The Hiring Process

Chapter Study Outline

• Previous chapters have assumed that firms simply select a level of employment and take the wage as given in perfect competition.
  • In reality, both firm recruitment activities and worker search efforts are costly and time-consuming, while the labor market is very dynamic.
  • Furthermore, firms and workers are heterogeneous, resulting in imperfect information for both parties.

13.1 The Economics of Hiring

• A number of economic mechanisms have evolved in the hiring process to bridge the information gap between workers and firms.
  • Workers can engage in job-market signaling by their educational attainment and by their visible accomplishments on the job.
  • Firms can use screening mechanisms such as offering different types of compensation packages in order to induce workers to reveal information about themselves through self-selection.
    • After signaling and screening, job seekers and employers are matched in the labor market.
    • Firms reject some candidates and rank those who remain.
  • Employers can obtain more information by using formal tests during the interview.
    • The firm then ranks the remaining candidates and makes job offers or rejections and restarts the process.
• Economists focus on two basic aspects of the hiring process.
  • Heterogeneity: Workers possess heterogeneous abilities, denoted ϵ A.
    • ϵ A is read as “Ability belongs to the set of possible abilities A.”
  • Information: Firms might not know workers’ abilities when they make their hiring decisions.
    • To effectively employ information economics it must be clear what agents know and when they know it.
    • Assumption 13.1: Information about Workers’ Abilities: PAS, the three common hypotheses about the information possessed by workers and firms.
      • (P) Each worker’s ability is public information and is observed by workers and firms.
      • (A) There is asymmetric information; workers know their own abilities but firms do not.
      • (S) Both workers and firms are symmetrically uninformed about workers’ true abilities.
    • P refers to a worker’s track record in the workplace, but it is less plausible for new entrants due to their limited experience.
    • A includes to a worker’s innate, unobservable abilities, which he knows better than prospective employers.
    • S is more relevant at the beginning of a worker’s career, since experience will reveal more about the worker’s true ability.

13.2 Public Information (The Roy Model)

• If there are no informational imperfections, only Assumption P is relevant.
  • If worker ability differs along a single dimension, earnings will simply be proportional to ability-unadjusted productivities.
• If workers’ abilities differ along several different dimensions, the Roy Model can help determine how much they earn and where they work.
  • Model 13.1: The Roy Model
    • (a) Roy (1951) assumes two occupations: hunting (H) for rabbits at a market price $r and fishing (F) for trout at a market price $t.
    • (b) There are a large number of workers who seek to maximize their earnings while working in only one occupation at a time.
    • (c) Each worker’s ability is described by (ρ, τ), where ρ and τ are, respectively, the number of rabbits or trout he can catch per week.
    • (d) The ability pairs (ρ, τ) are uniformly distributed over 0 ≤τ ≤ T₀ and 0 ≤ ρ ≤R₀.
    • (e) The inequality $r ∙ R₀ < $t ∙ T₀ holds.
  • A worker will be a hunter if r ∙ρ≥ t ∙τ and will fish if r ∙ρ < t ∙τ. If τ = (r ∙ρ)/t, she will be indifferent between the two occupations.
    • T* = (r/t)R₀ represents the highest potential fishing ability of hunters. Although these hunters are excellent hunters, they are also very good fishermen.
  • The Roy model demonstrates that the population will self-select on the basis of each worker’s comparative advantage and the prices that prevail in the marketplace.
  • The center of gravity of the triangle representing the hunters in the population, located at the intersection of the median bisectors, corresponds to the average values of ρ and τ in the self-selected population of hunters.
  • The corresponding averages in the population of fishermen will exceed the population average, (½)T₀, because so many talentless fishermen elect to become hunters in equilibrium.
    • Income inequality will arise in equilibrium, since on average fishermen will earn more than hunters. Policymakers cannot exploit this difference to make everyone better off—it would be a mistake to encourage hunters to become fishermen.
    • Akerlof (1981) argued that there is an opportunity cost to filling jobs. Once one worker takes a job, no one else can take that job, and if that worker is less skilled than the average worker, the firm will suffer an economic loss. Thus the firm will either charge the less-skilled worker a negative wage, or more realistically, leave the job open until it can find a more skilled applicant.

