关于「总额还是净额:按控制权判主体,而不是按平台怎么打款」,《多主体清分与审计就绪账本》给出的定量结论是什么?

结论:第一章按 IFRS 15 的控制判断给七类合成交易打分,「到账即收入」有 3 类出错,误差有正有负,合计高估 0.2763。

数据口径:书中参数为示意,数据为合成;全部计算由书中程序生成。 第一章按 IFRS 15 的控制判断给七类合成交易打分,「到账即收入」有 3 类出错,误差有正有负,合计高估 0.2763。

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: 0.2763

A. score, role, revenue by control vs revenue = cash landed
deal score role by-control landed gap
A1 principal 100.00 96.60 -3.40
B0 agent 14.00 14.00 0.00
C1 principal 20.00 19.02 -0.98
D1 principal 1,000.00 1,000.00 0.00
E0 agent 80.00 80.00 0.00
F0 agent 40.00 540.00 500.00
G1 principal 540.00 540.00 0.00
B. deals where cash landed differs from control view3 of 7
C. revenue by control 1,794.00 u; by cash landed 2,289.62 u
D. cash-landed habit overstates total revenue by 0.2763
E. note: deal A is understated, deal F overstated; errors run
both ways, so a net total can hide two opposite mistakes
查看完整验算程序
# Chapter 1. p11: principal-or-agent scoring of seven synthetic deals.
# Indicators follow IFRS 15 B37 (primary responsibility, inventory
# risk, pricing discretion); they support the control judgement and
# are not a checklist. Majority of the three is used here only as an
# illustrative screening rule. Amounts are illustrative units (u),
# exact Decimal with banker's rounding; no deal is a real contract.
from decimal import Decimal, ROUND_HALF_EVEN
from fractions import Fraction

Q = Decimal("0.01")

# code, form, resp, inv, price, paid_by_customer, kept_by_others, landed
DEALS = [
    ("A", "site report via PSP", 1, 1, 1, "100.00", "3.40", "96.60"),
    ("B", "ebook, store sells", 0, 0, 0, "20.00", "6.00", "14.00"),
    ("C", "ebook direct, PSP", 1, 1, 1, "20.00", "0.98", "19.02"),
    ("D", "B2B campaign", 1, 1, 1, "1000.00", "700.00", "1000.00"),
    ("E", "B2B referral fee", 0, 0, 0, "1000.00", "920.00", "80.00"),
    ("F", "buying agent", 0, 0, 0, "540.00", "500.00", "540.00"),
    ("G", "own-stock resale", 1, 1, 1, "540.00", "450.00", "540.00"),
]


def judge(resp, inv, price):
    score = Fraction(resp + inv + price, 3)
    return score, ("principal" if score >= Fraction(2, 3) else "agent")


def revenue(role, paid, kept):
    # principal: gross consideration; agent: the net fee it keeps
    return paid if role == "principal" else paid - kept


if __name__ == "__main__":
    print("A. score, role, revenue by control vs revenue = cash landed")
    print("  deal score role        by-control  landed     gap")
    tot_ok = Decimal("0")
    tot_cash = Decimal("0")
    wrong = 0
    for code, form, r, i, p, paid, kept, landed in DEALS:
        paid, kept, landed = Decimal(paid), Decimal(kept), Decimal(landed)
        score, role = judge(r, i, p)
        ok = revenue(role, paid, kept).quantize(Q, ROUND_HALF_EVEN)
        gap = landed - ok
        tot_ok += ok
        tot_cash += landed
        wrong += 1 if gap != 0 else 0
        print("  {}    {:<4} {:<10} {:>10}  {:>9}  {:>8}".format(
            code, str(score), role, "{:,}".format(ok),
            "{:,}".format(landed), "{:,}".format(gap)))
    print("B. deals where cash landed differs from control view: {} of {}"
          .format(wrong, len(DEALS)))
    print("C. revenue by control {:,} u; by cash landed {:,} u".format(
        tot_ok, tot_cash))
    diff = (tot_cash - tot_ok) / tot_ok
    print("D. cash-landed habit overstates total revenue by {}".format(
        diff.quantize(Decimal("0.0001"), ROUND_HALF_EVEN)))
    print("E. note: deal A is understated, deal F overstated; errors run")
    print("   both ways, so a net total can hide two opposite mistakes")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《跨国数字业务多主体清分、跨境税基合规边界与一键审计就绪账本母典》第 1 章「总额还是净额:按控制权判主体,而不是按平台怎么打款」(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-402 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026402,
  author    = {Fan, Yafu},
  title     = {Multi-Jurisdiction Digital Revenue Clearing, Cross-Border Tax-Base Compliance Boundaries and Audit-Ready Automated Ledgers},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论