Nearly 40% of Profitable Japanese Companies Fail to Earn Their Cost of Equity

Nearly 40% of profitable companies posted an ROE below their Ke

Of the companies in our universe, 2,113 posted positive LTM net income attributable to the parent. Of those, 834 — 39.5% — had an ROE below their Ke. Put differently: nearly four in ten profitable companies are earning an accounting profit, but the return on their equity capital still falls short of what our model estimates shareholders require. Looking at the size of the gap, 314 companies (37.6%) sit within a 2-point spread — arguably within estimation-error range — so a meaningful share of the "39.5%" reflects close calls rather than wide misses. At the other end, 24 companies (2.9%) show a gap of 10 points or more. The headline figure treats a 0.1-point miss and a 40-point miss identically, so read it as a screen, not a precision instrument.

All companies listed have confirmed positive equity.

Nippon Sheet Glass shows the widest gap

Nippon Sheet Glass — #4 in our earlier Ke ranking — takes the top spot here for the widest Ke − ROE gap. Its Ke of 43.08% is driven largely by financial leverage: D/E of 12.4x. Its trailing ROE, by contrast, is 2.93% — a gap of 40.2 points. Thinly capitalized companies tend to show extreme readings on both metrics at once: a high D/E pushes Ke up, while that same thin equity base sits in the denominator of ROE, pulling ROE toward zero (or below) whenever earnings are modest. So a wide gap here isn't purely a story about weak profitability — it can be substantially a function of capital structure.

How we calculate Ke

We calculate Ke using CAPM (Ke = Rf + β × ERP). Rf is the 10-year JGB yield (2.80%); ERP is Damodaran's Japan equity risk premium (5.14%, updated annually). For beta, companies whose 2-year weekly-return OLS regression against TOPIX yields R² ≥ 0.30 use their own directly regressed beta; companies below that threshold instead use the Damodaran Japan sector-average unlevered beta, relevered by the company's own D/E (interest-bearing debt at book value ÷ equity market cap) and the statutory effective tax rate (30.62%) via the Hamada formula (βL = βU × {1 + (1 − T) × D/E}), assuming a debt beta of zero. Of the 834 companies in this comparison, 184 (22.1%) use their own directly regressed beta, 649 (77.8%) use a relevered sector beta, and the remaining 1 uses a sector-median fallback. Companies with wider gaps skew heavily toward the relevered-sector-beta method — all 24 companies with a gap of 10 points or more use it. Nippon Sheet Glass's Ke of 43.08% is likewise a mechanical output of the sector-average unlevered beta (0.82) relevered at D/E = 12.4x — not a figure drawn from the company's own trading history. At D/E levels this high, the assumptions underlying the Hamada formula — including a stable cost of debt — become harder to sustain theoretically, so figures like this should be read as a single-point estimate that applies extreme financial leverage to a sector-level risk read, not as a precise measurement.

We don't yet have a long enough history (our official beta and market-rate snapshots begin only in March 2026) to test how this single-point estimate relates to subsequent PBR or share-price moves over time. This ranking is a point-in-time screen, not a sensitivity analysis across alternative assumptions or a time-series validation of the underlying relationship.

A forward-looking requirement compared against a trailing-year result

Ke is a forward-looking estimate of what shareholders require; the ROE used here is a trailing twelve-month accounting result — the two sit on different time horizons. In theory, a sustained gap where ROE trails Ke puts downward pressure on PBR (the Gordon growth model: PBR = (ROE − g) / (Ke − g)), but that's an implication of the formula, not something this single-year cross-section itself demonstrates. A fuller read would weigh multi-year average ROE, forward ROE estimates, and the durability of earnings alongside this snapshot.

In March 2023, the Tokyo Stock Exchange asked listed companies to pursue "management that is conscious of cost of capital and stock price," flagging that roughly half of Prime Market companies had ROE below 8% and traded below 1x PBR. In April 2026, it updated that request with a focus on "appropriate allocation of management resources." The relationship between ROE and Ke sits squarely within what investors and regulators are watching.

Data assumptions

• Universe: 2,252 Japanese listed companies with market cap ≥ ¥10bn
• Excluded: banks, securities firms, insurers, and other financial institutions; REITs; companies with equity at or below zero
• Data date: August 9, 2026
• ROE definition: LTM net income attributable to the parent ÷ period-end equity (same definition used on beluuga.ai company pages)
• Ke calculation: see "How we calculate Ke" above
• No individual adjustments were made for companies that changed their fiscal year-end
• References: Tokyo Stock Exchange, "Request regarding Actions to Implement Management that is Conscious of Cost of Capital and Stock Price" (March 31, 2023); Japan Exchange Group, "Update on the Request Concerning 'Management Conscious of Cost of Capital and Stock Price'" (April 28, 2026)
• Source/calculation: beluuga.ai