5D Library › 多阶段最后通牒议价、应收账款衰减动力学与最优施压序列 · English

第六章 加薪与提价:建立替代选择

把加薪看作纳什议价

费希尔与尤里教会了一代谈判者:一个人的力量在于谈判协议之外的最佳替代方案,即 BATNA (Fisher and Ury 1981)。纳什的议价解把这一点说得更精确:协议把双方替代选择之间的剩余分开 (Nash 1950)。p11 把它用于加薪。一位月薪 5,000 的员工,计入替换成本,对雇主值 7,000;议定的工资是员工的替代选择加上差距的一个份额。

推导

# Chapter 6. p11: A raise as Nash bargaining. An employee earns 5,000
# a month. To the employer the employee is worth V = 7,000 a month
# (replacing them would cost that much in output, hiring and
# training). The employee's best alternative (BATNA) is w0. The
# symmetric Nash bargaining wage splits the surplus between the two
# alternatives: w = w0 + 0.5 (V - w0). With no outside offer, w0 is
# the current wage; with an outside offer of 6,000, w0 = 6,000. An
# employee with bargaining weight beta < 0.5 gets w0 + beta (V - w0).
def wage(w0, v=7000.0, beta=0.5):
    return w0 + beta * (v - w0) if v > w0 else w0

if __name__ == "__main__":
    print("   outside offer   bargaining weight   wage    raise")
    for w0 in (5000, 6000, 6500):
        for beta in (0.5, 0.3):
            w = wage(w0, beta=beta)
            print("   {:13,d}   {:17.1f}   {:5,.0f}   {:5.1%}".format(
                w0, beta, w, w / 5000 - 1))
    print("   offer of 7,500 (above V): the employer cannot match it, and")
    print("   the employee leaves for {:,}".format(7500))
#    outside offer   bargaining weight   wage    raise
#            5,000                 0.5   6,000   20.0%
#            5,000                 0.3   5,600   12.0%
#            6,000                 0.5   6,500   30.0%
#            6,000                 0.3   6,300   26.0%
#            6,500                 0.5   6,750   35.0%
#            6,500                 0.3   6,650   33.0%
#    offer of 7,500 (above V): the employer cannot match it, and
#    the employee leaves for 7,500

没有外部录用、议价力量各半时,模型工资是 6,000,加薪 20%;议价力较弱(0.3)时是 5,600。6,000 的外部录用把工资推到 6,500 或 6,300;6,500 的录用推到 6,750 或 6,650。7,500 的录用高于员工对雇主的价值,雇主无法匹配,员工会离开去接受它。同样的算术也支配供应商的提价:客户的替代选择是转换成本,供应商的替代选择是次佳客户的价值。

两种议价权重下,员工最佳外部录用与议定工资;虚线标出员工对雇主的价值(p11)。

建立替代选择,还是虚张声势

替代选择需要去建立,而建立要花时间。p12 比较真实的求职与虚张声势。

推导

# Chapter 6. p12: Building a real alternative or bluffing about one.
# Searching costs 1,500 (time and effort) and yields an offer of 6,000
# with chance q. With a real offer the p11 wage is 6,500; without one
# it is 6,000 (a raise from 5,000 in an ordinary review). A bluff ("I
# have an offer of 6,000") works with chance 0.5; if the employer calls
# it, the employee keeps 5,000 and loses goodwill worth 2,000. Values
# over 12 months.
if __name__ == "__main__":
    base = 12 * 6000
    print("   staying with the ordinary raise: {:,}".format(base))
    print("   chance of an offer q   value of searching   gain over none")
    for q in (0.2, 0.5, 0.8):
        v = q * 12 * 6500 + (1 - q) * base - 1500
        print("   {:20.1f}   {:18,.0f}   {:14,.0f}".format(q, v, v - base))
    bluff = 0.5 * 12 * 6500 + 0.5 * (12 * 5000 - 2000)
    print("   bluffing: {:,.0f}, gain over none {:,.0f}".format(
        bluff, bluff - base))
#    staying with the ordinary raise: 72,000
#    chance of an offer q   value of searching   gain over none
#                     0.2               71,700             -300
#                     0.5               73,500            1,500
#                     0.8               75,300            3,300
#    bluffing: 68,000, gain over none -4,000

一年下来,留在原地拿普通加薪值 72,000。花 1,500 去求职,若只有五分之一的机会拿到录用,值 71,700;机会各半时 73,500;五分之四时 75,300。谎称有一份并不存在的录用,值 68,000,损失 4,000,因为雇主一旦戳穿,员工只能拿旧工资,还失去信任。真实的替代选择即使从未动用也能增强实力;捏造的替代选择一经检验就削弱实力。

假说 · 尚未验证

作者的诠释(尚未验证): 把工作表现固定之后,在薪资谈判前取得真实外部录用的员工,得到的加薪大于声称有录用而实际没有的可比员工。什么会证明它错: 在一项核实录用真伪的预先登记调查中,真实录用与声称录用的员工之间加薪没有差别。

小结

加薪或提价的条件主要由双方的替代选择决定。当真实替代选择有相当机会带来录用时,建立它值得付出的成本;捏造一个则是一场通常会输的赌博。

参考文献

Fisher, Roger, and William Ury. 1981. Getting to Yes: Negotiating Agreement without Giving in. Houghton Mifflin.
Nash, Jr., John F. 1950. “The Bargaining Problem.” Econometrica 18 (2): 155 to 162.

Fan, Y. (2026). Multistage Ultimatum Bargaining, Accounts-Receivable Decay Kinetics and Optimal Escalation Sequences: Ackerman Concession Curves, Credible Threat Thresholds and Default Recovery Bounds [多阶段最后通牒议价、应收账款衰减动力学与最优施压序列] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090245 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).