Chapter 5

Figures converted from Swiss francs at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

What the Price Implies

At $257 Sonova trades on roughly 20x forward normalized earnings, about half the multiple it carried at its 2021 peak — yet normalized continuing earnings per share, in the company's own francs, are essentially where they were then. The de-rating was almost entirely a re-rating. From here the price implies only mid-single-digit perpetual growth, roughly what the hearing-care market grows anyway, and the two-year outcome is governed less by earnings than by the exit multiple — the same cyclical-versus-structural question the rest of this report tests.

What you pay today

Sources: share price and market cap from market data (23 July 2026), ~59.6m shares; net debt $1,243.5m [1] and normalized EBITA $1,014.7m and operating free cash flow $649.2m per FY2025/26 Annual Report [2]; forward P/E on consensus FY2026/27 EPS; dividend $5.88 [3].

The market capitalization is about $15.3bn on roughly 59.6m shares at $256.6. Net debt of $1,243.5m — around 1.2x normalized EBITA, and falling — lifts enterprise value to about $16.5bn [4]. None of these headline multiples is demanding for a high-margin, cash-generative medical-device leader; none is cheap. The work is in deciding whether ~20x is the right number.

The de-rating was a re-rating

Sonova has lost roughly 48% of its value in francs since the November 2021 peak near $435. Almost none of that came from a fall in underlying earnings. Normalized earnings per share from continuing operations were $11.43 in FY2021/22 and $13.03 in FY2025/26; in the company's own francs they were flat across the whole de-rating, and the rise in dollars reflects the stronger franc translating into more dollars [5]. The multiple did the moving: from roughly 38x normalized earnings at the peak to about 20x today.

Source: normalized continuing EPS and reported EPS including discontinued operations, five years to FY2025/26, per the Annual Report Five Year Key Figures [6] and Sonova Group key figures [7], converted to dollars at each year's rate.

The chart also carries the honest counter to the "flat earnings" read. The lighter bars — reported EPS including discontinued operations — fell from $11.27 to $9.04, a 20% decline in dollars (31% in francs), because a strengthening Swiss franc translated local-currency progress away and the Sennheiser consumer business is being exited at a loss [8]. A franc-based owner's headline earnings genuinely stepped down; the underlying earning power of the retained business did not. Both statements are true, and the gap between them is the franc and the divested division — the subject of the Financials and Estimates chapter.

Source: peak/current prices from market data (converted to dollars); normalized continuing EPS per FY2025/26 Annual Report [9]. Peak-era EPS uses FY2021/22.

Decomposed, the ~48% franc price fall (~41% in dollars) is roughly a 47% compression in the multiple against a ~1% change in franc-denominated normalized earnings; in dollars the same multiple compression is partly offset by a ~14% rise in translated earnings as the franc strengthened. That framing matters for what comes next: a stock that de-rated on multiple, not on broken earnings, is worth a durable-quality assessment, because the re-rating either corrected an over-priced peak or over-corrected a still-good business. The Moat and OTC Threat chapter argues the underlying franchise held; the price says the market is not yet paying for that.

What the price implies

The multiple can be read back as an expectation. A simple perpetual-growth frame on trailing operating free cash flow of $649.2m [10] against a $15.3bn market value implies the growth rate the price is discounting, given a required return. At an 8.5% cost of equity — reasonable for a low-beta Swiss blue-chip — the price implies about 4% perpetual free-cash-flow growth.

Source: derived from FY2025/26 operating free cash flow and market value; Gordon-growth solution, illustrative [11].

The hearing-care market grows an estimated 4–6% a year on demographics and adoption — the tailwind established in the Moat and OTC Threat chapter. So at $257 the market is paying for Sonova to grow roughly in line with, or a touch below, its own industry: no premium for the share gains it has been posting since the Infinio/Sphere launch, no credit for the normalized-margin recovery off the FY2024 trough, and no value for the buyback optionality of a lapsed programme and a de-levering balance sheet. This is a low bar for a company guiding to 5–8% sales growth and 7–10% core-EBIT growth for FY2026/27, and to $7.4bn of revenue by FY2030/31 [12] [13]. It is not, however, an obviously wrong bar: if the franc stays strong and reported earnings keep lagging local-currency ones, a franc-based owner may indeed compound at roughly the market rate.

