1651 — Precision Tsugami (China) Corporation Limited

Precision Tsugami (China) Corporation Limited — 1651.HK

Deep-value analysis · 2026-09-18 · reference price … (day’s close, …)

Reporting currency CNY (RMB), trading currency HKD. Rate used throughout: 0.8538 RMB per … (CNYHKD/HKDCNY, 2026-09-18). Fiscal year ends March 31; “FY2026” = the 2025-04-01 – 2026-03-31 exercise, reported 2026-07-23. Semi-annual reporter (HKEX Main Board) — there is no quarterly data, neither in filings nor in yfinance; every “quarterly” series in this report is SEMI-ANNUAL.

Primary sources: 2025-2026 Annual Report (HKEXnews, 2026-07-23, 215 pages), 2023-2024 Annual Report (2024-07-17), 2025 Interim Report (2025-12-12, half-year ended 2025-09-30), plus the data pack data-pack-1651-20260918.md and the research briefing research-1651-20260918.md.

Executive summary (1 page: thesis, estimated value, verdict)

Precision Tsugami (China) is the Chinese manufacturer of precision CNC lathes for the Japanese Tsugami group (TYO: 6101), which holds 65.89% of the shares (247,530,000 shares, Directors’ Report, p. 75). FY2026 was the best year in the company’s history: revenue RMB 5,184.1 million (…), net profit RMB 1,094.4 million (…), gross margin 34.9% (record) and net margin 21.1% — both the highest of the five published years. The balance sheet is practically debt-free: RMB 1,486.2 million cash and deposits, against RMB 3.0 million of lease obligations; debt/equity 0.08%. ROE 32.4%, ROIC (after-tax EBIT on invested capital, i.e. equity minus net cash) 50.5%. The company bought back 9,878,000 shares (2.6% of the capital) for …lion and paid dividends of RMB 369.0 million; total shareholder return was 66% of net profit.

The thesis, in one sentence: the business is better than the price. At … the stock trades at 14.1× a cyclical company’s peak profit and 4.27× book value, after climbing from … (Dec. 2022) — 8.6 times in less than four years, of which earnings-per-share growth explains roughly a doubling, and the rest was multiple re-rating (from ~3.3× to ~14×). The valuation models don’t converge: the base DCF, on assumptions I consider defensible (normalized owner earnings RMB 820 million, stage-1 growth of 8%, discount …, terminal 3%), gives an intrinsic value of …, i.e. MOS …; EPV Greenwald (no growth) gives …, the Graham Number (…) gives …, and the bull scenario …. The Monte Carlo simulation over 20,000 scenarios on the same model gives median MOS ** …**, with an undervaluation probability of …% and a probability of margin of safety above 30% of only …%. In other words: at today’s price, the buyer pays almost exactly the central intrinsic value and is betting on the optimistic end of the distribution.

The FCF bridge has one wrinkle, but it matters: the FY2026 reported FCF (CFO 1,064.6 − capex 53.9 = RMB 1,010.6 million, yield 6.5% on market cap) is inflated by capex of only RMB 53.9 million — below the RMB 62.2 million of depreciation — and by an RMB 50.8 million release of inventory that won’t recur. Budgeted FY2027 capex is RMB 200 million, and the two new plants in Pinghu (34 mu of land bought in March 2026) cost RMB 150 million and add 3,000 units/year of capacity by the end of 2027. In addition, every RMB of new revenue consumes working capital: notes receivable have reached 35.1% of revenue (from 14.8% in FY2023). Normalized owner earnings — profit + depreciation − normalized capex − the structural investment in working capital — are RMB ~820 million, not 1,010.6. That’s the figure I put into the DCF and the Monte Carlo.

Verdict: INTERESTING, but not at this price. The recalculated GBL score is …% (… weighted points out of 60; Graham 6.0 / Buffett 6.0 / Lynch 7.5; F-Score 6/9, cyclical F-adj 4/9), up from 60.8% MONITOR in the tracker (2026-09-06) — the quality of the business is real and had been undervalued by the mechanical score. But criterion B20 (margin of safety ≥20%) fails outright, and G2, G3, and G10 likewise. The zone where this business deserves buying, on my assumptions, is below … (P/E ~10 on FY2026 profit, positive median MOS); below … it would be a real-size position. The main risk is not bankruptcy — impossible with this balance sheet — it is the fact that FY2026 is a cycle peak AND a multiple peak at the same time, and the company’s own history shows revenue can fall 23% in a single year (FY2024).