Chapter Study Outline

  • The hedonic framework abandons the assumption that the heterogeneous quality of workplaces is exogenously determined.
    • Instead, the framework assumes that employers consciously select nonwage attributes that affect employees’ well-being.
    • Since the early 1990s, workplace injury rates have decreased even as employment has increased.

10.1 The Hedonic Framework I: Building Blocks

  • Hedonic: Quality-adjusted: Hedonic pricing can model the size of compensating wage differentials and the choices of workers concerning undesirable job attributes.
    • The labor market transaction can be analyzed as two sales.
      • The worker sells the services of her labor in return for a wage.
        • The worker receives payment for labor.
      • The employer sells unpleasant work conditions (a bad) in return for a fringe benefit (a good).
        • The worker makes a payment for any benefits of the job by accepting a lower wage.
  • Assumption 10.1: Workers: The job is described by the package (w, z), where w is the age and z is a generic nonwage attribute.
    • (a) Each worker’s utility, u, depends on his consumption, c, measured in dollars, and a single job attribute, z, according to u = u(c, z).
    • (b) The marginal utility of consumption is MUc = Δ/Δc > 0, and the marginal utility of the job attribute, z, is MUz = Δuz.
      • The essence of the hedonic approach is that there is more to a job than its wage.
    • -MUz > 0 is the worker’s marginal disutility, assuming MUz is negative since the job attribute, z, is bad.
      • The worker’s indifference curves for c and z are positively sloped because an increase in c must be met with an increase in z in order for utility to remain constant.
      • The curves are convex because of the worker’s diminishing MRS for c and z.
      • Timid workers have steep indifference curves because they require more compensation, c, per increased unit of undesirable work conditions, z.
  • Assumption 10.2: The Firm’s Technology: Using the available technology, a firm produces output, y, and undesirable byproducts, z.
    • (a) The firm chooses the job package, (w, z), that it offers its workforce.
    • (b) The firm employs a fixed number of worker hours, L. Its profits depend on the job package, (w, z), according to Π = Π(w, z).
    • (c) Ceteris paribus, Π(w, z) is decreasing in w and increasing in z.
    • (d) Reducing the level of z is subject to the law of diminishing returns.
      • These assumptions state that wages and working conditions are endogenously determined, while the level of employment, L, is exogenous.
      • An increase in the wage, w, lowers profits, while an increase in z raises them.
    • Iso-profit curves: the infinite number of curves, Πn, along which the firm will generate the same level of profit for each combination of w and z.
      • The iso-profit curves are positively sloped because increases in w must be met with increases in z so that there is no net change in profit.
      • The curves are concave to the origin because improved working conditions have diminishing marginal returns.
      • Industries with disagreeable working conditions will have steeper iso-profit curves than industries with salubrious working conditions.
  • Assumption 10.3: Entry, Exit, and Mobility
    • All firms will operate on the zero-economic profit curve, Π* = 0, assuming free entry and exit of firms and perfect mobility of workers.

