The Unwinding Wave: Parent-Subsidiary Listings 2020-2025

The 45-firm cohort

The framework's B-05 indicator scores listed-subsidiary configurations. The 45-firm cohort represents firms with disclosed minority-protection language at parent ownership ≥50% — the structural condition the METI 2019 guideline targeted.

Sector distribution:^[2]

The 45 firms are diffuse — top sector (Banking) has only 7 firms. Banking subsidiaries reflect the post-1990s Megabank consolidation where regional and specialty banks remained sub-listed under FG holding structures. IT, services, and retail are mostly tech-parent-and-brand-sub patterns. Real estate is the developer-parent / sub-trust REIT pattern.

The Poison Apple over-representation

The 45-firm listed-sub cohort registers Poison Apple at 18% — a 4–5× over-representation relative to the universe-wide 4% baseline.^[2] This is the centerpiece finding for Note 3.

Archetype distribution within the cohort:

The mechanism is structural. The Poison Apple archetype captures firms with high T-axis disclosure — yuho quality, audit opinion, J-SOX attestation, comply-or-explain depth on the CG Code — paired with B-axis or R-axis weakness underneath. Listed subsidiaries with majority-controlling parents satisfy the high-T condition almost by construction. Dual disclosure (parent and sub) creates redundant scrutiny on the formal layer; the parent's audit firm typically also audits the sub; Prime listing carries the same disclosure obligations as standalone Prime firms.

But the underlying axes carry the parent's footprint. The board of a listed subsidiary cannot be fully independent because the parent typically nominates a majority of directors. Related-party transactions with the parent are routine and dwarf transactions with unrelated parties. Capital allocation decisions reflect the parent's strategic priorities, not minority shareholders'. The B-axis and R-axis indicators read these conditions and pull the underlying axis scores down even as T-axis stays high.

The 8 Poison Apple firms in the listed-sub cohort sit at this exact structural junction. The framework's reading is: high formal disclosure, weak substantive governance, archetype label that flags the gap.

The grade distribution echoes the same finding:^[2]

Grade C share is 38% — substantially above the universe-wide 15%. Listed-sub firms cluster in the watchlist tier, again reflecting the structural T-axis-high / B-axis-or-R-axis-weak signature that pulls grade down without triggering KS.

The size signature

The 45-firm cohort is concentrated in mid-cap and below:^[2]

Only 2 of 45 listed-subs (4%) sit at TOPIX Large70; none reach Core30. The remaining 43 firms (96%) are mid-cap or smaller. The pattern is unambiguous: large-cap parent firms have mostly already absorbed (or never had) listed-subsidiary structures; the remaining listed-sub firms are mid-cap downward.

This reflects the secular trajectory. Cross-listed firms at the Toyota / Hitachi / Recruit / Sony Financial scale have been consolidated in the past decade or are in active consolidation now. The remaining 45-firm Prime cohort represents the smaller-cap residual that the unwinding wave has not yet reached — or, in some cases, has not been targeted by it because the parent's strategic value of a listed sub is weighed against minority cost differently at smaller scale.

The Toyota Group restructure

The single largest parent-child listing-style unwind currently in motion is the Toyota Group restructure announced on April 26, 2025.^[3] The architecture: Toyota Fudosan announced a tender offer for Toyota Industries (sec code 6201), with the explicit goal of taking Toyota Industries private and progressively unwinding the Toyota Motor / Toyota Industries / wider Toyota Group cross-shareholding architecture.

A nuance matters here. Toyota Industries does not sit in the 45-firm B-05 cohort — its parent relationship is bilateral cross-shareholding rather than majority-controlled subsidiary. The framework's production reading places Toyota Industries at Grade B, Chameleon archetype with the B-weak sub-tag, and B-01 at the near-residual band (10–20% of net assets). The reading is consistent with a firm carrying meaningful keiretsu cross-shareholding but not above the KS-1 threshold.

What makes Toyota Industries the anchor case for Note 3 is not its own framework reading but the structural movement it represents. Toyota Group's bilateral cross-shareholdings — Toyota Motor (7203), Toyota Industries (6201), Denso (6902), Aisin (7259), Toyota Tsusho (8015), JTEKT (6473), Toyota Boshoku (3116), and others — collectively constitute the largest corporate cross-shareholding cluster in Japanese listed equities. The April 2025 announcement signals that the Toyota Group has begun restructuring this architecture in earnest. Two Large-Shareholding Report (大量保有報告書) filings during the v02 window (June 9 and June 25, 2025) by Toyota Industries on Toyota Motor are the public-record early signature of the trajectory.^[4]

The framework reads Toyota Industries today as a near-residual Chameleon B-weak. This is where the cross-sectional snapshot meets the time-sensitive trajectory — and why Note 3 is necessarily a 2024–2026 window piece rather than an evergreen analysis.

