What does Debiasing Collectible Returns conclude, in numbers, about “After the Correction: Selection Bias in Repeat-Sales Samples”?
Answer: A repeat-sales index sees only the works that sell again, and whether a work sells again depends on how much it has gained. A large study that corrects for this selection moves the average annual return of paintings from 8.7% to 6.3% and the Sharpe ratio from 0.27 to 0.11; other studies that handle their samples carefully also report long-run returns below those of equities. The smallest model shows that the more sensitive the sale decision is to the gain, the more the index overstates, and the bias can be as large as the true return. The next chapter returns to the method itself: how the regression of Bailey, Muth and Nourse turns pairs of prices into an index, and when it is unbiased.
Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not personalised legal or financial advice, and fee schedules and rules are subject to manual review before release. Marked boxes separate published evidence, runnable Python calculations (20 programs), classical sources and the author's hypotheses.
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: 0.11, 0.27
| Published estimates (annual, as reported) | |
| study period basis return | |
| KKV 2016 repeat-sales, uncorrected | 1960-2013 nominal 0.0870 |
| KKV 2016 selection-corrected | 1960-2013 nominal 0.0630 |
| Renneboog-Spaenjers 2013 hedonic | 1957-2007 real 0.0397 |
| Dimson-Spaenjers 2011 stamps | 1900-2008 nominal 0.0700 |
| Dimson-Spaenjers 2011 stamps | 1900-2008 real 0.0290 |
| KKV sample | 32928 paintings, 53 years |
| return drop | 0.024 absolute, 0.276 of the headline |
| Sharpe drop | 0.27 -> 0.11, 0.593 of the headline |
| 1 unit compounded 53 years | 83.2 vs 25.5, ratio 3.27 |
| headline minus corrected, per year | 0.0240 |
| if index volatility were unchanged: sd 0.150, implied riskless 0.0465 |
Show the full program
# Chapter 1. p11: a ledger of published art and collectible return estimates
# Model: no simulation. Published point estimates are laid out in one table
# and re-used in plain arithmetic (sources, not illustrative):
# Korteweg, Kraussl & Verwijmeren (2016, RFS): 32928 repeat-sold
# paintings, 1960-2013; uncorrected index 8.7% a year, corrected for
# selection 6.3%; Sharpe ratio 0.27 -> 0.11.
# Renneboog & Spaenjers (2013, Management Science): hedonic index,
# 1957-2007, real 3.97% a year in US dollars.
# Dimson & Spaenjers (2011, JFE): British stamps 1900-2008, nominal
# 7.0%, real 2.9%.
# Terminal multiples compound each rate over 53 years; they only translate
# a yearly rate into "what one unit became". The last line is algebra
# only: it asks what volatility and riskless rate the two Sharpe ratios
# would imply if the index volatility had not changed (an assumption).
import math
N_PAINTINGS = 32928
YEARS = 2013 - 1960
R_RS, R_SC = 0.087, 0.063
SH_RS, SH_SC = 0.27, 0.11
ROWS = [
("KKV 2016 repeat-sales, uncorrected", "1960-2013", "nominal", R_RS),
("KKV 2016 selection-corrected", "1960-2013", "nominal", R_SC),
("Renneboog-Spaenjers 2013 hedonic", "1957-2007", "real", 0.0397),
("Dimson-Spaenjers 2011 stamps", "1900-2008", "nominal", 0.070),
("Dimson-Spaenjers 2011 stamps", "1900-2008", "real", 0.029),
]
def main():
print("Published estimates (annual, as reported)")
print("{:<36}{:<11}{:<9}{:>7}".format("study", "period", "basis",
"return"))
for name, per, basis, r in ROWS:
print("{:<36}{:<11}{:<9}{:>7.4f}".format(name, per, basis, r))
drop = R_RS - R_SC
print("KKV sample: {} paintings, {} years".format(N_PAINTINGS, YEARS))
print("return drop: {:.3f} absolute, {:.3f} of the headline".format(
drop, drop / R_RS))
print("Sharpe drop: {:.2f} -> {:.2f}, {:.3f} of the headline".format(
SH_RS, SH_SC, (SH_RS - SH_SC) / SH_RS))
w_rs = math.exp(YEARS * math.log1p(R_RS))
w_sc = math.exp(YEARS * math.log1p(R_SC))
print("1 unit compounded {} years: {:.1f} vs {:.1f}, ratio {:.2f}".format(
YEARS, w_rs, w_sc, w_rs / w_sc))
gap = math.fsum([R_RS, -R_SC])
print("headline minus corrected, per year: {:.4f}".format(gap))
vol = gap / (SH_RS - SH_SC)
rf = R_RS - SH_RS * vol
print("if index volatility were unchanged: sd {:.3f}, implied riskless"
" {:.4f}".format(vol, rf))
if __name__ == "__main__":
main()Single source and citation
Fan, Y. (2026). Repeat-Sales Selection Bias in Collectibles, Forgery Priors and the Illiquidity Discount of Alternative Assets (Version 1.0.0), Chapter 1: After the Correction: Selection Bias in Repeat-Sales Samples. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-208 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026208,
author = {Fan, Yafu},
title = {Repeat-Sales Selection Bias in Collectibles, Forgery Priors and the Illiquidity Discount of Alternative Assets},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 2 · The Bailey-Muth-Nourse Regression: from Pairs of Sales to a Price Index
- Chapter 3 · Who Sells When: Resale Timing, Exit and Censoring
- Chapter 4 · Forgery Priors: Likelihood Ratios of Authentication Evidence and the Value of Information
- Chapter 5 · The Round Trip: Break-Even Years and the Chance of Loss on a Single Piece
- Chapter 6 · The Illiquidity Discount: a Perfect-Timing Bound and Sales under a Deadline
- Chapter 7 · Classics East and West: Mi Fu's History of Painting and Vasari's Lives
- Chapter 8 · Closing the Loop: an Allocation Cap, a Due-Diligence Checklist and a Falsification Plan