What dividends mean for your retirement


The romantic vision: living off passive dividend income in retirement. The reality: it works — but differently than most people think. Here's the honest math, with Swiss examples, the yield-on-cost effect, and the moment when the 4% rule is actually better.
There are few ideas in the private investor universe as romantic as "living off passive dividend income". The picture is clear: you have a portfolio of quality stocks that send you money to your account quarter after quarter, you never have to sell a share, the capital stays intact, and you can even pass it on to your children. Sounds like a perpetual money machine — and in certain cases it really is. But only if you truly understand the math behind it.
This article explains how dividend income actually works in retirement, why the yield-on-cost effect is so powerful, and in which situations a classical 4% rule strategy still performs better. With Swiss data and examples — Roche, Novartis, Nestlé — and an honest assessment of the pros and cons.
Before we get into the math, an important clarification: a dividend is not free money. If a stock is worth CHF 100 and the company distributes a dividend of CHF 3, on the ex-dividend day the stock is mechanically worth CHF 97 — the company's value has fallen by the amount distributed. The dividend is therefore a conversion of "value in the stock" to "cash in the account", not an extra return.
So why pursue dividends as a strategy at all? Three reasons:
1. Psychology. Dividends don't force you to sell shares. For many retirees, exactly this is the crucial point: they feel safer living "off income" rather than "off capital". Mathematically it's a small difference; psychologically it's a big one.
2. Company discipline. Companies that have raised their dividend annually for 20+ years are typically more disciplined in capital management. They have to know they have long-term predictable cash flows, otherwise they can't maintain the dividend. That's a quality signal — and it systematically selects more robust companies.
3. Yield-on-cost. And this point is the real game-changer — and we dedicate the entire next chapter to it.
The yield-on-cost (YoC) is the dividend yield calculated against your original purchase price, not the current price. And that's the crucial difference between "dividend yield" in the newspaper and what really matters to you as a long-term investor.
An example. You buy CHF 100'000 of shares in a Swiss quality company at CHF 100 per share in 2026. The current dividend is CHF 3 per share — a dividend yield of 3% today. The company raises the dividend by an average of 4% per year over the following 20 years.
Year 0 (purchase): 1'000 shares @ CHF 100, dividend CHF 3 = CHF 3'000/year → YoC 3.0%
Year 5: dividend grown to CHF 3.65 = CHF 3'650/year → YoC 3.65%
Year 10: dividend grown to CHF 4.44 = CHF 4'440/year → YoC 4.44%
Year 15: dividend grown to CHF 5.40 = CHF 5'400/year → YoC 5.40%
Year 20: dividend grown to CHF 6.58 = CHF 6'580/year → YoC 6.58%
Your original investment stays unchanged (CHF 100'000), but your annual dividend income has more than doubled. And that's without additional contributions, without selling a single share.
The effect becomes even more dramatic over longer periods. With a 30-year holding period and 5% dividend growth, the YoC lands at almost 13% — meaning CHF 100'000 purchase price generates CHF 13'000 in annual income. At 6% growth over 30 years, it's over CHF 17'000 per year. This effect is so strong it requires a completely different mentality than "dividend yield today".
The condition: This only works with quality companies that can actually raise their dividend over decades. A company that cuts or suspends the dividend even once (which happens frequently with weaker names) immediately destroys the YoC effect. Selecting the right names is therefore the entire strategy.
Switzerland has some of the best examples of long-term dividend continuity in the world. Three of the most well-known:
At the 2026 AGM, Roche paid out a dividend of CHF 9.80 per share — the 39th annual increase in a row. That makes Roche one of the world's most stable dividend aristocrats. Anyone who bought in 1987 and held the stock enjoys a YoC of several hundred percent on the original purchase price today.
Novartis raised the dividend in 2026 to CHF 3.70, up 5.7% from the prior year. That's the 29th annual increase in a row — a remarkable track record that also places Novartis in the dividend aristocrat category.
Nestlé had long been considered the most reliable dividend compounder in Switzerland — year after year of increases since around 2007. But in 2025 the dividend was raised from only CHF 3.05 to CHF 3.10 — an increase of just 1.64% and well below inflation. That's a warning signal: even the most robust dividend stocks aren't immune to periods of weaker performance. Anyone relying exclusively on dividends for income needs to plan for this volatility.
A long dividend streak is an indicator of quality, but not a guarantee. Even companies with 20+ years of increases can enter a phase where growth stagnates or the dividend is cut. Diversification across 20–30 quality companies is therefore mandatory, not optional.
Here it gets important — because Swiss tax logic treats dividends differently from capital gains. While capital gains on sales are tax-free for private individuals (as long as you're classified as a "private investor" under ESTV Circular 36), dividends are fully taxable as income. This is a structural disadvantage of the dividend strategy compared to a pure accumulation strategy in free assets.
On Swiss dividends, 35% withholding tax is automatically deducted at source. When Roche pays out CHF 9.80, only CHF 6.37 directly arrives in your account. You can reclaim the CHF 3.43 per share through your annual tax return — but only if you cleanly declare your securities holdings. Anyone who forgets or ignores this gives away these CHF 3.43 per share.
Annual dividend gross: CHF 15'000
Withholding tax 35%: −CHF 5'250
Net inflow during the year: CHF 9'750
Reclaim via tax return: +CHF 5'250
Effectively you get the full CHF 15'000 — but with a delay of up to 12 months.
Important: Dividends are declared as taxable income and subject to your marginal rate. At CHF 130'000 income in Zurich, that's approximately 32% on every dividend franc.
