The Trinity Study meets the Abgeltungsteuer
The 4 percent rule is the most famous rule of thumb in financial independence, and it is one of the most widely misunderstood. It promises an agreeable simplicity: save twenty-five times your annual spending, withdraw four percent each year, live off your wealth. Yet the rule was designed in another country, for another tax regime, and for a shorter time horizon. Anyone who imports it into Germany unchanged is planning on an assumption that was never tested here.
This article sets the rule in its place: where it comes from, what it actually says, and why a careful plan in Germany sits closer to three or three and a half percent. That is not a correction of an arithmetic error. It is the acknowledgement that three independent factors stand differently in Germany than they did in the original study.
The rule has two roots. The first is a 1994 paper by the US financial planner William Bengen, published in the Journal of Financial Planning. Using historical US data from 1926 onward, Bengen tested which initial withdrawal rate a portfolio of US stocks (the S&P 500) and intermediate-term US government bonds would have sustained over a minimum horizon of 30 years. His method is the often-forgotten core of the rule: you withdraw four percent in the first year and keep that euro amount constant in real terms thereafter, that is, adjusted for inflation, regardless of how the portfolio performs. Bengen himself was more cautious than the headline his work later became: for horizons of 50 years and more, he considered something closer to three percent to be “absolutely safe.”
The second root is the so-called Trinity Study of 1998, named after Trinity University in Texas and written by Cooley, Hubbard and Walz. It examined US data from 1926 to 1995 over horizons of 15, 20, 25 and 30 years and across various stock-bond mixes. The much-quoted finding: an equity-heavy portfolio survived a real withdrawal of four percent over 30 years almost always in the past, on the order of roughly 95 percent of the starting points examined.
From this comes the familiar inversion: anyone who wants to withdraw four percent needs twenty-five times their annual spending, because 25 is simply the reciprocal of four percent (1 divided by 0.04).
The 4 percent rule is not a law of nature. It is a historical US backtest with clear limits.
Two of these limits are decisive for German investors. First, the study models neither taxes nor fund costs: the four percent is a gross withdrawal, before tax and fees. Second, it is a look back at a particular period of the US markets, not a promise and not a globally valid rule. A past that would have worked almost every time is a valuable orientation, but it is no guarantee for the future.
The 4 percent rule is a US backtest: before tax, before fund costs, over 30 years. In Germany a more conservative plan of around 3.0 to 3.5 percent is in order, and not on account of the Abgeltungsteuer (Germany flat tax on investment income) alone. Three independent factors stack up: the tax burden on investment income, a typically longer horizon of 40 to 50 years rather than the tested 30, and the cost of health insurance before reaching pension age. Tax on its own does not pull the rate down to 3.5 percent.
It is worth looking at these three factors separately, because they are so often thrown into one pot.
The first factor is tax. The 4 percent rule calculates gross. In Germany, investment income is subject to the Abgeltungsteuer (the German flat tax on investment income). It amounts to 26.375 percent, made up of 25 percent Kapitalertragsteuer (capital gains tax, § 32d EStG) plus a 5.5 percent Solidaritätszuschlag (solidarity surcharge) on top. With Kirchensteuer (church tax) it comes out a little higher. What matters, however, is this: the tax does not fall on the entire withdrawal, but only on the gain contained within it. That is precisely what makes the effect smaller than the bare percentage would suggest, as the next section shows.
The second factor is the time horizon. The Trinity Study tests 30 years. Anyone who leaves working life in their mid-forties or earlier may be planning for 40 to 50 years. Over so long a span the historically sustainable rate falls, because more unfavourable market phases have to be endured. Analysis from the research around Kitces, Pfau and the blog Early Retirement Now points to something closer to 3.0 to 3.5 percent for such horizons.
The third factor is health insurance. In Germany it is often the largest hidden item for those who leave their careers early, in the years before the regular pension age, and it has nothing to do with tax. Anyone voluntarily insured under the gesetzliche Krankenversicherung (statutory health insurance) pays contributions on their entire income, investment income included. This item reduces what is genuinely left to live on, on top of the tax.
The honest conclusion is therefore this: it is not a single effect that pushes the German rule of thumb below the American one, but the interplay of three factors. Anyone who attributes the decline to the Abgeltungsteuer alone is oversimplifying.
The most common error in reasoning is the naive subtraction: withdraw four percent, deduct 26.375 percent tax from it, done. On a withdrawal of EUR 40,000 that would be EUR 10,550 in tax. This calculation overstates the real burden substantially, because it overlooks that only the gain portion of the units sold is taxed, never the returned capital that was originally invested (§ 20 Abs. 4 EStG).
A deterministic worked example makes this tangible. It is illustrative and not a forecast.
Worked example, tax burden of a withdrawal
Portfolio EUR 1,000,000, withdrawal 4% = EUR 40,000. Assumed embedded gain portion: 50% → raw gain EUR 20,000. Teilfreistellung for equity funds, 30% tax-free: EUR 20,000 × 0.70 = EUR 14,000. less Sparerpauschbetrag EUR 1,000 = EUR 13,000 to be taxed. Tax: EUR 13,000 × 26.375% = EUR 3,428.75. That is 8.57% of the withdrawal, a return reduction of roughly 0.34 percentage points. Net, around 3.66% remains instead of 4%.
