The bridge problem: how to fund early retirement before pension age

Most early retirement plans have a gap in the middle. The saver builds a pot in tax-advantaged accounts, runs a 4% withdrawal calculation against the total, and treats the project as complete. But the tax wrappers that make the pot grow efficiently are the same wrappers that lock the money away until 55, 57, 59½, or later. Retiring at 50 and waiting seven years to access your own savings is not a plan. It is a problem in search of one.

This is the bridge problem: you need income from the date you stop working to the date you can touch your pension without penalty. How you build that bridge depends almost entirely on where you live, and the mechanics differ enough between the UK, France, Germany, and the US that a strategy that works in one country can be illegal or ruinously expensive in another.

What follows is a walk through the specific instruments available in each market in 2026, the order to use them in, and the traps to avoid.

What the bridge actually has to do

Before picking instruments, it helps to be clear about what the bridge has to deliver.

The bridge has to provide enough after-tax income to cover your expenses from your last working day until the earliest moment you can access pension-wrapped money without penalty. In the UK that gap is typically 5 to 20 years. In the US it is typically 5 to 15. In Germany it is typically 5 to 17. In France it can be anything from zero (because the PER has an early-access route for residence purchase) to 15 years.

The bridge also has to do this without forcing you to liquidate pension assets you have not yet reached, without triggering penalties, and without pushing you into tax brackets that erase the point of the exercise. And it has to survive the market environment you happen to retire into, which you cannot predict.

These constraints narrow the instrument list considerably.

The UK bridge: ISA to SIPP

The UK is the cleanest of the four systems because of one structural feature: the ISA. You can withdraw from an ISA at any age, in any amount, for any reason, with zero tax and zero penalty.

That single property makes the UK bridge straightforward. You fill your ISA alongside your SIPP during your working years, and in retirement you live off the ISA until you reach the minimum pension age. For 2026, minimum pension age is 55. It rises to 57 in April 2028, so anyone retiring from 2028 onwards will have a longer bridge than the current generation.

The mechanics are simple. If you retire at 50 with £400,000 in an ISA and £700,000 in a SIPP, you draw from the ISA for seven years (at 57) or five years (under current rules, at 55). A £40,000 annual spend covered by ISA withdrawals generates zero taxable income during the bridge, which is why some people choose to crystallise small SIPP tranches during the same period to use up their personal allowance. Withdrawing just enough SIPP to fill the £12,570 personal allowance each bridge year is tax-free income that reduces the total SIPP balance facing 40% withdrawals later.

The trap to avoid is the SIPP minimum age change. If you are 52 in 2026, you cannot rely on accessing your SIPP at 55 in 2029, because the minimum age rises to 57 in April 2028. Plan your bridge length against 57, not 55, if you will turn 55 on or after 6 April 2028.

The other UK consideration is the Cash ISA cap falling to £12,000 for under-65s in April 2027. This does not change the Stocks and Shares ISA allowance (still £20,000 overall), but it does constrain how much of an emergency buffer you can hold in cash inside the wrapper. Most bridges are equity-heavy anyway, so this matters less than it sounds.

The German bridge: there is no wrapper

Germany does not have the UK's ISA or the US's Roth. The closest thing to a personal tax-advantaged vehicle is the Sparerpauschbetrag, an annual investment income allowance of €1,000 per person. That is not a bridge. That is a rounding error.

Practically, German early retirement planning runs almost entirely through taxable brokerage accounts. The saver holds equity ETFs in a German broker, accepts the annual Vorabpauschale (a pre-tax on unrealised gains based on the federal basis rate, 3.20% for 2026), and draws down when needed. Capital gains are taxed at 26.375% (Abgeltungsteuer plus Soli), with a 30% partial exemption (Teilfreistellung) for equity funds holding at least 51% equities. That brings the effective tax on equity ETF gains to around 18.5% plus church tax if applicable.

The bridge strategy in Germany is therefore:

Build a large taxable brokerage position during working years. Accept that you are taxed more heavily on accumulation than a UK or US saver would be. When you retire, draw from the brokerage account until state pension age, which for someone born in 1972 is currently 67. There is no shortcut.

There are two wrinkles worth knowing about. First, the Grundfreibetrag (€12,348 for 2026) means the first chunk of any income is tax-free. If your only income in retirement is capital gains realisations, you can coordinate your drawdowns to stay under or close to this threshold for several years, particularly early in the bridge when your total expenses may be lower. Second, the Riester and Rürup systems do exist but are not useful bridges because their payouts are pension-like and start at statutory ages, not earlier.

For a German saver retiring at 50 with state pension at 67, the bridge is 17 years funded entirely from taxable investments. This is why German FIRE plans typically require a larger pot for the same after-tax retirement income: the bridge years take a heavier tax load than the wrapper-protected years that come after.

