The stock option grant that taxes your Tokyo engineer at employment-income rates — 7 steps to structure equity in Japan

Equity is the one lever a Tokyo startup has against a megabank or a foreign tech company that can simply pay more in cash. Used well, it is decisive. Used carelessly, it produces a specific and memorable failure: an engineer exercises their options, discovers they owe a substantial tax bill on a gain they cannot access, and concludes that the equity was never real.

That outcome is not bad luck. It is the default result of granting ordinary options without structuring them to meet Japan’s qualification requirements. The alternative — a tax-qualified stock option, 税制適格ストックオプション — defers tax to the point of sale and taxes the gain at a flat capital gains rate rather than progressive employment-income rates.

The difference is large enough to change whether an offer is competitive. Here is the seven-step sequence, with the usual caveat: this is an operational guide written from hiring practice, not tax advice. The thresholds move, and your plan should be reviewed by a Japanese tax adviser before any grant is made.

Step 1: Understand the two possible outcomes before you design anything

Everything else follows from this distinction, so it is worth stating precisely.

Under a non-qualified option, the taxable event is exercise. The spread between the exercise price and the market value of the shares at that moment is treated as employment income and taxed at progressive rates. Combined national and local rates at the top of the scale in Japan are among the higher ones internationally. A subsequent sale then produces a separate capital gain or loss on any further movement.

Under a tax-qualified option, there is no tax at exercise. Taxation occurs only when the shares are sold, and the entire gain from the exercise price to the sale price is taxed as a capital gain at a flat separate rate — substantially below the top employment-income rate.

Two differences follow, and the second matters more than the first.

The rate difference is significant but comprehensible. The timing difference is what causes real damage: under the non-qualified route, the tax is due when the engineer exercises, which is precisely when they have paid money out and received nothing liquid in return. If the company is private with no secondary market, they must fund the bill from savings while holding shares they cannot sell.

Step 2: Check that the grantee is eligible

Qualification is not only about the terms of the option — it is also about who receives it. The regime applies to defined categories of recipients: directors and employees of the issuing company and, subject to conditions, certain related companies.

The rules were widened in recent years to bring in certain external contributors — specialists and advisers who are not employees — which is genuinely useful for startups that depend on a small number of senior outside contributors. But the categories are defined, not open-ended.

There is also a shareholding constraint: recipients who are substantial shareholders, or closely related to them, are generally excluded. This catches founder relatives more often than people expect.

The practical failure mode here is a well-intentioned grant to someone outside the eligible population — a contractor who has been essential for two years, an adviser who introduced the lead investor. The grant does not fail visibly at the time. It fails at exercise, which is the worst possible moment to discover it.

Step 3: Set the exercise price at or above fair market value at grant

The exercise price must be at least equal to the fair market value of the shares at the time of the grant. A discounted strike price disqualifies the option from the regime.

For a listed company this is mechanical. For an unlisted company it requires a defensible valuation, and that is where startups run into difficulty: the temptation to set a low strike price so the option feels more valuable to the recipient is strong, and it is exactly the wrong move. A low strike price makes the option look attractive and simultaneously destroys the tax treatment that made it attractive.

Japan has published guidance in recent years intended to make valuation of unlisted startup shares more workable for this purpose, which has reduced — though not eliminated — the practical friction. Get the valuation documented at grant, not reconstructed later.

Step 4: Respect the exercise window

Qualified options must be exercised inside a defined period measured from the date of the grant resolution. Exercise before the window opens or after it closes takes the option outside the regime.

The window has been extended for certain unlisted companies in recent reforms, recognising that the original period was too short for companies whose path to liquidity is long. This is a meaningful improvement for deep-tech and hardware startups in Japan, where the interval between grant and any realistic exit routinely exceeded the old limit.

Two operational consequences. First, vesting schedules must be designed against the window, not independently of it — a four-year vest that pushes final tranches past the closing date creates a cliff nobody intended. Second, departing employees need explicit guidance: an engineer who leaves and forgets about their options until after the window closes has lost the treatment, and often the options.

Step 5: Track the annual exercise limit

There is a ceiling on the total exercise value an individual may realise in a calendar year under the qualified regime. Exercising above it breaks qualification for the excess.

This ceiling has been raised in recent reforms, with higher limits available to qualifying companies, in direct response to the complaint that the original figure was too low to be meaningful for senior hires at companies that had grown substantially.

Because the figure has moved more than once, do not rely on a number quoted in an article written in an earlier year — including this one. Confirm the limit applicable to your grant with a current adviser.

The operational point that does not change: somebody has to track it. If several engineers exercise in the same year following a liquidity event, the limit is per person and the monitoring is per person. Companies that treat the option ledger as a spreadsheet nobody owns discover the problem after the exercises have been processed.

Step 6: Arrange compliant custody and non-transferability

Two structural conditions complete the picture.

The options must be non-transferable. A plan that permits assignment falls out of the regime. This is normally handled in the plan document and rarely causes trouble, provided nobody drafts a well-meaning exception for estate planning.

Shares acquired on exercise must be held under an approved custody arrangement. Historically this meant deposit with a designated financial institution, which was a genuine obstacle for unlisted companies whose shares no securities firm wanted to hold. Reforms have made issuer-side custody workable for unlisted companies, removing what was one of the most common practical reasons that startup plans failed to qualify.

If your plan was drafted before those changes and never revisited, this is the clause most likely to be out of date. It is worth a specific look rather than a general assumption that the plan is fine.

Step 7: Document the grant and explain it in language engineers can use

The final step is the one most often skipped, and it determines whether the equity does any work at all.

On documentation: the grant resolution, the plan terms, the valuation basis and the individual grant notice all need to exist and to be consistent with each other. Inconsistencies between the board resolution and the individual notice are a standard finding in pre-listing due diligence.

On communication: an equity package the recipient cannot value has no retention effect. Most engineers — including very strong ones — have no framework for evaluating an option grant, and a grant they do not understand is treated as worth zero in their decision-making.

What we have seen work is a one-page summary in plain language covering: how many options, what percentage of the current fully-diluted share count that represents, the exercise price, the vesting schedule, the exercise window with concrete dates, what happens on departure, and a worked example at two or three illustrative company valuations.

That last element does the heavy lifting. A grant expressed only as a number of options is abstract. The same grant shown against a realistic outcome becomes a reason to stay. For international candidates weighing Tokyo against other markets, this comparison is decisive — our colleagues at HireDeveloper.sg cover the equivalent equity and reporting mechanics in Singapore, and HireDeveloper.ae covers the UAE, where the absence of personal income tax changes the calculation entirely.

The three mistakes that cost the most

• Granting ordinary options by default. Usually not a decision at all — a template was reused, and nobody asked whether it met the qualification conditions. The cost lands years later, on the engineer, at exercise.
• Setting a strike price below fair market value to make the grant look generous. Destroys the treatment that made the grant genuinely generous. The intent is good and the effect is the opposite.
• Never explaining the plan. An unexplained grant has no retention value, which means the dilution was accepted for nothing. This is the cheapest of the three to fix and the most frequently neglected.

If you are structuring an engineering compensation package in Japan more broadly, our guide to shakai hoken, payroll and equity for engineering hires covers the social insurance and payroll side that sits alongside this.

Frequently asked questions