Incentives by mamapeirong


									Compensation and incentives
        Managing compensation:
          A map of concepts and tools
            (Baker, Jensen & Murphy’ 88)
▪ Compensation level
  ♦  where people work: recruitment, turnover
▪ Compensation composition
  ♦  the value (cost, utility) of a certain level
▪ Compensation function
  ♦  performance
                Compensation level
▪   = Value of “job package”:
     ♦ effort, money, term, human capital, specificity, fringe benefits, etc.
▪   How to set up? Multidimensional, lack of prices  Look at:
     ♦ turnover
     ♦ queues of qualified candidates
     ♦ costs and benefits of turnover
▪   Job rating systems (“Hay” plans)  aggravates bureaucracy
▪   “Efficiency wages” = higher than market wages  Two origins:
     ♦ Market  compensation level or function?
     ♦ Regulation  rent seeking
       Compensation composition
▪   Should not trigger conflict between parties
▪   Effects on labor relations
     ♦ Self-selection: e.g., health plans, deferred  low subjective
       discount rate
     ♦ Moral hazard: paying with product discourages theft
     ♦ Self-control: pensions, work safety
▪   Effects on other relations
     ♦ Adverse selection: reduces transaction costs by grouping: health
     ♦ Moral hazard: e.g., easily-to-enforce credit to employees
     ♦ Tax avoidance by paying in kind
▪   Growing demand for flexibility:
     ♦ Demographic variety and change  ‘Flexible compensation’
▪   Marketing view of compensation package: selling jobs
      Compensation function (1)
▪ How compensation level depends on
▪   Basic types of variable compensation
    ♦ Piece rates, commissions, bonus plans, etc.
    ♦ BUT not only money, not only short term:
       •   Task allocation between subordinates
       •   Promotions and salary revisions
       •   Firing and penalties
       •   Non-vested pensions and deferred compensation
              Compensation function (2).
    Effects and limitations of pay for performance
▪   Big problem in economics literature, small in real life:
    “inefficient” risk allocation (principal-agent model)
▪   Main real problem: produces harder effort and smarter effort—
    BUT poses measurement problems:
     ♦   Gap between performance and contribution to firm value
     ♦   Quality: Moscow taxis; MBO and killings of civilians in Viet Nam
     ♦   Gaming: reallocation of performance between periods or units
     ♦   Externalities:
          •   Sabotage of others in relative performance, between shifts  Pay
              compression useful to enhance cooperation
          •   Damage to clientele: reputation effects
     ♦  Important to group together activities in which performance can
       be measured in a similar way
    1st party                           2nd party
  enforcement            Effort