Other large-cap unwinds in the v02 window

The Toyota Group is the largest, but it is not the only major unwind in motion. Public-record events during the v02 window (March 2025 → April 2026) include:

NTT Data Group (sec code 9613) was taken fully private by NTT through a tender offer running May 9 to June 19, 2025 at a price of ¥4,000 per share.^[5] NTT, which had previously held approximately 58% of NTT Data, acquired the remaining shares; the Extraordinary General Shareholders Meeting on August 29, 2025 approved the share consolidation, and NTT Data delisted from the TSE Prime Market on September 26, 2025. (A separate 2023-05 transition had restructured NTT Data into a holding-company form, NTT Data Group Corporation, but did not change the parent-subsidiary listing structure — that is a distinct event from the 2025 take-private.) The case represents the cleanest structural path: parent acquires 100%, listed-sub delists, dual-listing structure terminates.

Sony Financial Group completed a partial spin-off effective October 1, 2025.^[6] This is the reverse pattern — Sony Group separated SFGI into independent listing rather than absorbing it. Sony distributed approximately 83.6% of SFGI shares to Sony shareholders as dividends in kind and retains approximately 16.4% post-spin; SFGI listed on TSE Prime under sec code 8729. The framework reads spin-off as eliminating the parent-sub configuration through separation rather than consolidation. Both directions retire the listed-sub structure.

Hitachi Construction Machinery ownership was restructured in 2022 (announced January 14, 2022, closed June 2022): Hitachi sold 26% of HCM to a 50/50 JV between ITOCHU and Japan Industrial Partners (HCJI Holdings G.K.), reducing Hitachi's stake from 51.5% to 25.4% and converting HCM from consolidated subsidiary to equity-method affiliate.^[7] Hitachi Astemo was restructured in late 2023: JIC Capital took a 20% stake, with Hitachi's stake declining from 66.6% to 40% and Honda's rising from 33.4% to 40% (announced March 2023, closed Q3 FY2024).^[8] In December 2025, a follow-up restructuring shifted Astemo further: Hitachi sold an additional 21% to Honda, moving the structure to Honda 61% / Hitachi 19% / JIC 20%, with Astemo expected to become a Honda consolidated subsidiary in Q1 FY2027 subject to regulatory approvals.^[8] Hitachi has been progressively unwinding its listed-sub portfolio for over a decade.

Multiple sogo-shosha (Itochu, Mitsui) consolidations and JR group sub-trust restructurings have moved through 2025 and 2026. The aggregate effect: the listed-sub population has been actively contracting throughout the v02 window. The 45-firm production reading captures the cross-section at FY2025; the subsequent year is unlikely to leave the cohort unchanged.

The trajectory: METI 2019 to present

External public statistics document the secular decline of parent-child listing in Japan.^[9] Per Jefferies Financial Group estimates cited in international media, Japan had approximately 285 parent-child listings in 2020, declining to approximately 212 by 2025 — a roughly 25% reduction over five years. Nikkei Research has separately documented the long-term downward trajectory and noted that the composition of the top parent-firms has shifted: classic conglomerates such as Hitachi have effectively eliminated listed subsidiaries, while a smaller set of newer groups (GMO Internet, RIZAP) has actively built them up.^[10]

The METI 2019 guideline arrived in the middle of an already-declining trajectory and accelerated it. The roughly 212-firm 2025 figure spans all TSE tiers (Prime + Standard + Growth); the framework's Prime-only production cohort of 45 represents the fraction of that population currently listed at the most senior tier and meeting the framework's B-05 disclosed-protection criterion.

The decline has not been uniform. The largest unwinds have come from the top-cap firms — Toyota / Hitachi / Sony / Panasonic / Recruit / NTT — where the strategic costs of dual listing began to outweigh the historical benefits. The 45-firm Prime residual reflects what has not yet been touched: smaller-cap firms where the parent's strategic value of the listed sub is judged differently.

The 2023 METI Guidelines for Corporate Takeovers, published August 31, 2023, added a complementary pressure.^[11] The new guidelines on board duties under unsolicited acquisition scenarios raised the cost of maintaining anti-takeover defense plans (買収防衛策), and listed-sub configurations could function as informal defense in some cases. Combined with the 2019 group-governance guideline and the 2019 Fair M&A Guidelines (which addressed conflict-of-interest transactions including controlling-shareholder acquisitions), the regulatory pressure has flowed toward both the listed-sub and the keiretsu defensive cross-shareholding architecture.^[12]

Remaining backlog

Public-record candidates for upcoming unwinds — not framework predictions, simply firms whose parent-sub configurations are publicly disclosed and structurally vulnerable to consolidation:

The Toyota Group residual extends beyond Toyota Industries to multiple major affiliates: Denso (6902), Aisin (7259), Toyota Tsusho (8015), JTEKT (6473), Toyota Boshoku (3116). All carry bilateral Toyota Group cross-shareholdings that the April 2025 announcement implies will be progressively unwound, though the precise timing and structure of each affiliate's transition is not announced.

The Honda Group's restructuring of Hitachi Astemo, described above, is already in motion. Honda Lock was a wholly-owned (non-listed) Honda subsidiary sold to MinebeaMitsumi in January 2023 and renamed Minebea AccessSolutions — that transaction sits outside the listed-sub category but is consistent with the broader Honda portfolio rationalization.

