关于「三种混淆:整数堆积、识别极限与申报反应」,《门槛聚束与营业额弹性》给出的定量结论是什么?

结论:基准比值是 0.6007。R 世界里,98 到 100 那一格挤进了 884 家,比值 0.5991,与基准无异;M 世界里,同一格有 843 家,比值升到 0.6315,比基准高出 13.0 个标准误。真实反应让销售与刷卡一起下降,比值不变;申报反应只压低了分母,比值就抬了起来。

数据口径:书中参数为示意,门槛一律记作 T;全部计算由书中程序生成。 第一章在一份固定种子的示意分布上估出超额质量 942.0 家、观察弹性 0.1378,结构弹性为 0.25,差距来自摩擦与反事实近似。

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: 0.5991, 0.6007, 0.6315, 100, 13.0, 843

baseline traced/declared (60-90): 0.6007
binn(R)ratio(R)z(R)n(M)ratio(M)z(M)
92- 943300.5971-1.13210.5966-1.4
94- 963150.5911-3.13170.5928-2.5
96- 983030.5983-0.73450.60160.3
98-1008840.5991-0.98430.631513.0
100-1021800.60270.41720.5939-1.6
102-1042070.5952-1.41660.5987-0.5
104-1061930.5962-1.11850.5974-0.8
106-1081910.60320.61690.60280.5
查看完整验算程序
# Chapter 5. p53: real response or reporting response?
# Bunching below T can come from firms that really sell less or from
# turnover that is under-declared. A third-party trace (here the
# card-payment share of sales, phi, drawn between 0.5 and 0.7)
# separates them: a real response keeps traced / declared at phi; an
# under-declaration raises it. World R: a share 0.3 of firms with
# true sales in (T, T + 15] really stop just below T. World M: the
# same firms keep selling but declare just below T. Illustrative.
import math
import random

SEED, N, T, SHARE = 20261113, 20000, 100.0, 0.3


def world(rng, misreport):
    rows = []
    for _ in range(N):
        s = rng.lognormvariate(math.log(80.0), 0.5)
        phi = rng.uniform(0.5, 0.7)
        moved = T < s <= T + 15 and rng.random() < SHARE
        below = T - abs(rng.gauss(0.0, 1.0))
        if moved and misreport:
            rows.append((below, phi * s))
        elif moved:
            rows.append((below, phi * below))
        else:
            rows.append((s, phi * s))
    return rows


def by_bin(rows, lo, hi):
    vals = [t / d for d, t in rows if lo <= d < hi]
    m = math.fsum(vals) / len(vals)
    sd = math.sqrt(math.fsum((v - m) ** 2 for v in vals) / (len(vals) - 1))
    return len(vals), m, sd / math.sqrt(len(vals))


if __name__ == "__main__":
    rng = random.Random(20261113)
    wr, wm = world(rng, False), world(rng, True)
    base = by_bin(wr, 60.0, 90.0)[1]
    print("baseline traced/declared (60-90): {:.4f}".format(base))
    print("bin        n(R)  ratio(R)    z(R)   n(M)  ratio(M)    z(M)")
    for lo in range(92, 108, 2):
        nr, mr, sr = by_bin(wr, lo, lo + 2)
        nm, mm, sm = by_bin(wm, lo, lo + 2)
        print("{:3d}-{:3d} {:7d} {:9.4f} {:7.1f} {:6d} {:9.4f} {:7.1f}"
              .format(lo, lo + 2, nr, mr, (mr - base) / sr, nm, mm,
                      (mm - base) / sm))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《起征点聚束估计与营业额分布、门槛下方堆积揭示的行为弹性母典》第 5 章「三种混淆:整数堆积、识别极限与申报反应」(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-319 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026319,
  author    = {Fan, Yafu},
  title     = {Bunching at Tax Thresholds and the Distribution of Turnover: Behavioural Elasticities Revealed by Piling-Up below the Line},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论