关于「个人连带担保:重尾损失下的破产放大与单次大跳」,《责任图渗流与担保尾部破产》给出的定量结论是什么?

结论:第四章说明个人连带担保把经营者十年内破产的概率放大 7.0 倍(轻尾)到 48.6 倍(重尾),破产由单独一年的大损失主导。

数据口径:书中的资产负债表、损失分布与法律参数均为示意;全部计算由书中程序生成;本书不是个人化的法律、税务或财务建议,各法域规则不同,书中也不提供任何规避合法债务或逃避执行的方法。 书中以标注框区分已发表的证据、可运行的 Python 计算(17 段程序)、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。 程序只用 Python 标准库与结论吻合的打印值: 48.6, 7.0

P(owner ruined within 10 years) by guarantee cap G
loss lawG = 0G = 1G = 4unlimitedratio
exponential0.00520.01600.03490.03627.0
lognormal0.00970.03390.11240.145815.0
Pareto a=2.50.00040.00180.01520.022349.5
Pareto a=1.50.00200.01010.06160.098648.6
查看完整验算程序
# Chapter 4. p41: how a personal guarantee turns a business's loss tail
# into the owner's ruin. Each year the company (capital E) takes a loss L
# with mean 1. If L > E the company fails: the owner refounds it (pays E)
# and pays min(L - E, G) under a personal guarantee capped at G. The
# owner starts with W0 outside the company and earns Y a year; ruin =
# personal wealth below zero within T years. Four loss laws share the
# same mean. Seeded Monte Carlo, common random numbers across G.
# All parameters are illustrative.
import math
import random

E, W0, Y, T, PATHS = 3.0, 4.0, 0.8, 10, 100000
CAPS = (0.0, 1.0, 4.0, math.inf)
rng = random.Random(20261112)
S_LN = 1.2


def draw(kind):
    u = 1.0 - rng.random()
    if kind == "exponential":
        return -math.log(u)
    if kind == "lognormal":
        return math.exp(rng.gauss(-S_LN * S_LN / 2, S_LN))
    if kind == "Pareto a=2.5":
        return 0.6 * u ** (-1 / 2.5)
    return (1 / 3) * u ** (-1 / 1.5)          # Pareto a=1.5, mean 1


def ruin_row(kind):
    hits = [0] * len(CAPS)
    for _ in range(PATHS):
        losses = [draw(kind) for _ in range(T)]
        for c, cap in enumerate(CAPS):
            w = W0
            for x in losses:
                w += Y
                if x > E:
                    w -= E + min(x - E, cap)
                if w < 0:
                    hits[c] += 1
                    break
    return [h / PATHS for h in hits]


print("P(owner ruined within {} years) by guarantee cap G".format(T))
print("{:14s}{:>10s}{:>10s}{:>10s}{:>11s}{:>9s}".format(
    "loss law", "G = 0", "G = 1", "G = 4", "unlimited", "ratio"))
for kind in ("exponential", "lognormal", "Pareto a=2.5", "Pareto a=1.5"):
    r = ruin_row(kind)
    print("{:14s}{:10.4f}{:10.4f}{:10.4f}{:11.4f}{:9.1f}".format(
        kind, *r, r[3] / r[0]))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《多子公司责任隔离图渗流、个人连带担保尾部破产放大与经营资产风险仓切分》第 4 章「个人连带担保:重尾损失下的破产放大与单次大跳」(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-177 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026177,
  author    = {Fan, Yafu},
  title     = {Liability Percolation across Subsidiary Graphs, Tail-Ruin Amplification from Personal Guarantees and Risk-Bucket Partitioning of Operating Assets},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论