Two mechanisms keep the burden below the naive estimate. First, the Teilfreistellung (partial exemption): for equity funds invested more than 50 percent in stocks, 30 percent of the income remains tax-free for private investors (§ 20 InvStG), so the effective rate on fund gains falls to 18.4625 percent. Second, the Sparerpauschbetrag (saver's lump-sum allowance) of EUR 1,000 for single filers, EUR 2,000 for jointly assessed couples (§ 20 Abs. 9 EStG). The order matters here: first the Teilfreistellung, then the allowance.
In the example, the naive subtraction (EUR 10,550) overstates the true burden (EUR 3,428.75) by about threefold. This factor is not constant, however. It depends on the embedded gain portion, and that portion rises over the years. For each securities account the FIFO principle applies without exception (§ 20 Abs. 4 S. 7 EStG): the oldest units are sold first. Over time those are the ones with the highest relative gain, so the taxed share of the withdrawal grows across the decades. Early in retirement the tax burden is therefore small; later it grows.
| USA (Trinity) | Germany (planning) | |
|---|---|---|
| Rule-of-thumb withdrawal rate | 4% | around 3.0–3.5% |
| Multiple required | 25× | around 29× to 33× |
| Tax on income | not captured in the model | Abgeltungsteuer 26.375%, on the gain only |
| Teilfreistellung for equity funds | not relevant | 30% tax-free (§ 20 InvStG) |
| Horizon tested | 30 years | often 40–50 years targeted |
| Health insurance | not captured in the model |
The table sets a historical US backtest figure against a German planning figure. The left column is a past result, the right a cautious forward plan, not a measured number.
Kernaussagen
Three to three and a half percent is to be understood as a planning anchor, calibrated against historical data and against the German particulars named here. It is expressly not a guaranteed value. Which point within the range makes sense depends on your own horizon: anyone planning for 30 years can lean towards three and a half percent; anyone planning at 40 or 45 for half a century, towards three. Expressed in multiples, that corresponds roughly to 29 to 33 times annual spending, with about 30 times as a serviceable midpoint. This number is an orientation for the order of magnitude of the goal, not a licence to withdraw that much blindly in any given year.
The calculator shows how a chosen withdrawal rate affects wealth over time. It is no substitute for individual planning; rather it makes the levers visible: wealth, rate and horizon. Anyone who lowers the rate from four to three and a half percent sees immediately how much additional staying power that single half percentage point creates. Assumptions about return and inflation remain exactly that, assumptions, not a prediction.
Two topics often decide success in the withdrawal phase more than the precise level of the starting rate.
The first is inflation. The original rule does not withdraw a fixed euro amount but a fixed real amount: the initial four percent is adjusted year by year to inflation. Anyone who overlooks this and withdraws a nominally constant amount appears to be planning more conservatively, but loses purchasing power in real terms over the years. An honest plan has to decide whether the withdrawal is to preserve purchasing power, and then factor inflation in.
The second is sequence risk, also called sequence-of-returns risk. It describes an effect that operates only in the withdrawal phase: not only the size of the returns counts, but also their order. Weak returns in the early years do lasting, disproportionate damage, because units sold during a downturn are missing when the market later recovers. Those very units can then no longer rise with it.
This explains why no fixed withdrawal rate can be entirely safe. An unfavourable start can endanger even a cautious rate, and a favourable start forgives even one that is somewhat too high. From this insight follow the practical strategies of the next section.
Since sequence risk chiefly affects the early years, the most effective approaches aim at not having to sell into a downturn in precisely those years.
A liquidity buffer. Anyone who holds around two to three years of spending in liquid, low-volatility funds can live from that buffer in a weak market year, instead of selling equity units at depressed prices. The buffer is no source of return but an insurance against the most unfavourable timing.
Flexible spending. Anyone who can lower the withdrawal in weak years, for instance by deferring larger, postponable outlays, relieves the portfolio exactly when it is most vulnerable. Flexibility is often worth more than a few tenths of a percentage point on the starting rate.
Dynamic guardrails. Rule-based approaches such as the dynamic guardrails of Guyton-Klinger allow a higher starting value and define guardrails of about plus or minus 20 percent around the path. In bad years the withdrawal is cut by around ten percent, and after a losing year the inflation adjustment is suspended. This improves the longevity of the portfolio, but it buys that improvement with a fluctuating income.
These three strategies can be combined. Their shared core: a rigid rate is more fragile than a plan that is allowed to respond to the order of returns. The rule of thumb sets the order of magnitude of the goal; the strategy decides whether the plan also survives a poor start.
The worked examples in this article are illustrative and not a forecast. Rates, allowances and the Basiszins (base rate used to compute the Vorabpauschale) reflect the position as of 2026 and may change. The article explains general principles and the applicable tax law; it is not individual investment or tax advice. The sustainable withdrawal rate depends on personal circumstances, and sequence risk means that no fixed rate can be guaranteed.
Before you set a withdrawal rate
In Investboard you see what a withdrawal really means in tax and costs, and how a given rate carries your wealth across the years.
Work through your withdrawal in Investboard →| a separate item before pension age |