13.3 Asymmetric Information I: Signaling

• In assumption 13.1A, workers (the sellers of labor) have more information about their abilities than prospective employers (the buyers).
  • High-ability workers have a strong incentive to convey their ability to employers.
  • Low-ability workers have a strong incentive to conceal their ability from employers.
    • High-ability workers need a means whereby they can signal their ability to employers in a credible manner. Credibility requires that the signal cannot be copied by low-ability workers or that it is too costly to do so.
    • Spence demonstrates that education may function as a signaling mechanism since high-ability workers cannot simply state that they are have high abilities.
• In the theory of education as a signaling mechanism, it is less costly for high-ability students to obtain a given set of credentials.
  • Assumption 13.2 The Labor Market
    • (a) The labor market is competitive and populated by equal numbers of high-ability, H, and low-ability, L, college graduates.
    • (b) Each high-ability worker produces output valued at a_H = $100, and each low-ability worker produces output valued at a_L = $40.
    • (c) Assumption 13.1A holds; workers know their own abilities but employers do not.
      • Without any market signaling, everyone will be paid the common equilibrium wage, which is equal to the productivity of the average worker.
  • Assumption 13.3 Education as a Signal
    • (a) In terms of their time and effort, it costs $50 for high-ability students to increase their GPAs from 2.75 to 3.75. It costs low-ability students $65 to effect the same increase.
    • (b) The higher GPA has no effect whatsoever on workers’ productivities.
      • Assumption 13.3(b) simply allows us to focus on the message of the signaling model.
      • Firms will offer a_H = $100 to any worker with a 3.75 GPA and a_L = $40 to any worker with a 2.75 GPA based on what the firms expect about ability and productivity.
      • High-ability students will study harder to increase their net income to ($100–$50) = $50, rather than earning $40 with a low GPA
      • Low-ability students will not study harder because they will have a net income of ($100–$65) = $35 with a high GPA and a net income of $40 with a low GPA.
    • This is a separating equilibrium. Nobody has any incentive to change his behavior. The signal truthfully reveals two types of ability.
  • The same type of signaling mechanism may apply to high-ability workers who are laid off. Their optimal choice may be to stay unemployed instead of taking a low-wage job that would signal low ability in the future.
  • If there are three or more ability types, high ability students may also engage in countersignaling, whereby they act “too cool for school.”
    • Medium-ability students also answer questions in class and study hard to get good grades on tests to differentiate themselves from low-ability students.
    • High-ability students can try to distinguish themselves from medium-ability students by selectively answering only the most difficult questions.
    • Moreover, the Bill Gates phenomenon may explain why gifted students drop out of school in order to distinguish themselves from ordinary students.
• Weiss (1983) noted that it is optimal for firms to give students job offers before they graduate, since the students have already signaled their ability by showing up at school.
  • In testing the signaling model, economists have difficulty distinguishing the signaling role of education from the human capital function of education. Both predict a positive association between education and earnings.
    • Wolpin (1977) found the self-employed find only slightly less time in school than their non-self-employed peers, since signaling is not important for business owners. More recent research indicates that the self-employed may still use signaling to signal their creditworthiness.
    • Albrecht and Van Ours (2001) suggested that educational signaling is much more important in formal recruiting channels.
  • Cameron and Heckman (1993) tested the signaling capacity of the General Educational Development (GED) credential and found that passing the GED increases the earnings of White high school dropouts by between 10% and 19% but has no effect on the earnings of minorities.
  • Lang and Kropp (1986) found some support for the signaling hypothesis, since compulsory attendance laws (CALs) increase the enrollment rates of students in the age groups they do not directly affect.
    • Older students must voluntarily stay in school a year longer in order for the mechanism to work.