The KDDI Group (sec code 9433, parent) includes regional Standard-listed subsidiaries such as Okinawa Cellular (sec code 9436), the structurally-distinct regional carrier serving Okinawa Prefecture under KDDI majority ownership.

The SoftBank Group structure includes SoftBank Corp (9434), Prime-listed with SoftBank Group holding majority — an unusual dual-Prime-listed parent / sub configuration at scale.

The Toshiba Group, post-Toshiba's own delisting in 2023, has Toshiba Tec (6588) as a residual structure.

The Mitsui Real Estate group has several sub-trust REITs that have been consolidating progressively. Various sogo-shosha groups carry similar REIT and trading-firm sub structures.

Each of these is publicly announced or publicly observable. The framework does not forecast which will move next; it reads the cross-sectional snapshot at the FY2025 yuho window. Notes published in 2027 against the FY2026 window will likely show a smaller listed-sub cohort.

Minority protection: disclosure presence vs substance

The framework's B-05 indicator scores listed-sub firms with disclosed protection at parent ≥50%. All 45 firms in this band satisfy the disclosure-presence test — they have explicit minority-protection language in their CGSO and yuho. What the framework does not currently measure directly is the substantive quality of that protection.

The METI 2019 group-governance guideline and the 2019 Fair M&A guideline together established the standard:^[1][12] disclose fairness opinion at TOB / wholly-owned conversion (完全子会社化) / squeeze-out events, give independent committees real authority, ensure minority shareholder representation in major decisions. Substantive compliance varies. Some firms produce detailed fairness opinions with independent investment-bank evaluations; others produce template language with minimal substantive content. Public TOB premium statistics show 30–50% premiums typical at squeeze-out events but with wide variance — and the variance correlates loosely with the substance of the protection language disclosed.

The framework defers boilerplate-vs-substantive classification to v1.2 NLP scope. The current production reading captures the cohort that satisfies disclosure presence; the substantive split awaits the parser upgrade.

What the unwinding wave implies

Three readings.

First, the parent-subsidiary unwinding wave is real and measurable. The METI 2019 guideline arrived in the middle of a declining trajectory and accelerated it. The 45-firm Prime residual is what remains after a decade of structural pressure has already retired the largest-scale listed-sub configurations. Forward attention should focus on the 45-firm cohort specifically, the Toyota Group restructure, and the residual flagged in the backlog.

Second, the listed-sub configuration carries archetype risk that runs higher than the universe baseline. The 18% Poison Apple share — 4–5× the universe-wide 4% baseline — reflects the structural T-axis-high / B-axis-or-R-axis-weak signature inherent to majority-controlled subsidiaries. An analyst valuing a listed-sub firm should weight the parent-sub structural factor explicitly, not assume that the Prime listing standardizes governance quality across firms.

Third, the framework's reading of the unwinding wave is cross-sectional, not predictive. The framework does not forecast which firm will be acquired or spun off next. It reads the configuration at the yuho window and assigns archetype based on the observable structure. The forward-looking question — which of the 45 will move within the next 24 months — is a separate analytical task that requires firm-specific reading of public announcements, parent strategic priorities, and acquisition-related signaling. The framework provides the static map; the moving-parts assessment is downstream.

The parent-child listing unwinding is, in the broader frame, one piece of the cross-shareholding compression that Note 2 traced. Cross-shareholding aggregate has fallen substantially over the past decade — by ACGA's measure, the simple-average policy-held (政策保有) ratio for TOPIX 500 firms moved from 13.5% to 8.4% of net asset value between FY2015 and FY2023. Listed-sub structures have followed the same downward arc, falling from 285 to 212 firms across all TSE tiers between 2020 and 2025. Both trajectories reflect the same broad governance reform pressure, and both carry the same kind of residual: the firms that have not responded sit in distinct sectoral and structural pockets.

The cross-shareholding architecture has been compressed. The listed-sub architecture has been compressed. What remains is the residual that the secular reform has not reached — and that is what the framework's cross-sectional reading captures most accurately.

Apex G-Score™ Japan Foundation Series. Production cohort: TSE Prime, 1,576 issuers (1,556 rated; 20 NR). The "FY2025" label denotes a rolling-window cohort — each firm's most recent yuho filed within 2025-03-11 to 2026-04-15. Fiscal year coverage by 決算月 (fiscal year-end month): 3月 (March) ~76.5% (reporting period 2024-04 to 2025-03); 12月 (December) ~13.7% (calendar year 2025); 9月 (September) ~5.1% (2024-10 to 2025-09); 6月 (June) ~4.6% (2024-07 to 2025-06). The December-end sub-cohort shares the calendar-year reporting window of the Korea cohort. Framework: Apex G-Score v2 (T 0.30 / B 0.30 / R 0.40); archetype classifier: unified-v1. Sample firm-level scores in this article are limited to the IP-guardrail-v2.0 disclosed set (Toyota Motor; Samsung Electronics and Reliance Industries appear in cross-market comparisons only). All other firm-level scores remain unpublished. Multi-year indicators where cited are noted explicitly with source vintages. This article is research output, not investment advice; readers should consult their own advisors before making investment decisions.