The Discount That Hasn't Closed: Why Daiichi's 18% EV Growth Left P/EV at 0.53x

Daiichi Life Group told its investors on 13 February 2026 that group embedded value had risen 18% in thirteen months, to ¥9.65 trillion. The shares moved with it, up 22%. P/EV barely moved – 0.53x today against 0.51x a year ago, while European peers change hands at 0.8 to 1.0 times.

An earlier piece on this site argued that Japan’s listed lifers were structurally cheap, and that a four-year J-curve in spread income would re-rate Daiichi Life Group (8750), renamed from Dai-ichi Life Holdings on 1 April 2026, and T&D Holdings (8795). The case rested on disclosed sensitivities and one early data point from Fukoku Mutual. Daiichi’s December disclosure has now turned the hypothesis into a measured fact for one of the two listed names. The market has accepted the EV growth at face value. It has not closed the discount.

Two dates will calibrate the next step. T&D discloses its annual MCEV in mid-May, three weeks from now. Daiichi reports its audited fiscal-2025 EV in late June.

The composition is the part the market underweighted. Group adjusted net assets fell by ¥250 billion across the period, dragged down by a domestic bond unrealised loss that widened from ¥2.05 trillion to ¥3.42 trillion at Dai-ichi Life, the operating subsidiary. Group value of in-force rose by ¥1.73 trillion, to ¥8.11 trillion. The headline number is the sum; the substance is the second component.

Value of in-force rose because higher rates discount future liabilities more heavily. Long-dated insurance liabilities lose more present value than long-dated bond assets, given the duration mismatch on a Japanese life insurer’s book. J-ICS, the new economic-value solvency regime introduced in fiscal 2025, makes the asymmetry visible for the first time in three decades. The ¥13 trillion bond-loss number that dominated coverage in early 2026 was looking at the asset side only.

Four channels supported the move over the same thirteen months and all four pointed up: equities rose 70%, the 30-year yield rose 97 basis points, the yen weakened 6.7% and operations added another ¥750 billion of EV through new business and expected return on the existing book. Daiichi’s Group EV sensitivity to a 50bp rate rise is a fall of ¥130 billion against an ¥8.16 trillion base, under 2%. The bond losses are real; they are not the EV story.

T&D discloses MCEV once a year. The most recent figure is ¥3.95 trillion at end-March 2025. Applying the same channels and T&D’s disclosed sensitivities gives a pro-forma estimate against which the May number can be checked.

Pro-forma midpoint implies +22% growth – in the same range as Daiichi’s disclosed +18%, and verifiable against the actual MCEV figure in mid-May.

Daiichi’s P/EV was 0.51x at end-March 2025 – ¥4.17 trillion of market capitalisation against ¥8.16 trillion of EV. Now it is 0.53x against the December disclosure (¥5.11 trillion against ¥9.65 trillion). T&D was at 0.41x in March 2025 (¥1.63 trillion against ¥3.95 trillion); today, against pro-forma, 0.38 to 0.40x. The 0.51 to 0.53 creep absorbs the gap between share appreciation (+22%) and EV growth (+18%) almost exactly. The 30-point gap to European peers has not closed across thirteen months of equity rally, rate rise, yen depreciation and 18% EV growth.

If P/EV moves to 0.85x over four years and EV continues to compound at the disclosed run-rate, Daiichi delivers around 21% annualised from price-and-EV combined, plus a 3.7% prospective dividend yield (¥52 against the recent close): total return in the mid-twenties annualised. T&D, starting from 0.39x and compounding at its four-year average ROEV of 7.9%, reaches around 31% – plus 3.4% yield, low thirties total. The variance comes entirely from where the terminal multiple lands. At 0.7x rather than 0.85x, Daiichi annualises in the mid-teens and T&D in the mid-twenties.