The French bridge: PER exceptions plus assurance-vie

France is the most structurally flexible of the four for people who can use its escape hatches, and the most rigid for people who cannot.

The French PER (Plan d'Épargne Retraite) is nominally locked until retirement age, but article L224-4 of the Monetary and Financial Code lists six cases of early release. Five of them are "accidents de la vie": invalidity, death of a spouse or PACS partner, expiration of unemployment benefits, over-indebtedness, and judicial liquidation of a non-salaried activity. Money released under any of these five cases is exempt from income tax on the contributions (social charges of 17.2% still apply to the gains portion).

The sixth case is the one that matters for early retirement planning: purchase of a primary residence. A PER holder can unlock the plan to buy their main home, regardless of whether they are in financial distress. The tax treatment is less favourable than the accidents-de-la-vie cases (deducted contributions are reintegrated into taxable income in the year of release, and gains face the 31.4% flat tax), but the access is real.

This creates a specific French strategy: if you plan your early retirement around buying your primary residence at the transition, you can unlock a large portion of your PER without waiting. For someone retiring at 50 who has been renting and plans to buy, the timing is workable. For someone who already owns, it is not available.

The fifth accident-de-la-vie case, expiration of unemployment benefits, deserves attention. If you leave employment and exhaust your ARE (Allocation d'aide au Retour à l'Emploi, which typically runs 18 to 24 months for workers in their 50s), you become eligible to unlock the PER without tax on the principal. A calculated exit from salaried work, followed by the standard unemployment period, followed by PER release, is a legitimate bridge strategy. It is also well-known to the French tax authority, so the timing must be genuine: you cannot quit with a prearranged plan to claim unemployment and then unlock, without risking the claim being recategorised.

Outside the PER, the main French bridge instrument is assurance-vie. This is a flexible, taxable-but-efficient savings wrapper with no age restrictions on withdrawal. After 8 years of holding, gains benefit from an annual allowance of €4,600 for a single person or €9,200 for a couple, and the marginal rate on gains above that is 7.5% plus 17.2% social charges. An early retiree with a mature assurance-vie (opened 8 or more years before retirement) can draw down gains at an effective rate of around 24.7%, which is competitive with any tax-advantaged wrapper in Europe outside the UK.

The French bridge stack, for someone retiring at 50, is therefore: assurance-vie first (the flexible layer), PER unlocked via primary residence purchase if applicable (the large one-off layer), taxable brokerage as a topper-up. At around 62 to 64, the PER becomes accessible normally alongside the state pension.

The US bridge: four instruments, strict rules

The US has the most options and the strictest rules. There are four routes to early access on pension money, each with specific mechanics.

Roth IRA contribution withdrawals. Contributions (not earnings) can be withdrawn from a Roth IRA at any age, tax-free and penalty-free. If you have contributed $7,000 a year for 20 years, $140,000 in contributions is available without restriction, regardless of the earnings sitting on top. For savers who have maxed out Roth contributions since their 20s, this alone can bridge several years.

Roth conversion ladder. You convert traditional IRA or 401(k) money to Roth each year, pay ordinary income tax on the converted amount in that year, and wait five years. After five years, the converted amount (not the earnings on it) can be withdrawn penalty-free. By converting $40,000 a year starting at age 45, you have $40,000 available at 50, $40,000 available at 51, and so on — a rolling annual bridge. This is the standard FIRE playbook for people whose main accounts are traditional.

72(t) SEPP (Substantially Equal Periodic Payments). Section 72(t)(2)(A)(iv) of the tax code, governed by IRS Notice 2022-6, allows penalty-free withdrawals from an IRA or separated-from-service 401(k) before 59½ if you take a series of substantially equal periodic payments. There are three calculation methods: RMD, fixed amortization, and fixed annuitization. The 2026 interest rate for the amortization and annuity methods is 120% of the federal mid-term rate, currently around 4.5%. Payments must continue for the longer of five years or until you reach 59½. You cannot add to the account, cannot take additional withdrawals, and cannot modify the payment amount — a single misstep triggers retroactive penalties plus interest on every payment you have ever received under the plan. A one-time switch from amortization or annuity method to RMD method is permitted; nothing else is.

The practical implication: 72(t) is a blunt instrument. You calculate your annual payment based on the balance of a specific IRA, and you are committed to that payment for 5 to 14 years depending on your starting age. If your expenses change, you cannot change the payment. A common workaround is to split a large IRA into two before starting the SEPP — you base the 72(t) on one account, leaving the other fully flexible.

Rule of 55. If you leave your employer in or after the calendar year you turn 55, you can take penalty-free withdrawals from that employer's 401(k) or 403(b) — but only from that specific plan, and only if the funds stay in the plan (rolling to an IRA eliminates eligibility). Public safety workers qualify at 50. The Rule of 55 is useful for someone leaving their final job at 55 or later, but it does not help a 45-year-old retiree.