Triangulating the number

No single lens settles it, so it is worth seeing several at once. The methods below fit a stable, cash-generative compounder with modest leverage — earnings multiples and cash yield, cross-checked against the enterprise value.

Sources: normalized EBITA $1,014.7m, EPS and dividend per share per FY2025/26 Annual Report [14] [15]; forward EPS from consensus estimates.

The lenses agree on a range rather than a point. On earnings the stock is neither cheap nor expensive against its own quality: 20x forward for a business that held gross margins above 73%, converts about 95% of profit to operating cash, and earns a 19.0% return on capital employed [16]. The 4.2% operating free-cash-flow yield, rising toward 5% as net debt falls, and a 2.2% dividend that has grown through the de-rating (with one FY2024 cut to $4.76 before rebuilding to the record $5.88) give the holder a mid-single-digit cash return before any growth [17]. With roughly 8% consensus normalized-EPS growth and the multiple unchanged, the expected total return is around 10% a year — respectable, not spectacular, and entirely dependent on the multiple not falling further.

A clean peer cross-check is not available in this corpus: reliable enterprise-value multiples for Demant, Amplifon and GN Store Nord are not present in the run's data, and one staged peer snapshot resolved to the wrong company. What can be said from primary filings is that Sonova's closest pure-play peer, Demant, has been re-rated down alongside falling margins (EBIT before special items 20.9% to 17.2% over three years), so Sonova's ~20x sits above a struggling comparable — a premium the share-gain and margin-recovery evidence has to keep earning.

A two-year range

The valuation is best expressed as a range whose width is set by the exit multiple. The table below anchors on fiscal 2028 normalized EPS — roughly two years out — and pairs each earnings path with the multiple that plausibly accompanies it, plus about $12 of dividends collected along the way.

Source: scenario construction on consensus and normalized earnings; consensus FY2028E EPS ~$14.5. Illustrative, derived from reported financials [18].

Source: derived from the scenario table above; current price $256.6 for reference.

The range brackets the published analyst spread closely — a mean target near $275–293, a high around $368 and a low near $208–228 — which is a reasonable sanity check on the construction rather than a coincidence. The bear case (about -15% over two years) needs the franc to stay strong, competitors to ship credible deep-neural-network platforms that erase the Infinio/Sphere lead, and US managed-care to keep pressing average selling prices — and the multiple to fall to a Demant-like 16x. The bull case (about +48%) needs the share gains to persist, the normalized margin to recover toward 24%, buybacks to resume at today's depressed prices, and a partial reclaim of the historical premium to 24x.

The leverage sits in the multiple. From bear to bull, the earnings assumption moves about 19% ($12.87 to $15.32); the exit multiple moves 50% (16x to 24x). The multiple dominates the outcome — and the multiple is the market's verdict on whether the FY2022–24 step-down was cyclical or permanent. The valuation, in other words, does not resolve the report's central question; it prices it, and hands most of the two-year return to whichever way that question is answered.

What would change the read

The read here is deliberately a range, not a target, because the inputs that would tighten it are genuinely unresolved. Three are checkable against future filings:

Whether the local-currency growth premium survives the next competitive cycle — the durability question the moat chapter could not close — is the single largest swing factor, because it decides the exit multiple. A sustained slowdown of Wholesale growth toward the 4–6% market rate, visible in the half-year releases, would validate the bear multiple; continued outperformance would support a re-rating.

Whether management restarts buybacks at today's prices is the cleanest near-term signal. The $1.8bn 2022–2025 programme lapsed roughly 70% unused after the front-loaded, buy-high repurchases documented in the Owners and Stewards chapter; a new authorization deployed near $257 would be the first evidence of counter-cyclical capital allocation and would directly lift per-share value.

Whether the franc's translation drag reverses would flatter the reported earnings a franc-based owner actually banks. The FY2025/26 result absorbed a $276.4m hit to reported sales from currency [19]; a stable-to-weaker franc would let local-currency progress reach the income statement the market capitalizes.

Two limitations bound this chapter. The run holds only about six months of daily prices, so the long-run multiple history beyond the sourced 2021 peak is not reconstructable here, and the peer enterprise-value comparison is thin for the reasons noted. Neither changes the core arithmetic: at $257, the price pays for market-rate growth and leaves the recovery, the share gains, and the buyback optionality as unpriced upside — against a real risk that the franc and the competition keep the reported numbers from ever showing it.