Ask anyone with a retirement plan what they are afraid of and they will say a crash. The 2008 they remember, the portfolio cut in half the year they stop working. It is a reasonable fear, and it is the one the entire planning industry is built around: the 4% rule, the Monte-Carlo simulators, the “what if the market drops 40% in year one” stress tests all exist to fortify that one wall. They do it well — and for someone growing old abroad, they are fortifying the wrong wall.
A market crash has two properties worth naming precisely. It is uncertain (it may not happen in your window at all) and it is, historically, recoverable, which is the whole reason a 4% withdrawal rate survives the worst sequences on record. The shocks that actually end expat retirements have the opposite properties. They are certain, or close to it, and they are permanent. This page is about testing your plan against those instead. It is decision analysis, not financial advice; verify your own figures with a licensed professional.
The shock you test, and the shock you should
Lay the shocks out on two axes (how certain each is, and whether you recover from it) and the misallocation of worry becomes obvious.
| Shock ↓ | Certain? | Recoverable? | Tested by standard tools? |
|---|---|---|---|
| Market crash / sequence risk | Uncertain may not hit your window | Historically yes what the 4% rule survives | Yes the main thing they model |
| Insurance cliff / lapse | Near-certain by ~75 premiums ×2/decade, caps at 70–75 | No cannot re-enter the market | No |
| Long-term-care tail | High probability most reaching 65 need some care | No ongoing, ≈ a whole income | Rarely |
| Frozen pension + adverse FX | Certain by country binary; plus likely FX drift | No does not mean-revert | No |
Source: Synthesised from the site's cliff, care, FX and pension work (sourced per piece) · checked 2026-05-22
Read the bottom three rows. Each is near-certain, none is recoverable, and standard planning tests none of them. Take them one at a time, because the certainty is the part people resist.
The insurance cliff is not a maybe. Expat health premiums roughly double each decade to 75 and can rise five-fold after, new-applicant age caps cluster at 70 to 75, and the real failure is lapse-by-attrition: you are priced out of your own grandfathered policy at the age you most need it, with no way back into the market. That is not a tail event. It is the default trajectory of holding cover into your eighties on a flat income.
The long-term-care tail is a high-probability event, not a remote one (a large share of people reaching 65 will need some long-term care), and a single care line, a Thai full-time nurse around 37,500 THB a month or Philippine assisted living from roughly ₱62,000, can absorb an entire base income. The frozen pension is binary and certain: frozen for life in a non-agreement country, with sterling having fallen 35 to 40% against the baht over eighteen years on top. Neither the freeze nor the currency drift bounces back the way a market does.
So the planning convention is exactly inverted. It rigorously models the one shock that might not happen and that you would recover from, and ignores the three that probably will happen and that you will not.
The stress-test protocol
The correction is not a new calculator. It is a posture: price the certain, permanent shocks as the base case, not the adverse case, and see whether the plan still stands. Take the plan you would otherwise run through the drawdown worksheet and apply all five of these at once.
- Cover gone by 75. Set private health insurance to zero from age 75 (lapse-by-attrition) and self-fund the health line thereafter at the 11–14% medical trend.
- A care line from 80. Add ongoing care costs from age 80, sized at roughly one base income.
- Pension frozen. If the destination has no uprating agreement, hold the state pension flat in nominal terms for the whole run.
- Adverse FX. If your income and spending currencies differ, apply a −2% real drift per year to the pair.
- A 3.3% draw. Lower the withdrawal rate from 4% to 3.3%, Morningstar’s 2021 low-water mark (below even its current 2026 reading of ~3.9%), used as a conservative stress floor.
| Date | THB per GBP | Range | Basis | Note |
|---|---|---|---|---|
| 2007 | 70 | 65–75 | triangulated | pre-2008 sterling peak (GBP ≈ $2.00, USD/THB ≈ 33–36) |
| 2016 | 45 | 40–45 | triangulated | settled into the ~40–45 band after the Brexit vote |
| 2026-05 | 42.98 | 41.69–44.3 | sourced | May spot; 2026 range 41.69 (21 Jan) to 44.30 (4 May). Re-read 24 May ~43.9, within range; value held at the representative mid-May figure. |
Source: Bank of England spot series (via poundsterlinglive GBP-THB history) · latest 42.98 THB per GBP (2026-05) · as of 2026-05-24
That is not five separate scenarios. It is one scenario, because for the modal expat these conditions co-occur — they are the same retirement seen at 80, not five different unlucky retirements. A plan that stays solvent to about 90 under the combined load is genuinely robust. A plan that fails the moment you apply them is not weak; it is a plan that was only ever solvent in the brochure’s benign case, which is the one case guaranteed not to be the one you live.