A US early retiree at 45 typically stacks these: Roth contributions come out first (tax-free, penalty-free, unlimited timing). A Roth conversion ladder runs in parallel, converting traditional money each year so it becomes accessible five years later. Taxable brokerage fills any remaining gap. 72(t) is usually avoided unless the other routes are insufficient, because of its inflexibility.

The HSA deserves a footnote. After age 65, HSA funds can be withdrawn for any purpose at ordinary income tax rates, which makes it effectively a Traditional IRA with better accumulation tax treatment. Before 65, only qualified medical expenses are penalty-free. It is not a bridge instrument but a late-stage one.

Common mistakes across all four markets

Three errors turn up regularly regardless of jurisdiction.

The first is confusing "I have enough money" with "I can access enough money." A £1.2M pot that is 90% locked in pension wrappers is not the same as a £1.2M pot that is 50/50 locked and liquid. Bridge planning starts during accumulation, not at the point of retirement, because the allocation between wrapped and unwrapped (or flexibly-wrapped) money is what determines whether you can actually retire when you want to.

The second is underestimating the tax load on bridge years. In every system, the bridge draws mean taxable income, and if you structure them badly you can end up in a higher bracket during retirement than you were during work. The UK case is partial but important: filling the personal allowance each year during the bridge with small SIPP crystallisations is almost always worth doing. The US case is larger: Roth conversions during low-income bridge years before Social Security and RMDs kick in are often the single most valuable tax move an early retiree makes.

The third is not stress-testing the bridge against a bad sequence of returns. If you retire into a 2000 or 2008-style market, your bridge portfolio can shrink by 40% in the first two years at exactly the moment you are withdrawing from it. A bridge designed for average returns can fail completely under bad ones. We cover this in detail in a separate post in this series.

What we are building in Endute

Bridge planning is mechanically specific: the instruments differ by country, the rules change frequently, and the consequences of getting it wrong compound over decades. Endute's FIRE simulator treats wrapped and unwrapped balances separately in the projection, respects minimum pension ages by jurisdiction, and flags when a plan has enough total savings but not enough bridge liquidity to actually stop working on the target date.

The simulator does not attempt to be a tax advisor. For PER early-release strategies, Roth ladders, or 72(t) SEPP plans, the details matter enough that professional advice pays for itself. But it does show you the shape of the problem honestly, which is more than most FIRE spreadsheets do.

Sources and notes

UK

• Minimum pension age 55 rising to 57 April 2028: Finance Act 2022, HMRC guidance updated 2025
• ISA allowance and rules: gov.uk, 2026/27 tax year
• Personal allowance £12,570: HMRC 2026/27
• Cash ISA £12,000 cap for under-65s (April 2027): Autumn Budget 2025

France

• PER early-access cases: Article L224-4 CMF (Code monétaire et financier), six cases as defined by loi PACTE (2019)
• Tax treatment of early release: Article 81 4° bis a CGI for accidents-de-la-vie; PFU 31.4% for gains on primary-residence release
• Prélèvements sociaux 18.6% from 2026: LFSS 2026, loi 2025-1403 art. 12 (applies to PER; assurance-vie remains 17.2%)
• Assurance-vie 8-year allowance €4,600/€9,200: Article 125-0 A CGI

Germany

• Abgeltungsteuer 25% + 5.5% Soli = 26.375%: §32d EStG
• Teilfreistellung 30% for equity funds: §20 Abs. 1 InvStG
• Vorabpauschale Basiszins 3.20% for 2026: Federal Ministry of Finance announcement, January 2026
• Sparerpauschbetrag €1,000: §20 Abs. 9 EStG
• Grundfreibetrag €12,348 for 2026

US

• 72(t) SEPP rules: Section 72(t)(2)(A)(iv) IRC; IRS Notice 2022-6 for calculation methods
• 2026 federal mid-term rate for SEPP calculations: approximately 4.5% (IRS Rev. Rul. monthly)
• Rule of 55: Section 72(t)(2)(A)(v) IRC; public safety worker exception Section 72(t)(10)
• Roth contribution withdrawal ordering rules: Publication 590-B
• Roth conversion 5-year rule: Section 408A(d)(3)(F) IRC

General note on bridge design

• "Roth conversion ladder" terminology popularised by the Mad Fientist blog, though the mechanics derive directly from the IRS Roth ordering rules. The strategy predates the FIRE movement's adoption of it by roughly a decade.

This is the third post in a cluster on early retirement planning across UK, EU, and US markets. See also: "How much do you need to retire early?" (the anchor), "Tax-advantaged accounts compared" (the wrappers themselves), and the forthcoming posts on sequence of returns risk and the 4% rule outside the US.