Advantest vs Tokyo Electron: 9.1% slide leaves a 31% valuation premium

Advantest fell 9.1% and Tokyo Electron dropped 7.5% in Thursday morning trading in Tokyo on November 5. At nearly simultaneous Yahoo Finance observations around 9:59 a.m. JST, the semiconductor equipment stocks traded at ¥31,420 and ¥8,940, respectively, against Wednesday's closes of ¥34,570 and ¥9,663.

The selloff leaves a revealing difference: Advantest costs about 30.6 times its company's forecast earnings for the year ending March 2027, versus 23.4 times for Tokyo Electron. That is a roughly 31% premium. Yet Tokyo Electron needs substantially faster second-half operating-profit growth to deliver its outlook. The choice is between two different execution risks, rather than two interchangeable ways to buy the AI investment cycle.

What the Tokyo selloff changed

The Advantest observation was timestamped 9:58:49 a.m. JST; the Tokyo Electron observation followed three seconds later. These are fixed, dated morning-session snapshots used at publication, in Japanese yen. They were already about 15 minutes old when retrieved; the provider's formal delay specification is unavailable. They are not closing prices or a claim of live quotations.

Those moves use the previous regular-session close as the denominator: ¥31,420 divided by ¥34,570, minus one, gives a 9.11% decline for Advantest. The equivalent Tokyo Electron calculation gives a 7.48% decline. Merely returning to those closes would require gains of about 10.0% and 8.1% from the observed prices, before transaction costs. A large fall does not make the recovery mathematically symmetrical.

The market observations establish the selling, but do not identify one definitive cause. Nor do they establish that either company's earnings forecast has changed. Advantest's guidance used here was released October 28; Tokyo Electron's was released October 30. This comparison applies those dated forecasts to a new market development. It does not present the financial releases as Thursday announcements.

A 31% premium, with two important qualifications

Advantest's October 28 forecast puts annual basic earnings per share at ¥1,027.08. Tokyo Electron's October 30 financial release forecasts ¥381.69. Dividing the observed share prices by those figures produces 30.59 and 23.42 times earnings. Dividing the first multiple by the second, then subtracting one, produces the 30.6% premium.

Both forecasts cover April 1, 2026, through March 31, 2027. Advantest calls that fiscal 2026; Tokyo Electron calls it fiscal 2027. Comparing the fiscal labels without the actual dates would put an unnecessary obstacle in front of an otherwise matched-period calculation. These are company-guidance multiples, rather than analyst-consensus or next-twelve-month estimates.

The accounting is less perfectly matched. Advantest reports under IFRS and Tokyo Electron under Japanese GAAP. Their company EPS figures are useful published benchmarks, but do not become a standardized measure of recurring operating earnings merely because they cover the same period. This is an equity-price comparison, not an enterprise-value calculation adjusted for every balance-sheet asset and liability.

Tokyo Electron also completed a five-for-one share split on October 1. Its forecast EPS is explicitly adjusted for that split, matching the post-split price. Comparing ¥8,940 directly with Advantest's ¥31,420 would say nothing useful about which business is cheaper: nominal share prices depend on how many shares divide the ownership.

Advantest has the smaller profit step, but a margin retreat

Advantest's forecast calls for ¥1,863 billion of annual revenue and ¥967 billion of operating profit. Against first-half revenue of ¥845.2 billion and operating profit of ¥456.8 billion, the company projects ¥1,017.8 billion and ¥510.2 billion, respectively, in the second half. The implied sequential operating-profit increase is 11.7% using the company's rounded figures.

That increase comes with a lower operating margin: 50.1% in the second half versus 54.0% in the first. Revenue growth can therefore exceed profit growth without signaling a missed forecast. An investor who extrapolates the first-half margin through March would be assuming something stronger than management's published outlook.

The spending assumptions help explain the hurdle. Advantest budgets ¥69.1 billion of research and development in the second half, against ¥44.9 billion in the first, and ¥37 billion of capital expenditure versus ¥23 billion. Research expenditure affects operating earnings; capital spending affects cash and future depreciation under the applicable accounting. They should not be subtracted from profit in the same way.

Advantest attributes tester demand to increasing volumes and complexity of high-performance AI semiconductors, including advanced packaging. Its half-year results also describe efforts to expand production capacity across the supply chain. Those are reasons its earnings might remain resilient, but meeting customer demand still depends on equipment delivery and component availability. A favorable end market is not the same as completed revenue.