13.4 Asymmetric Information II: Screening

• To screen applicants employers design job packages to encourage applications from some types of workers and discourage other types.
  • A long and onerous probationary period before promotion may discourage less skilled lawyers from applying to a law firm.
  • Model 13.2: Screening and Self-Selection
    • (a) The market for lawyers is competitive.
    • (b) After attaining a law degree a lawyer’s career has a junior and senior phase.
    • (c) Half of the lawyers possess intrinsically high ability worth $200 per hour. The other half has low ability worth $100 per hour.
    • (d) Assumption 13.1A holds. Junior lawyers know their abilities but law firms do not. A senior lawyer’s track record publicly reveals his ability.
      • Law firms offer graduates a career package (w_j, w_s), where w_j and w_s denote the earnings of junior and senior lawyers, respectively.
      • Type-H firms hire high-ability graduates; type-L firms hire low-ability graduates.
  • If type-H firms offer a ($200, $200) package while type-L firms offer a ($100, $100) package:
    • Junior lawyers of all types will apply to the type-H firm.
      • Low-ability workers will be fired and will move to the type-L firm, but not before they have earned $200.
    • Type-H firms will not be able to turn a profit as a result of the masquerading low-ability workers.
  • If type-H firms offer a ($90, $310) package while type-L firms offer a ($100, $100) package.
    • High-ability workers will apply to the type-H firm and earn ($90 + $310) = $400 upon promotion
      • If low-ability ability workers apply to the type-H firm, they will be fired and forced to move to a type-L firm, earning a net ($90 + $100) = $190.
    • Instead, low-ability workers will apply only to type-L firms, earning a net ($100 + $100) = $200
      • The result is a screening equilibrium.
• Gausch and Weiss (1980) interpret tests and fees in the application process as a probationary periods and low initial starting wages, respectively.
  • Both function as screening and sorting mechanisms.
  • Despite constant productivity, earnings of high-ability workers are initially low in the probationary period and high after they pass the test.
  • This explanation for why wages rise with tenure is radically different from the human-capital model.
• According to the signaling hypothesis, workers are the first to act.
  • They undertake the type of education corresponding with their ability and then search for work at firms who respond to the signals in their educational choices.
• According to the screening hypothesis, employers are the first to act.
  • Firms offer different career packages, and workers respond by revealing their information via market choices.

13.5 Symmetric Information

• Contrary to parts P and A of Assumption 13.1, it is possible that both employers and workers are equally ignorant about workers’ abilities at the time of hiring, as is the case in assumption 13.1S.
  • As their careers continue workers’ true abilities are revealed.
  • Lazear (1998) proposed a model of risky and safe workers.
  • Model 13.3: Risky vs. Safe Workers
    • (a) There are two time periods and two types of workers: safe, s, and risky, r. At the start of the first time period the firm must decide which to hire.
    • (b) Safe workers have a constant ability worth ā per period.
    • (c) A risky worker’s true ability is unknown at the beginning of the first period. Everyone knows that there is a 50% chance he is worth a_H and a 50% chance he is worth a_L. His track record reveals his ability at the start of the second period.
    • (d) The firm can hire a risky worker for $w_r per period and a safe worker for $w_s.
      • Worker riskiness declines with experience.
      • Risky workers have better option values than safe workers.
        • After learning the risky worker’s ability, the firm exercises the option of terminating the worker only if the worker is not profitable.
          • The firm can profit from hiring potentially high-skilled risky workers.
          • The firm experiences no downside losses from firing low-skilled risky workers.
      • w_r > w_s since risky workers will capture some of their higher expected returns in their higher wages.
  • Lazear’s framework predicts that new firms in growing industries prefer younger, riskier workers. Firms in declining industries prefer older, safer workers.
• MacDonald’s person-specific information framework also provides an information-based theory explaining why earnings increase with experience.
  • As the firm gathers information about a worker, the firm assigns the worker to the task for which her skill set is most valuable.
  • Even though her innate ability remains constant, her productivity and earnings increase with experience.