Reading the result
Two outcomes, and they mean different things.
If the plan survives the combined stress to 90, you have something rare: a retirement that does not depend on the kindness of the exchange rate, the insurer, or your own continued good health. That is the profile for whom moving abroad is a sound financial decision rather than a wager, and you should proceed knowing the stress test, not the brochure, cleared you.
If it fails (and for the modal case of a single frozen pension or a couple on a sliding currency, it fails in the seventies), the instinct is to look for the input you can soften. Resist it on the permanent shocks. You can argue the withdrawal rate or the FX drift, because those are genuinely uncertain. You cannot argue the cliff or the care tail down to zero, because they are not volatility you are unlucky to meet; they are the structure of growing old far from a system that would have absorbed them. The honest response to a failed stress test is not a more optimistic assumption. It is a smaller plan, a different destination with an uprating pension and reachable care, or the recognition that the move only works on the benign path and the benign path is not a plan.
The market crash you were taught to fear is the recoverable one. The shocks you were never taught to test are the ones that do not let go. Fortify accordingly, and run the test before the move makes the test moot — because the alternative is to discover which wall was load-bearing at 80, in a second language, with no margin left to rebuild it.
This article is analysis, not financial advice. The stress-test protocol is a deterministic illustration built from this site’s sourced cliff, care, currency and pension figures; “certain” and “near-certain” describe a planning posture toward high-probability permanent shocks, not a prediction about any individual. Treaty relief, an uprated destination, a larger income or genuine self-insurance can change the result. Verify your own assumptions with a licensed financial professional before acting.
Questions
How do you stress-test a retirement plan for living abroad?
Differently from how you stress-test one at home. The standard approach tests a portfolio against a market crash and a bad sequence of early returns, which is why the 4% rule and its 3.3% reassessment exist. For an aging expat that misses the shocks that actually end retirements, which are certain and permanent rather than uncertain and recoverable: the insurance cliff that prices you out of cover by 75, the long-term-care tail, and a frozen pension eroded by currency drift. The protocol is to re-run the plan with those priced as the base case (cover gone at 75, care from 80, pension frozen, adverse exchange-rate drift, a 3.3% withdrawal rate), and check whether it stays solvent to about 90.
Is a market crash the biggest risk to retiring abroad?
No, and that is the costly misconception. A market crash is genuinely frightening, but it is uncertain in timing and historically recoverable over a retirement horizon — which is exactly what the 4% withdrawal rule is built to survive. The shocks that more reliably end an expat retirement are the ones standard planning ignores because they are not market events: being priced out of health cover by 75 with no way back into the market, a care need that costs a full income, and a pension frozen by treaty and eroded by exchange rates. Those do not bounce back. Fortifying against the crash while ignoring the cliff is defending the wrong wall.
What is lapse-by-attrition in expat health insurance?
It is the real shape of the insurance cliff. The headline fear is being refused new cover at 70, but the more common failure is subtler: you hold a policy from your sixties, it renews for life, and its premium roughly doubles each decade — until the renewal cost overtakes your flat or frozen pension at an age when no insurer will take you as a new applicant. The policy does not get cancelled. You can no longer afford to renew it, and you cannot replace it. You lapse out of your own cover at the age you are most likely to need it, which is why a stress test must model cover as gone by 75.
What withdrawal rate should I use to stress-test retiring abroad?
Use 3.3% as the stress case rather than the classic 4%. Morningstar reassessed the safe starting rate down to roughly 3.3% in 2021 on high valuations and sequence-of-returns risk, then back up to about 3.9% by 2026; the 3.3% low-water mark is a deliberately conservative stress floor, not the current figure. But the withdrawal rate is only one input, and for a retirement abroad it is not the highest-leverage one. The certain-and-permanent shocks (the insurance cliff, the care tail, the frozen-pension-and-FX compression) move the failure year more than a half-point on the withdrawal rate does. Lower the rate to 3.3% as part of the stress test, then spend your real attention on pricing the permanent shocks.