Repository logo
Research Outputs
Projects
People
Statistics
  1. Home
  2. HSG CRIS
  3. HSG Publications
  4. Nixed in Nippon: Index Deletions, the TOPIX Reform, and the Upside of Getting Dumped in Japan
Details

Nixed in Nippon: Index Deletions, the TOPIX Reform, and the Upside of Getting Dumped in Japan

Series
SSRN Working Paper Series; 7518118
Type
working paper
Date Issued
2026
Author(s)
Andreas Grünbichler  
Abstract
Japan's TOPIX reform turns index deletion from a surprise into pre-announced policy. In October 2022 the exchange designated 493 small constituents for a phase-down in ten quarterly steps ending in January 2025. A second stage begins with the review announced on 7 October 2026. This paper uses the reform, together with thirty years of Nikkei 225 deletions, to test the thesis of Arnott and Henslee (2024) that stocks discretionarily deleted from major indexes outperform for years after their removal. The Nikkei 225 evidence replicates the US anatomy. Among 87 tradable deletions since 1995, 65 have complete five-year windows. These 65 lose 26 percent against the market in the year before removal at the mean and 31 percent at the median. They leave at 0.78 times the universe median price-to-book, against 2.14 times for the stocks that replace them. A yen invested at the effective date grows to 1.97 times its TOPIX-relative value within five years, with a median of 1.73. One episode, the April 2000 overhaul, accounts for four fifths of the summed excess. Against a control matched on size and book-to-market with point-in-time fundamentals the premium is smaller: a mean excess ratio of 1.20, a median of 1.00 and a ratio of mean wealth of 1.19, about 3.5 percent per year. Most of the deletion premium in Japan is the premium of what deleted stocks have become. The TOPIX evidence is descriptive. The designated issues trailed size-matched non-designated peers by 8 to 12 points during the phase-down. About a quarter of that shortfall accrued inside the step windows, the rest between the steps. After final removal the cohort rebounded by 3 points in one month and has led by 5 to 8 points since. The control is twice the cohort's size, so the shortfall is measured, not identified. For the second stage the paper screened 605 likely transitional issues on the actual review window and registered five falsifiable predictions before the list was published. The list of 7 October 2026 names 683 transitional issues. The screen held 585 of them, and the misses sit just above its full-capitalization boundary, where the exchange ranks on free float. A first addendum records cohort, control and reference prices before any outcome is known.
Language
English
Keywords
index deletions
index reconstitution
price pressure
TOPIX
Nikkei 225
passive investing
pre-registration
HSG Classification
contribution to scientific community
Refereed
No
Official URL
https://ssrn.com/abstract=7518118
URL
https://alexandria.unisg.ch/handle/20.500.14171/143553
Subject(s)

finance

Division(s)

SoF - School of Finan...

Additional Information
JEL classification: G11, G12, G14, G23
File(s)
Thumbnail Image

open.access

Name

Nixed_in_Nippon_SSRN_Oct2026_v2.pdf

Type

Main Article

Size

534.46 KB

Format

Adobe PDF

Checksum (MD5)

635c537c697447045e245c91083a809f

Support
HSG researchers can find instructions here for adding or importing publications (DOI, ORCID). Please send questions to alexandria@unisg.ch

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science

  • Accessibility settings
  • Privacy policy
  • End User Agreement
  • Send Feedback
Repository logo COAR Notify