What does Negative Working-Capital Flywheels conclude, in numbers, about “Eastern and Western Sources: Jiran, Bai Gui, Smith and Bagehot”?

Answer: Chapter 7 examines Jiran and Bai Gui in the Shiji, Smith and Bagehot; in an illustrative simulation, same-week refunds lead to 216.7 requests and stalling to 686.9.

Scope of the numbers: Revenues, margins, payment terms, refund rates, acquisition costs and prices are illustrative, and all calculations are generated by the book's programs. Chapters 4 to 6 use fixed-seed Monte Carlo to show that single shocks rarely break a presale business while stacked shocks do; under illustrative parameters the cautious reserve-ratio red line is 0.474.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 216.7, 686.9

policyrequests in 60 dayspeak queuerefunds paid
free216.70.0216.7
stall686.9507.6179.3
refunding freely costs less in total: stalling saves cash for a
few days and multiplies the claims that arrive afterwards
Show the full program
# Chapter 7. p72: Bagehot's rule for a presale - refund freely or stall?
# After a delivery delay is announced, 1,000 prepaid buyers (1 unit each)
# may ask for refunds over 60 days. Daily request hazard per open buyer:
#   h = 0.004 + 0.05 * (unpaid queue / 1,000) + 0.0002 * (days of queue)
# where "days of queue" counts days on which some request went unpaid.
# Policy "free": every request is paid the same day from the reserve.
# Policy "stall": at most 3 refunds a day are paid, the rest wait.
# 40 seeded runs each (seed 20261120 + run). Coefficients illustrative.
import math
import random

N, DAYS, RUNS = 1000, 60, 40


def run(policy, r):
    rng = random.Random(20261120 + r)
    open_, queue, paid, qdays, peak = N, 0, 0, 0, 0
    for _ in range(DAYS):
        h = 0.004 + 0.05 * queue / N + 0.0002 * qdays
        new = 0
        for _ in range(open_):
            if rng.random() < h:
                new += 1
        open_ -= new
        queue += new
        pay = queue if policy == "free" else min(queue, 3)
        queue -= pay
        paid += pay
        if queue:
            qdays += 1
        peak = max(peak, queue)
    return paid + queue, peak, paid


print("policy   requests in 60 days   peak queue   refunds paid")
for policy in ("free", "stall"):
    req, peaks, paid = [], [], []
    for r in range(RUNS):
        total, peak, out = run(policy, r)
        req.append(total)
        peaks.append(peak)
        paid.append(out)
    print("%-7s %16.1f %16.1f %14.1f" % (
        policy, math.fsum(req) / RUNS, math.fsum(peaks) / RUNS,
        math.fsum(paid) / RUNS))
print("refunding freely costs less in total: stalling saves cash for a")
print("few days and multiplies the claims that arrive afterwards")

Single source and citation

Fan, Y. (2026). Negative Working-Capital Flywheels, Upfront Customer Float and Zero-Capital Bootstrap Scaling (Version 1.0.0), Chapter 7: Eastern and Western Sources: Jiran, Bai Gui, Smith and Bagehot. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-365 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026365,
  author    = {Fan, Yafu},
  title     = {Negative Working-Capital Flywheels, Upfront Customer Float and Zero-Capital Bootstrap Scaling},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book