Tokyo Electron's discount comes with a steeper delivery requirement

Tokyo Electron forecasts ¥3,700 billion in annual sales and ¥1,169 billion in operating income. Its first-half release reports ¥1,667.455 billion and ¥506.776 billion. Subtracting those actual results leaves ¥2,032.545 billion of sales and ¥662.224 billion of operating income to be delivered from October through March.

That means second-half sales need to rise about 21.9% from the first half, while operating income needs to rise 30.7%. The implied operating margin is approximately 32.6%, versus 30.4% in the first half. Unlike Advantest, Tokyo Electron needs both a larger revenue base and a higher margin to reach its full-year operating target.

This is why the lower multiple cannot settle the comparison. Tokyo Electron offers a less expensive claim on the earnings it forecasts, but a larger portion of the annual operating result still lies ahead. Shipment timing, customer acceptance, product mix and spending discipline can all affect whether revenue growth produces the required profit growth. The calculations describe the hurdle; they do not establish that management will miss it.

TECHi's earlier Tokyo Electron analysis examined that profit requirement during the November 2 rally. Thursday's new issue is relative valuation after a sharp decline in both stocks. Tokyo Electron could close part of the valuation gap by delivering its forecast while Advantest loses some of its premium. It could also remain cheaper because investors demand compensation for the larger execution step.

Earnings quality and cash complicate the apparent winner

Advantest's first-half net income was ¥368.1 billion, but its release identifies approximately ¥57 billion of financial income, primarily valuation gains on investment securities. That income is outside operating profit. It is not equivalent to cash earned by selling more testers, and its future contribution depends in part on investment values.

The premium comparison therefore deserves extra caution: a headline EPS multiple can look more attractive when non-operating gains boost the earnings denominator. The release does not provide a directly comparable, tax-adjusted recurring EPS measure for both companies. Removing the entire ¥57 billion from net income would mix pre-tax and after-tax quantities, so this comparison does not manufacture an adjusted multiple.

At September 30, Advantest reported ¥401.785 billion of cash and cash equivalents, ¥89.486 billion of bonds and ¥17.760 billion of current and non-current lease liabilities. Tokyo Electron reported ¥578.601 billion of cash and cash equivalents; its balance sheet does not separately list a bank-borrowing or bond line. That absence should not be expanded into a claim that every form of financing obligation is zero.

The cash figures also are not an argument for subtracting one company's total cash from its share price. They are historical company-wide balances, while the prices are per-share November observations. A proper cash-adjusted valuation would require matching outstanding shares, debt, leases and other relevant claims on the same basis. Here, the figures show liquidity context and the limits of using a simple P/E ranking.

What would change the forecast comparison?

If both companies achieve their current annual EPS forecasts and the observed valuation multiples hold, neither assumption by itself produces a gain from the snapshot prices: those prices already equal forecast EPS multiplied by the calculated multiples. A bullish forecast needs an additional argument for higher earnings, a higher valuation multiple, or both.

A useful stress test reverses the question. If annual EPS were 10% below each company's guidance while prices stayed unchanged, Advantest's multiple would rise to about 34.0 times and Tokyo Electron's to 26.0 times. Those are hypothetical denominator changes, calculated by dividing each current multiple by 0.9. They are neither revised forecasts nor estimates of the probability of a miss.

The strongest case for paying Advantest's premium is that tester demand tied to chip complexity proves more durable and its smaller second-half profit step leaves less execution distance. The strongest objection is that its forecast already embeds exceptionally high operating margins, while investment gains and exchange rates complicate the earnings base. A smaller hurdle is not an inexpensive hurdle.

Tokyo Electron's case improves if its shipments and margins deliver the required second-half acceleration without a corresponding valuation increase. The countercase is that lower valuation is compensation for a more demanding forecast, rather than a mispricing. Evidence of weaker profit conversion would strengthen that objection. Better operating delivery at both companies could support both stocks; their relative performance still depends on what investors pay for those earnings.

What to watch

The next dated market checkpoints are Tokyo's 11:30 a.m. JST morning close and 3:30 p.m. cash close on November 5, under the exchange's trading schedule. Those observations will show whether the early declines persist; they will not, on their own, validate or invalidate the March 2027 earnings forecasts. For the business comparison, watch subsequent company guidance and reported second-half revenue, operating margins and non-operating income. The decisive question is whether each company earns the profit its own valuation now asks investors to assume.