关于「委托-代理问题与本地团队的激励」,《跨文化契约与出海博弈》给出的定量结论是什么?

结论:两类合伙人各自选择了为他们设计的合同:长期型选择合伙型合同(价值 378.9 对 318.0),短期型选择工资型合同(103.7 对 79.2)。递延款与较高的分成对长期型更有吸引力,因为他们更可能拿到递延款,也更不在意每月收入的波动。代价是:对长期型合伙人,合伙型合同的总成本(474.0)高于工资型(360.0)。这部分额外成本是总部为「区分类型、留住长期伙伴」付出的价格;它是否划算,取决于一个长期伙伴能为企业多创造多少价值,而这在模型中没有计算。

数据口径:本书不是法律、财务或投资建议。 书中以标注框区分已发表的证据、可运行的 Python 计算(15 段程序,全部核对)、原典与作者的假说。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 与结论吻合的打印值: 103.7, 318.0, 360.0, 378.9, 474.0, 79.2

long-horizon : A salary 318.0, B partner 378.9 -> picks B partner
short-horizon: A salary 103.7, B partner 79.2 -> picks A salary
cost of A salary for a long-horizon partner:360.0 over 36 months (still there at month 36: 100%)
cost of A salary for a short-horizon partner:139.8 over 36 months (still there at month 36: 11%)
cost of B partner for a long-horizon partner:474.0 over 36 months (still there at month 36: 100%)
cost of B partner for a short-horizon partner:142.0 over 36 months (still there at month 36: 11%)
查看完整验算程序
# A contract menu that lets local partners sort themselves (screening). Two kinds of partner apply:
#   long-horizon  - expects to stay 3+ years, monthly discount factor 0.995, risk aversion 1.0
#   short-horizon - may leave within a year (leaves with probability 0.06 a month), discount 0.98,
#                   risk aversion 2.0
# Monthly revenue the partner influences: mean 100, SD 40 (illustrative units, e.g. thousand MYR).
# Menu, 36 months:
#   A 'salary'  : fixed 8 a month + 2 % of revenue
#   B 'partner' : fixed 5 a month + 4 % of revenue + a deferred tranche of 150 paid at month 36
#                 to a partner who is still there (forfeited on leaving)
# Each type's value = expected discounted pay - risk premium (CARA, r/2 * variance of monthly pay).
def value(contract, beta, leave, r):
    fixed, share, tranche = contract
    v, alive = 0.0, 1.0
    for m in range(1, 37):
        alive *= 1 - leave
        monthly = fixed + share * 100 - r / 2 * (share * 40) ** 2
        v += alive * beta ** m * monthly
    return v + alive * beta ** 36 * tranche

MENU = {"A salary": (8, 0.02, 0), "B partner": (5, 0.04, 150)}
TYPES = {"long-horizon": (0.995, 0.0, 1.0), "short-horizon": (0.98, 0.06, 2.0)}
for t, (beta, leave, r) in TYPES.items():
    vals = {c: value(k, beta, leave, r) for c, k in MENU.items()}
    pick = max(vals, key=vals.get)
    print(f"{t:13s}: " + ", ".join(f"{c} {v:6.1f}" for c, v in vals.items()) + f"  -> picks {pick}")

# Expected cost to headquarters per hire (undiscounted, with each type's chance of staying):
for c, (fixed, share, tranche) in MENU.items():
    for t, (beta, leave, r) in TYPES.items():
        stay = sum((1 - leave) ** m for m in range(1, 37))
        cost = stay * (fixed + share * 100) + (1 - leave) ** 36 * tranche
        print(f"cost of {c} for a {t} partner: {cost:6.1f} over 36 months "
              f"(still there at month 36: {(1 - leave) ** 36:.0%})")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《跨文化契约、实物期权与东南亚出海博弈论》第 6 章「委托-代理问题与本地团队的激励」(版本 1.0.1). 5D Library. https://doi.org/10.5281/zenodo.23765345

DOI: 10.5281/zenodo.23765345 · 阅读本章所在专著 · 5DL-B-016 · v1.0.1 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026016,
  author    = {Fan, Yafu},
  title     = {Cross-Cultural Contracts, Real Options and Game Theory of Southeast Asian Market Entry},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.1},
  doi       = {10.5281/zenodo.23765345},
  url       = {https://doi.org/10.5281/zenodo.23765345}
}

本书其他章节结论