The math is plain; the assumption that buyers eventually take the multiple to 0.85x is not. Daiichi reports adjusted ROE (a P&L-based measure of return on book equity, currently around 12%); T&D’s 7.9% is ROEV (an EV-growth-based measure). The two are conceptually different, used here as proxies for the same thing because that is what each company chooses to disclose.

Three reasons the discount may persist. Demographics: Japan’s new-policy market is shrinking, which over a long horizon constrains the EV-growth assumption. Cross-shareholdings: equity gains that have offset bond losses are partly the product of TSE-mandated unwinds, and the cushion thins as the programme runs down. Credibility: J-ICS is new, and investors who priced lifers on accounting earnings for thirty years may discount disclosed EV numbers as untested for several reporting cycles. None of this kills the case. They are reasons re-rating may stall at 0.7x rather than reach 0.85x.

External validation matters too. Listed Japanese banks moved from 0.5x P/B to above 1.0x between 2023 and 2025, but only after the Tokyo Stock Exchange demanded that companies trading below book either explain themselves or close the gap. Lifers do not yet have an equivalent push. M&A, a regulatory initiative or a major institutional adoption of J-ICS as the primary lens could each provide the trigger. None is forecastable; all are plausible. Without one, the discount may close part-way and stop.

Operating insurers are behaving as the J-curve thesis would predict. Of the ten largest Japanese lifers surveyed by Reuters on 27 April 2026, four plan to increase JGB holdings in fiscal 2026 – against none a year ago. Meiji Yasuda, the second-largest, has reversed its position from holding back purchases to buying by the trillion, according to Kitamura Kenichiro, its head of investment planning. Nippon Life rotated ¥3.9 trillion of low-coupon bonds in fiscal 2025 – the peak of its programme – and plans a smaller rotation this fiscal year, according to a separate Reuters report. The lifers’ own forecasters place the 30-year yield at year-end in a 3.50% to 3.90% range; none expect a return to January’s 3.88% high. Operating insurers are positioning for the spread arithmetic. Listed-share investors have not yet repriced for it.

Daiichi’s June disclosure starts from the December figure and adjusts for three more months of equity strength and a small further rate rise. Realistic landing: ¥9.5 to ¥10.5 trillion, putting P/EV around 0.50 to 0.55x if the share price stays where it is. T&D’s May disclosure has the pro-forma table above as its anchor; an audited figure of ¥4.7 to ¥4.95 trillion would put P/EV around 0.40x. Both are scheduled. Both will turn estimates into audited numbers.

Beyond that, the path depends on how investors choose to read the disclosures. The first move is mechanical: those who have not done the work see audited figures and adjust, taking P/EV to perhaps 0.55 to 0.60x. The second is structural: peer comparison enters mainstream sell-side research, the 2023-25 banking re-rating becomes part of the conversation, and fiscal-2027 results show spread expansion in reported earnings rather than only in EV. P/EV around 0.7x in this phase, twelve to eighteen months out. The third – full re-rating to 0.85x – depends on whether something analogous to the TSE’s PBR push reaches the lifers. The first move is small, the second plausible. The third is the trade.

The 1997-2001 period, when several Japanese life insurers failed and policyholders took losses, sits in the institutional memory of every Japanese investor in their late thirties or older. The ¥13 trillion bond-loss number revives that memory. The market is pricing the memory. Daiichi’s own February materials show Group EV up 18%, ESR strengthening to 213%, the positive interest margin expanding, equity sales accelerating to fund bond rotation. The substance has moved. The multiple has barely moved with it.

The next data points are on the calendar. The question is whether anyone is paying attention.

This piece is the author’s analysis of public information and not investment advice. The end-December 2025 Group EV figure for Daiichi Life Group is from the company’s Q3 FY2025 results call materials, dated 13 February 2026, page 34. T&D’s pro-forma MCEV is the author’s calculation from disclosed sensitivities and observed market moves; the audited figure is expected in mid-May 2026. Daiichi’s audited FY25 EV is expected in late June 2026.

— Gyokuro