Four Key Squares
← Portfolio · position sizing lives on the allocation strategy pageIn a chess endgame, a few squares decide the whole game. We score every company on four. Everything is scored from filings — never from hope. Each company's four squares →
- How fast is the industry growing, and why should that continue — a government budget, an aging population, a rule that forces spending?
- How much of next year's revenue is already promised under contracts customers renew?
- What is the one substitution that could make this product unnecessary? That is the break we watch for.
- What lets it win customers from competitors — a dataset, a network of users, a cost of switching away?
- When it raised prices, did customers stay? That is pricing power you can check in filings.
- This is the square we most often leave open — the profiles keep working on it.
- Tokyo Stock Exchange listing standards push companies priced below book value to explain themselves — and the usual fix is paying out more.
- Governance investors vote against managers who hoard. An activist on the register raises the cost of doing nothing.
- A founder approaching retirement often wants dividend income, and tax rules make dividends attractive to a controlling holder.
- We score the disclosures as they land: buybacks, dividend-policy changes, statements on the cost of capital, new 5% holders.
- Every name here passed the screen: priced at 2–12 years of operating profit, capital earning 15% or more, real cash generation.
- The cheaper it gets, the closer the cash-pile floor — see below.
- The only valuation exit: the multiple re-rates out of the screen. We publish no target prices.
The valuation floor
A company that earns more than it pays out adds the difference to its balance sheet every year. Run that arithmetic on several of these holdings and it ends somewhere unusual: net cash equal to the entire market capitalization. That is Benjamin Graham’s net-net — reached not by the price falling, but by the cash rising. A price like that carries an assumption: the cash will never reach shareholders. The four pictures below show why that assumption is unstable — whatever management does with the payout, the outcome converges on the shareholder. One holding as the worked example:
The same clock, every holding
| Name | EV/EBIT | ROCE | Net cash / mcap | Payout | Yield at 100% payout | Years to cash = mcap |
|---|---|---|---|---|---|---|
| 5843 Nippon Insure | 4.0x | 29.0% | 40.3% | 10.2% | 10.4% | 6.4 |
| 2477 Temairazu | 4.5x | 23.8% | 46.6% | 21.3% | 8.1% | 8.4 |
| 4168 Yappli | 8.1x | 22.6% | 11.2% | 20.6% | 10.2% | 11.0 |
| 5570 Jenoba | 6.7x | 22.9% | 36.8% | 17.1% | 6.6% | 11.6 |
| 2301 Gakujo | 5.4x | 15.4% | 31.8% | 50.4% | 9.7% | 14.1 |
| 4194 Visional | 10.8x | 30.8% | 22.6% | 0.0% | 5.0% | 15.5 |
| 3984 User Local | 8.6x | 22.7% | 33.0% | 25.0% | 5.4% | 16.7 |
| 7378 ASIRO | 5.8x | 42.3% | 14.7% | 47.0% | 9.6% | 16.7 |
| 4058 Toyokumo | 9.2x | 38.5% | 17.8% | 22.6% | 6.1% | 17.4 |
| 6037 Rakumachi | 10.5x | 29.1% | 3.7% | 19.5% | 6.7% | 17.9 |
| 4377 ONE CAREER | 9.0x | 38.4% | 19.8% | 29.4% | 6.3% | 18.0 |
| 4493 Cyber Security Cloud | 12.2x | 23.9% | 18.8% | 7.1% | 4.8% | 18.2 |
| 4776 Cybozu | 9.2x | 56.0% | 10.8% | 31.1% | 6.9% | 18.8 |
| 4475 HENNGE | 13.6x | 42.6% | 17.8% | 11.7% | 4.7% | 19.9 |
| 6947 Zuken | 9.1x | 13.1% | 37.3% | 55.5% | 5.9% | 23.9 |
| 2353 Nippon Parking Development | 8.5x | 18.8% | 5.3% | 49.3% | 7.5% | 25.1 |
| 6532 BayCurrent | 12.7x | 41.7% | 7.4% | 41.0% | 5.4% | 28.9 |
| 4071 Plus Alpha Consulting | 13.1x | 42.6% | 13.6% | 40.8% | 4.6% | 31.8 |
How to read the table: “Years to cash = mcap” counts the years until today’s net cash plus retained profit equals today’s market capitalization, holding this year’s profit and dividend flat. Buybacks are excluded — they would only shorten the count. “Yield at 100% payout” is profit divided by market capitalization: the dividend yield at today’s price if every yen of profit were paid out. Above 6% — the level where Japanese small-cap yields have historically found buyers — a full payout alone would argue for a higher share price. Figures restate as prices and filings change.