Dividends distributed within Pillar 3a are tax-free. This is one of the underrated advantages of invested Pillar 3a: you can hold dividend stocks or dividend funds without any income tax on distributions. Compared to free assets, this is a structural advantage of ~30% per dividend franc.
The classical alternative to the dividend strategy is the 4% rule (also called Safe Withdrawal Rate, SWR): you hold a diversified stock/bond portfolio and withdraw 4% of the original portfolio value per year (inflation-adjusted). Research shows that such a portfolio historically lasts at least 30 years in almost all cases — without ever running empty.
Which strategy is better? It depends:
| Aspect | Dividend strategy | 4% rule |
|---|---|---|
| Income certainty year 1 | ~2–4% of portfolio | 4% guaranteed |
| Income growth | Strong (YoC effect) | Inflation adjustment |
| Capital preservation | High (stocks held) | Moderate (gradual drawdown) |
| Diversification needed | Concentrated on div payers | Broadly diversified possible |
| Taxes (Switzerland) | Dividends fully taxable | Capital gains tax-free |
| Psychology | Very reassuring | Requires discipline |
The honest answer: for most Swiss retirees, a combination of both approaches is best. A broadly diversified quality portfolio in free assets (with reinvested dividends until retirement), then in retirement a mix of running dividend cash flows and selective selling for the 4% rule. The psychology of dividends + the flexibility of SWR = the best of both worlds.
Before you convert your entire strategy to "dividend retirement", know the three most common traps:
A stock with 7% dividend yield sounds attractive — but it's often a warning signal, not an opportunity. High dividend yields often arise when the share price has fallen because the market anticipates the dividend will soon be cut. Classical examples: real estate REITs before rate hikes, energy companies before restructurings, or telecom giants with shrinking market shares. Rule of thumb: when a single stock's dividend yield is significantly above the market average, something is usually off.
Anyone selecting specifically for dividend stocks often ends up in a very concentrated portfolio: many banks, many energy companies, many consumer staples. That's not inherently bad, but it reduces diversification compared to a broad market index. Anyone who additionally focuses on "Swiss dividends only" ends up with a portfolio of 5–10 names with strong correlation to the Swiss economy and Swiss franc.
A dividend that doesn't at least keep up with inflation loses real purchasing power. At 1.5% Swiss inflation and 1.5% dividend growth, you're real-terms flat. That's OK, but it's not "income that grows with you". Anyone who wants YoC effects needs dividend growth significantly above inflation — and that's often not the case with classical high-yield names.
It depends on your desired annual payout. At a conservative average dividend yield of 3%, you need roughly 33 times your desired annual income: CHF 33'000 annual income → CHF 1'100'000 portfolio. If you additionally count on YoC growth, the initial amount can be a bit smaller — but calculate conservatively.
US companies often have longer uninterrupted dividend streaks (Coca-Cola, Procter & Gamble, Johnson & Johnson have 50+ years) and typically pay quarterly instead of annually, which smooths cash flow. But: US dividends are more tax-complex for Swiss private investors (tax treaty, US withholding tax), and the USD/CHF currency effect can significantly influence the dividend stream over the years.
For most investors, a dividend ETF is the more robust choice — immediately diversified, automatic rebalancing, low cost. Individual stocks only make sense if you're willing to analyse and monitor 20–30 companies over years. Both can work, but individual stocks are more work.
In severe crashes, some companies cut dividends (2008/09 and 2020 were examples). Quality companies with strong balance sheets weather even these phases and maintain or even raise the dividend. The question of which names can do this is the core of stock selection.
During the accumulation phase: reinvest, always. That's the compound effect that works miracles over 30 years. In retirement: spend or partially reinvest, as needed. Many brokers offer DRIP (Dividend Reinvestment Plan) programs automatically.
Very well. Dividends within Pillar 3a are tax-free, and 3a is anyway the most tax-optimal vehicle in Switzerland. Anyone pursuing a dividend strategy should seriously consider placing most of their dividend names in the 3a — especially if they're Swiss stocks with withholding tax.
Yes — it's actually the logical combination. Quality companies with strong cash flows are typically also good dividend payers. arvy's quality approach selects ~30 companies based on fundamental criteria, and many of them are also dividend continuers. The strategies don't contradict each other — they complement each other.
Further reading & calculators
The dividend strategy works, but it's no panacea. Used correctly — with quality companies, a long horizon, diversification, and a tax-optimal setup (especially in Pillar 3a) — you end up at the end of your working life with robust income that grows over time and leaves the capital largely intact. Used incorrectly — with yield traps, concentration, and without regard to taxes — you lose money compared to a simple, diversified equity strategy. The difference lies in the execution, not the concept.
Written by Thierry Borgeat, Co-Founder of arvy, and reviewed by Patrick Rissi, CFA and Florian Jauch, CFA. Dividend data on Roche (39 years of increases, CHF 9.80 at 2026 AGM), Novartis (29 years, CHF 3.70 at 2026 AGM, +5.7%) and Nestlé (CHF 3.10, +1.64%) based on official 2026 AGM announcements. Yield-on-cost calculations: original purchase price constant, dividend growth as geometric average. Last updated April 2026.
Disclaimer: This article is for general educational purposes and does not constitute personal investment or tax advice. Companies mentioned are examples, not buy recommendations. Historical dividend payments are no guarantee of future distributions. arvy is a FINMA-supervised asset manager with a CISA licence. Imprint & Legal Notice.