10.2 The Hedonic Framework II: Equilibrium

  • Model 10.1: The Labor Market
    • (a) The labor market is competitive and each job is described by the package (w, z).
    • (b) There is complete information about each job package.
    • (c) Workers are equally skilled and have no savings. Their preferences are described by Assumption 10.1. The firm’s technologies are described by Assumption 10.2.
    • (d) Entry, exit, and worker mobility satisfy Assumption 10.3.
      • In this model the labor market is competitive.
      • In this discussion the undesirable job attribute, z, is the risk of injury or death.
      • The no savings assumption implies that w = c.
  • In equilibrium, firms only offer one job package. The package is located at the unique point of tangency between a firm’s zero-profit locus, Π*, and the worker’s indifference curve, u*.
    • The level of workplace safety, z*, in the particular job package is Pareto efficient since any improvement in one party’s well-being must come at the expense of a reduction in others’ utility.
  • Relaxing the assumption of worker and firm homogeneity to assume there are two types of workers and two types of firms, assume one firm is dangerous (a mine) and the other is safe (a bookstore), while one type of worker is brave (Rambo) and the other is timid (Dave).
    • There is still a unique equilibrium. Two distinct job packages are offered at two points of tangency.
    • Dave works at the bookstore, and Rambo chooses to work at the mine.
    • Although they are equally skilled, Rambo earns a greater wage. This is a compensating wage differential for the extra risk he bears.
      • Positive assortative matching: the behavioral pattern by which braver workers choose more dangerous jobs and vice versa
      • Economically relevant: points that constitute rational economic choices
      • Economically irrelevant: points that no one would choose if the packages were offered
        • Workers will only select points along the single curve formed by combining the economically relevant portions of each firm’s iso-profit curve.
    • Timid Dave’s steeper indifference curve reaches a point of tangency on the initially flatter part of the bookstore’s iso-profit curve, while Rambo’s flatter indifference curve reaches a point of tangency on the latter, steeper part of the mine’s iso-profit curve.
  • Hedonic wage risk locus (hedonic locus): The smooth curve describing all of the (w, z) job packages that are offered by many types of firms to many types of workers in equilibrium
    • On this curve Rambo still earns a compensating wage differential greater than Dave’s wage because of Rambo’s greater fatality risk.
      • Fatality risk is endogenously determined.

10.3 Death and Cost-Benefit Analysis

  • Since reducing fatality rate is costly, the optimal level of workplace fatalities is not zero.
    • Eliminating fatalities would leave no money to provide other human necessities for life.
  • Assumption 10.4: The Effects of Regulation X: Suppose that by imposing regulation X, policy makers can eliminate a known work hazard and reduce the workplace fatality rate.
    • (a) A total of L = 10,000 identical workers are employed in a given industry at a wage of $40,000 per year.
    • (b) Every year D = 200 workers die from injuries on the job.
    • (c) By mandating compliance with X, the government can save an average of S = 10 lives per year.
    • (d) The annual cost to employers of complying with X is $C.
      • VSL (value of a statistical life): the dollar value ascribed to a life in order to make informed policy decisions. Assume everyone believes he faces the same chance of dying on the job and that X will reduce everyone’s accident risk by the same amount.
        • (a) Let $B denote the maximum combined amount that the group of L workers is willing to pay to implement policy X.
        • (b) Define the value of a statistical life as VSL = $B/S = total willingness to pay divided by total lives saved
          • The VSL is the value that the group of workers collectively place on a one-person reduction in the annual average of workplace fatalities.
  • Policy X should be implemented only if $B > $C, since only then will the policy offer a potential Pareto improvement and raise the well-being of everyone.
    • CV (compensating variation): the worker’s maximum willingness to pay for a decrease in the probability of a fatality, z. The CV is equal to the dollar reduction that returns the worker to his original indifference curve.
    • In practice it is very difficult to determine either the VSL or the CV due to difficulty in measuring $B. Economists infer $B by observing people’s actual choices in the marketplace.
  • The VSL equals the slope of the hedonic locus, w(z).
    • For small changes Δc approximately equals the compensating variation, CV.
    • The slope of the indifference curve can be written as Δcz = CV ∙ (L/S) = (CVL)/S = B/S = VSL. In other words, the slope of the indifference curve equals the VSL at each of its points.
    • Since every point on the hedonic locus, w(z), is tangent to an indifference curve, the slope of the hedonic locus equals the slope of the indifference curve.
      • Δwz = Δcz
      • The advantage of determining the VSL from the hedonic locus is that the VSL can now be determined using data from actual choices in the marketplace.

