Labour Unions

Labour unions are a prominent societal and economic institution that are often politically relevant, thus they are often analysed politically. Whether it be from the left from a Marxian conflict theory framework, or from the right from a more individualist neoclassical economic framework. This post covers this basic economic theory and more, additionally incorporating some empirical evidence.

Some Basic Notes

• Behaviour of a standard union or a monopoly union maximises utility by demanding a higher wage, less working hours and higher leisure hours
• A firm that faces an elastic labour demand curve would cut employment, which is constraint to union behaviour
• If the firm faced an inelastic labour demand curve, the firm wouldn’t cut employment much and the union would gain more utility
• As per the Hicks-Marshall laws of derived demand, unions aim to manipulate labour demand elasticity by making it harder for substitution between union and non-union labour for firms and between goods produced by union and non-union firms for consumers
• Given workers choose to join unions, organising will be more effect in firms facing inelastic labour demand curve, with evidence suggesting union firms having more inelastic labour demand than non-union firms

Efficiency Costs of Unions

Monopoly unions cause inefficiency due to the disemployment effect

• The wage (value of marginal product of labour) differs between union and non-union jobs, employment would reduce in non-union firms and increase in union firms
• This is inefficient as the last worker hired by non-union firms would have greater productivity if they were hired in the union sector (as would happen under a competitive market wage)
• This also to a reduction in the total value of labour’s contribution to national income, as the employment distortion reduces labour output and economic value created by labour

Bargaining and Contracts

Unions bargain with firms to meet a sufficient wage-employment combination, they both have an incentive to move off the demand curve to a wage-employment combination where the wage-employment mix is better for both, unions through higher wages and/or more employment and firms through profits

• The contract curve has all the wage-employment combinations that would be Pareto optimal, with the resulting contract called an efficient contract

  • The contract curve lies to the right of the labour demand curve, for any given wage, an efficient contract leads to more employment than would be observed with monopoly unions
  • Efficient contracts also imply that unions and firms bargain over both wages and employment
  • Efficient contracts also suggest that the unionised firm is overstaffed, hiring more workers than it otherwise would at the “going” wage
  • If only two pilots are needed to fly a plane but the contract requires hiring three pilots, they may have to negotiate “make-work” to share the available tasks among the workers

• It’s important to point out the Pareto optimal wage-employment combinations are not efficient in an allocative sense because unionised firms are not hiring the number of workers they would have hired in a perfectly competitive market

  • It’s only efficient in an allocative sense with a strongly efficient contract, where the contract curve is vertical (amount of employment is fixed regardless of wage/absence of union).
  • Here is there is no efficiency cost, without a union the employment was the same and efficient, the firm just got to collect all the rents and with a union the rents can be shared without any efficiency cost

• Empirical evidence suggests wage-employment outcomes in unionised firms do not lie on the labour demand curve

  • There is disagreement over whether the contract curve is vertical though.
    • On the negative side, some evidence suggests union employment is sensitive to the union wage - contradicting the vertical contract curve
    • On the positive side, some evidence such as analysis showing an unexpected $1 increase in share of rents going to union workers reduces the value of the firm (shareholders’ wealth) by exactly $1 - as expected by the vertical contract curve
  • This means in reality, wage-employment outcomes in unionised firms are more efficient than in presence of a monopoly union, but unclear to whether they’re allocatively efficient

Union Wage Effects

There is a clear union wage gap, the wage differential or premium between similarly skilled workers in the union and non-union sectors

• Does this imply if the typical non-union worker just unionised, they could claim the wage gap? The answer is no.

  • The higher paid similar work in a union firm would be under a contract that makes it difficult to fire workers, the high cost of labour and flexibility in employment means firms screen their unionised employees more carefully - meaning the unionised firm be likely be composed of more productive workers than workers in non-union firms

• Union wages are also likely to influence the wages of both union and non-union workers through two ways:

• Threat effects: Profit-maximising employers have an incentive to keep the unions out and might be willing to share some of the excess rents in the hope the workers won’t unionise

  • Threat effects imply that unions, through their mere existence, have a positive impact on non-union wages
  • Evidence for this effect exists, such non-union police wages being higher in metropolitan areas where a powerful police union exists

• Spillover effects: As workers lose their jobs in unionised firms (like from a union-mandated wage increase), the supply of workers in the non-union sector increase and the competitive wage falls

  • Evidence of this effect also exists, such as the wage of non-union workers being lower in cities that have high unionisation rates

Public Sector Unions

Many essential public sector roles such as police officers, firefighters and teachers face inelastic labour demand curves, this is concerning for the following reasons: - If unions arose in these public sector jobs, they can behave like monopoly unions (wage-employment outcomes lie on the labour demand curve) - Marshall’s rules of derived demand imply that these unions could “extort” very high wages from taxpayers - Because these workers often become a potent political force, some politicians might be willing to grant high wages to those workers in exchange for political support

One example: Teacher Unions:

• Does teachers being unionised and getting higher wages improve student outcomes?
• Due to some states in the US banning collective bargaining by teachers and some states extending those rights, an influential study used the differences in the timings of the laws to determine teachers’ unions effects of education system outcomes (How Teachers’ Unions Affect Education Production - Hoxby 1996)
  • With regard to burden on taxpayers through spending, teachers’ unions increased per-pupil spending by 12%
    • Some of the increase goes to teachers directly through a pay rise of 5% (presumably allowing the hiring of better teachers), some of the increase spending goes to hiring more teachers, so that the pupil-teacher ratio falls
  • Surprisingly, the data suggests even with more teachers and with teachers that are paid more, the academic achievement of students does not improve
    • Instead, the dropout rate increases by 2%
    • It seems that adding more inputs to the education production function (such as more and better teachers or higher per-pupil spending) is not very effective in the rigid work environment implied by a unionised labour market