关于「多任务:二级斜率把力气从卖货推向拉人」,《合规分销与佣金斜率》给出的定量结论是什么?

结论:当上线可以把时间从卖货挪到拉人,二级佣金就不再是无害的纯抽成,而是一只把力气推向拉人的手。在示意参数下,二级斜率要超过一级斜率的五分之一,拉人才开始;二级斜率 0.3 时,每人每月多拉 2.03 人,报表毛利多 3.25 千元,真实毛利却少 0.20 千元。联合最优的二级斜率在下线报表的真实增量不超过一半时为零,只在真实增量很高时才为正,而这个增量恰恰最难核实。下线一层一层往下长,到底能长出多少人?最后一批加入的人又能拉到几个?下一章用分支过程与有限市场回答。

数据口径:最优的线性佣金斜率由努力成本、风险厌恶与噪音决定,示意基准下是 0.5263;给上线按下线业绩分成的二级佣金,主要买到的是拉人头而不是真实零售;按本书采用的最保守读法,以下线业绩给上线计酬本身落在 2005 年《禁止传销条例》列举的行为之内,合规方案的二级斜率应取零,转介绍改由只为真实购买、退货期满才发放的单层顾客推荐奖励来争取。 书中参数为示意;全部计算由书中程序生成;本书是非个人化法律、税务或理财建议,法规只述大意,以上架前人工复核的现行文本为准。

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: 0.20, 0.3, 2.03, 3.25

b10.5263 on own margin; second-tier slope b2 varies
b2 own s recruit h recruits reported real rep/real surplus
0.0010.53 0.00 0.00 10.53 10.53 1.000 5.263
0.1010.53 0.00 0.00 10.53 10.53 1.000 5.103
0.1510.34 0.93 0.47 11.27 10.48 1.076 4.836
0.2010.13 1.97 0.99 12.10 10.43 1.161 4.431
0.309.71 4.06 2.03 13.77 10.32 1.334 3.225
0.409.30 6.14 3.07 15.44 10.22 1.511 1.491
b20.3 vs 0: reported margin +3.25, real -0.20
recruits per month 0.002.03; own selling 10.53 -> 9.71
查看完整验算程序
# Chapter 2. p21: a second-tier slope moves effort from selling to
# recruiting (multitask LEN, after Holmstrom and Milgrom 1991).
# Efforts: s (selling to end consumers), h (recruiting sellers).
# Own margin x1 = s + e1, Var 36; margin reported by recruits
# x2 = h + e2, Var 64 (thousand yuan a month). Of x2 only a share
# LAM = 0.5 is real end-consumer retail (the rest is self-purchase
# and stock loading) and THETA = 0.35 displaces sales that existing
# sellers would have made, so one unit of x2 is worth V2 = 0.15.
# Cost c*(s^2 + h^2)/2 + d*s*h with c = 0.05, d = 0.01 (the two tasks
# compete for the same hours). Pay w = a + b1*x1 + b2*x2; r = 0.5.
# A recruit reports 2 thousand yuan a month on average. Illustrative.

C, D, R = 0.05, 0.01, 0.5
S1, S2 = 36.0, 64.0
LAM, THETA = 0.5, 0.35
V2 = LAM - THETA
PER_RECRUIT = 2.0


def effort(b1, b2):
    # 一阶条件 c*s + d*h = b1, d*s + c*h = b2;非负约束下取角点
    det = C * C - D * D
    s = (C * b1 - D * b2) / det
    h = (C * b2 - D * b1) / det
    if h < 0.0:
        return b1 / C, 0.0
    if s < 0.0:
        return 0.0, b2 / C
    return s, h


def cost(s, h):
    return C * (s * s + h * h) / 2.0 + D * s * h


def surplus(b1, b2, v2=V2):
    s, h = effort(b1, b2)
    prem = R * (b1 * b1 * S1 + b2 * b2 * S2) / 2.0
    return s + v2 * h - cost(s, h) - prem


B1 = 0.5263
print("b1 = {} on own margin; second-tier slope b2 varies".format(B1))
print("   b2  own s  recruit h  recruits  reported  real  rep/real"
      "  surplus")
for b2 in (0.0, 0.1, 0.15, 0.2, 0.3, 0.4):
    s, h = effort(B1, b2)
    rep = s + h
    real = s + V2 * h
    print("{:5.2f} {:6.2f} {:10.2f} {:9.2f} {:9.2f} {:5.2f} {:9.3f}"
          " {:8.3f}".format(b2, s, h, h / PER_RECRUIT, rep, real,
                            rep / real, surplus(B1, b2)))
s0, h0 = effort(B1, 0.0)
s3, h3 = effort(B1, 0.3)
print("b2 = 0.3 vs 0: reported margin {:+.2f}, real {:+.2f}".format(
    (s3 + h3) - (s0 + h0), (s3 + V2 * h3) - (s0 + V2 * h0)))
print("recruits per month {:.2f} -> {:.2f}; own selling {:.2f} -> "
      "{:.2f}".format(h0 / PER_RECRUIT, h3 / PER_RECRUIT, s0, s3))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《合规分销激励相容、二级佣金斜率委托代理模型与反传销层级红线内转介绍最大化母典》第 2 章「多任务:二级斜率把力气从卖货推向拉人」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-279 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026279,
  author    = {Fan, Yafu},
  title     = {Incentive-Compatible Compliant Referral Programmes: Commission Slopes in a Principal-Agent Model, a Second-Tier Slope of Zero and Maximal Referrals within Anti-Pyramid Limits},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论