10.4 Policy Application: OSHA

  • In 1970, President Nixon signed the Occupational Health and Safety Act (OSHA) to prevent workplace fatality, injury, and illness. In 2007 the new agency performed nearly 40,000 inspections, found nearly 90,000 violations, and imposed over $84 million in fines.
    • Using the assumptions from the hedonic model presented in this chapter, OHSA regulations will either lower overall welfare or leave worker utility unchanged.
      • The competitive market provides the efficient level of workplace safety, and if regulations require greater safety workers will be force to pay more than they value the increased safety.
  • If workers underestimate workplace risks, it is possible that the regulations improve welfare under the following alternative assumption, which replaces Assumption 10.1b.
    • Assumption 10.5: Incomplete Information: The true accident risk, z, is known to employers and the government. Workers systematically underestimate the true risk. Their beliefs, b, are given by b = z - k, where k > 0 is a positive constant.
      • If all parties are equally uninformed, the results provided under Assumption 10.1b are unchanged.
      • If the worker perceives the workplace to be safer than it actually is, her perceived indifference curve will be steeper than her actual indifference curve.
    • In this situation, the OSHA may increase workers’ utility even as workers perceive a reduction in their utility.
      • Benjamin, Dougan, and Buschena (1997) found that workers generally obtain accurate information about on-the-job risks.
  • Though workplace safety has improved dramatically in the years since the passage of OSHA, the decline in injuries and fatalities may have been due to a larger shift in the American economy from more dangerous manufacturing industries to inherently safer service industries.
    • Firms may respond to the existence of OSHA inspections by preemptively taking action to lower their injury rates, making comparison of change in rates between inspected and uninspected firms fruitless.
  • The Peltzman effect may have led to an increase in workplace accidents.
    • Peltzman (1975) argued that increased automobile safety standards would lead drivers to reduce their own safety efforts. This offsetting effect rendered stricter safety regulations ineffective at reducing the highway death toll.
    • The offsetting effect may be present in the workplace as safety regulations reduce the overall utility of workers, who respond by working with less caution.
    • OSHA standards could lead to an increase in workplace accidents if the offsetting effect outstrips employers’ increased safety efforts.

10.5 Policy Application: Workers’ Compensation

  • Workers’ compensation (WC) insurance is a no-fault scheme offering preordained benefits to workers who are injured on the job regardless of who was to blame for the accident. In return, the employee forgoes the right to sue the employer if he is injured.
    • Replacement rate: the ratio of benefit payments to the worker’s pre-injury earnings, usually about two thirds.
    • Temporary total incapacitation (TT): the most frequently filed claim, in which it is expected that the worker can resume normal working life after a few months
    • Permanent partial claims (PP): the larger share of WC costs, in which the worker is injured permanently but expected to return to gainful employment in the future
    • Experience rated: the method for increasing firm’s premiums due to the number of prior accidents. High-risk industries typically pay more for insurance. Employers are usually required to purchase insurance to protect against insolvency.
  • There are many hypotheses for how workers and employers will respond to an increase in the generosity of benefit payments and increased insurance premiums.
    • Workers will increase the number of legitimate and fraudulent claims.
      • The cost of filing is unchanged, but benefits have increased.
    • Workers will increase the duration of time they spend out of work.
      • Higher WC payments raise the value of not working.
    • Employers will attempt to entice workers back to work early in order to reduce their experience-rated work premiums.
    • Due to moral hazard workers exercise less caution on the job.
      • This is liable to both increase the number of accidents and change the distribution of accidents.
    • Employers will respond to the experience-rating system by improving workplace safety and reducing the number of accidents.
  • Many studies have attempted to estimate these effects.
    • Claim-frequency to benefit elasticity measures the percentage change in the average number of claims resulting from a 1% increase in benefit levels.
    • Claim-duration to benefit elasticity measures the percentage change in the duration of claims resulting from a 1% increase in benefits.
  • Market responses to legislation should lower the demand for labor, shifting down the equilibrium wage by the amount of WC insurance premiums.
    • Employers will transfer at least part of the costs of the program to workers as a result of this equilibrium offset.