirsch, Blanchflower, etc.) find wage gains + modest employment/productivity hits. Card/Krueger-type minimum wage work is sometimes invoked, but unions are broader cartel arrangements. Declines in union power coincided with rising overall employment and falling poverty by many measures; causation is multi-factor (trade, tech, education).
Context Matters
•Monopsony or boom sectors: Stronger case for unions (or just better antitrust/exit options).
•Competitive/tradable sectors: Weaker case—jobs exit.
•Modern alternatives: Occupational licensing, portable benefits, earnings subsidies (EITC-like), and tight labor markets often deliver gains with fewer cartel costs. Gig and tech organizing faces the reality that many workers value flexibility over dues and rigid schedules.
•History is not destiny: Early 20th-century industrial conditions differ from today's service/knowledge economy. Attributing the 40-hour week or weekends solely to unions ignores legislation, productivity growth, and competition for workers.
For workers who value the premium and stability and who keep their jobs, unions can be beneficial. For the broader workforce, consumers, and growth, freer entry, competition, and skill accumulation usually dominate. Policy that grants unions special legal privileges (exclusive representation, secondary boycotts, compulsory dues in non-RTW settings) amplifies the cartel downsides. For detailed evidence, start with Freeman & Medoff, Hirsch's reviews, and labor chapters in standard texts (Ehrenberg/Smith, Cahuc et al.).