       Uncertainty    Performance   Firm value

                      performance   Incentives

                       Verifiable     3rd party
                      performance   enforcement
3rd party, judicial
Compensation Function (3). Group or
▪ Key issue: relative advantage of specialized
    versus mutual performance evaluation
▪   Larger group size makes specialization
    relatively preferable
▪   Greater within-team interactions make mutual
    systems preferable
Compensation Function (4). Puzzles
    in variable compensation
▪   Short term not predominant, even in the US—but
    long-term variability hard to estimate
▪   Asymmetric use of prizes and punishments: wage
    reduction damages morale and is seen as unfair 
    better to dismiss underperformers
▪   Use of ceilings —to avoid aberrant effects? Aberrant
    results, exogenous risk, equality.
▪   Use of minimums despite causing reallocation (e.g.,
    in sales, public works)—as insurance?
         Compensation function (5).
          Performance evaluation
▪ Objective evaluation
  ♦ Objective aberrations
  ♦ Contractual problems: standards
    tricky to set and move
     •   Employees self-protect through social
         norms stipulating a certain effort
     •   Difficult to change: ‘ratchet’ effect with
         piece rate: moving standards   
           – Factors: technical change, trust
           – Famous ‘Lincoln Electric’ case Next
         Working / Careers at Lincoln Electric
▪   “Both entry-level and skilled positions require working a rotating shift on
    either a two or three shift basis. In addition, most plant operations positions
    pay on a piecework basis. That means pay is based directly on output.
▪   All of these positions offer excellent opportunities for quality-conscious,
    productivity-minded individuals who are interested in being rewarded for
    their excellent performance.
▪   Incentive Performance - A Cornerstone of Our Culture. The Incentive
    Performance System primarily attributed to James F. Lincoln has been in
    place at Lincoln Electric since the early twentieth century. It has resulted in
    one of the oldest "pay-for-performance" systems in the country, and is
    frequently used for benchmarking by other businesses and studied by
    academics around the world.
▪   The Incentive Performance System in place in the U.S. Lincoln operations
     ♦   An elected Advisory Board for direct and open communication with senior
         management since 1914.
     ♦   Piecework incentive rewards for all production work.
     ♦   A profit-sharing Bonus Plan for employees paid annually at the discretion
         of the Board of Directors since 1934.
     ♦   Guaranteed Employment after three years of service. The company has
         not exercised its layoff options in the U.S. operations since post war 1948.
                                                                                        “Through this well-defined group of
     ♦   401(k) plan offering the employee a variety of pre-tax investment options.     incentives, Lincoln encourages and
     ♦   Competitive compensation and other benefits.                                   compensates individual initiative and
     ♦   A Financial Security Program which includes company contributions based        responsibility. Employees work together
         on years of service from 4 - 10%.                                              to reduce costs and improve quality.
                                                                                        These individual and cooperative efforts
     ♦   An attractive vacation package based on comparable years of service.”          create a more profitable company, the
                                                                                        success of which each person shares
                                                                                        according to his or her own contribution.”
             Compensation function (5).
              Performance evaluation
▪   Objective evaluation
▪   Subjective evaluation
     ♦ Corrects objective aberrations (Baker/Gibbons/Murphy QJE ‘94)
     ♦ Requires relational contracting, 2nd party enforcement
     ♦ Causes subjective aberrations
         •   Systematic and unintentional failures  Questionnaires
         •   Rent-seeking ( “influence”) activities
     ♦ Rejected by middle managers (b/c we value ourselves too bad / too good in
       absolute / relative terms)  Important: amounts & expectations
▪   Frequency of evaluation:
     ♦ Cost: Measurement + Rent seeking
         •   E.g., promotions in Japanese firms, entry in Civil Service
     ♦ Value: before specific investments
    Compensation function (6). Careers
       & deferred compensation
▪   Compensating labor in the long term
▪   Examples: Seniority, non-vested pensions, pantoufles, audit
    firm’s partners, professions with entry barriers, etc.
▪   Long term  investment  expropriation risk if specific
Compensation function (7). Deferred
   compensation—Main issues
▪ Why?
  ♦ Performance unobservable in the short run (e.g.,
    quality, professionals)
  ♦ Also: “specific” human capital (e.g., info required
    for coordination); employees’ risk aversion (?)
▪ How?
  ♦ Compensation through quasi-rents: deferred
         Compensation function (7).
Consequences of deferred compensation
▪   Self selection of people with low discount rate 
    more fraud-adverse?  Lower expected turnover
▪   Stronger effort, if future compensation is not vested
    (threat of losing quasi-rents)
    ♦ What happens with professionals’ regulation?
▪   But quasi-rents can be expropriated 
    ♦  Might be better to determine promotions and deferred
      compensation by seniority than by evaluating performance
    ♦  Vested compensation  still self-selection, but not
      stronger effort: e.g., Civil Service, promotions in Japan
            Compensation function (8)
             How to contract careers
▪   Mandatory law
    ♦ Permanent employment: dismissal triggers established
      indemnity  3rd party, judicial enforcement  parties
      cannot opt out legally
        •   May trigger workers’ opportunism 
        •   Less incentive for employers to contract 
        •   More careful and personal selection, high capital intensity,
▪   Enabling law
    ♦ Idem unless parties contract out, agreeing e.g. on
      termination at will
    ♦ Employer performs “quasi-judicial” function often with the
      help of workers’ boards (e.g., Japan, Lincoln Electric)
           Compensation function (9)
            More issues on careers
▪   Mandatory retirement
     ♦ Japan: 54 years for “permanent” employees
     ♦ Needed in final years b/c compensation > disutility of work
▪   Promotions as a reward system
     ♦ Careful evaluation  controls for long term performance
     ♦ Matching problem (“Peter’s Principle”): also appointment:
       •  better when same abilities required in different ranks
       •  Otherwise, does not motivate technical, salespeople, etc.
       •  Two tracks needed
▪   More 
         Compensation function (&10)
         Still more issues on careers
▪   Tournaments
     ♦   Prizes fixed in advance, allocated through relative performance
     ♦   Spread between prizes is the motivating factor
     ♦   When luck is more important, more spread needed for same effort
     ♦   Advantages (M&R, 384): Ordinal information, Prize pool set in advance
     ♦   Compensation of “Stars”
▪   Tenure & Up or Out
     ♦ Common in consulting and professional firms and universities
     ♦ Explicit tenure helps recruiting good candidates: current workers do not feel
          •   (implicit tenure may result from the fact that matching improves with seniority.
              Institutional constraints very important in Europe (not in Japan)
     ♦ “Up or out”
          •   fresh ideas
          •   eliminates incentives to cheat by unfairly denying promotion: difference of gaps
              needed to encourage employees to invest and employers to promote employees
          •   less costly for “producers” of non-specific human capital: good outside Jobs
Managing our incentives

To top