“I’ll just go home if it goes wrong.” It is the most common thing a prospective expat says — and the least examined. It closes the conversation. It is offered as the floor under every other risk: the currency, the insurance cliff, the isolation, the long-term-care tail, all survivable, because there is always the door. This piece takes the sentence literally and tests it as what it grammatically is: a conditional. If it goes wrong, then I go home. The finding is narrow and exact. The conditional is false precisely when its antecedent is true. It is analysis of a logical structure, not advice, instruction, or a prediction about any individual.
The sentence as a conditional
A conditional is only load-bearing if the consequent is reliably available whenever the antecedent fires. “If the fire alarm sounds, I’ll use the stairs” is a plan because the stairs are there independent of the fire. “I’ll just go home if it goes wrong” is being used the same way: as a constant exit, available on demand, its presence unaffected by whatever triggered the need for it.
That independence is the thing to test. “Go home” is not a single action. It is a bundle of preconditions that all have to hold at once: the money to go, the health to travel, an entitlement to land into, a home that still receives you, and the capacity to decide and execute. The sentence treats all five as fixed. The question is whether they are fixed, or whether they move; and if they move, which way they move relative to the trigger.
Decomposing “go home” into five preconditions
Precondition 1: the money to go home
The implicit price of “go home” is a long-haul ticket. The actual price is the ticket plus resettlement plus re-housing into a market that moved while the person was away, net of the capital that the life abroad has immobilised. The Thai retirement-visa float (roughly 800,000 THB, seasoned in-country and held in baht) is not liquid on the day the decision is made; it is collateral for the status that has to be surrendered to leave. A foreign-currency drawdown that has fallen against the home currency crystallises its loss at exactly the moment it is converted back. The home housing and cost base did not wait. The liquid sum genuinely available to resettle is smaller than the sum the sentence assumes, and it is smallest in the scenario the sentence most needs it: the one where the trigger is that the money ran low. This is the money-doesn’t-last failure mode and the annual solvency test the visa quietly imposes, arriving together.
Precondition 2: the health to travel
While well, “go home” costs a ticket. The trigger, in the cases where it matters most, is a health event, and a health event reprices the journey. By aggregated medical-repatriation pricing, a commercial medical escort for a stable but unwell patient runs roughly $3,000–$25,000, and $15,000–$35,000 for a long-haul international airline escort. A commercial stretcher booking, with several seats purchased and removed, is about $15,000–$75,000, with long-haul in the $25,000–$30,000-plus range before the rest of the booking. A dedicated air ambulance for an unstable patient is $12,000–$25,000 short-haul and $50,000–$120,000-plus for long-haul intercontinental repatriation, with an Asia-to-home leg reported past $200,000 at the complex end. Those bands are triangulated across medical-transport providers, not quotes, and they move year to year; the structure they describe does not. The cost of getting home is a rising function of how wrong it has gone.
And there is a gate before the price even applies. The patient must be declared fit to fly by both a treating medical director and the airline. Past a certain severity the answer is no at any price, and the person does not go home by air at all; they go home, if at all, as a repatriated body, on a different and slower schedule. The clause’s consequent is not just expensive in the bad case. It can be clinically unavailable in the worst one.
Precondition 3: the entitlement to land into
The sentence assumes the home healthcare system is idling, waiting to receive its returning citizen. It was switched off by the move. UK NHS entitlement to free secondary (hospital) care depends on being “ordinarily resident” — living in the UK lawfully, voluntarily and for settled purposes — so it ends on permanent relocation. A citizen returning to resume properly settled residence will most likely meet the test from the date of arrival, but the test is assessed, not automatic, and anyone a trust treats as not ordinarily resident is an overseas visitor charged 150% of the cost of treatment. That is the live exposure for a hedged or ambiguous return rather than a clean one — and it sits on top of the harder fact that the entitlement was a thing the person stopped accruing the day they left. US Medicare is territorial: it does not lapse, but it pays essentially nothing in the country where the expat actually lives, so the person has been paying for, or forgoing, a benefit that produces nothing until they are physically back inside it. The mechanism is set out in full in what Medicare and the NHS won’t cover. The thing being returned to is not waiting intact. It was ended by the same act that created the need to return to it.
Precondition 4: the home to go to
“Home” in the sentence is a place. For the person who sold the house, surrendered the tenancy, and moved a life across the world, the place may no longer be a place that receives them. The home network that would have absorbed a return has dispersed on its own schedule: people aged, moved, or died while the expat was away, which is the structural attenuation that runs underneath the whole genre. “Go home” silently assumes a home that is still configured to take you back. Often what is left is the country, not the household: a destination in the abstract, with nothing specific at the end of it.
Precondition 5: the capacity to invoke it
This is the precondition the sentence cannot see, because the sentence is spoken by exactly the faculty the trigger removes. In cognitive decline, self-detection fails by the disease’s own mechanism: anosognosia is present in roughly 60% of mild cognitive impairment and 81% of Alzheimer’s cases, so the person cannot reliably register that it is time to go. In acute bereavement, the widower-mortality effect is concentrated in the first six months, the precise window in which a decisive international relocation is least feasible to plan and execute. The trigger and the loss of the capacity to act on the trigger are frequently the same event. The clause assumes a decider who is still there to invoke it. Several of the triggers work by removing the decider.
The artefact: the return-option decay schedule
Set the five preconditions in a column and, against each, the trigger that would invoke the clause and the direction the precondition moves as that trigger strengthens. The right-hand column is the finding.
| Precondition of "go home" | Backed by | What the trigger does to it |
|---|---|---|
| The money to go | Locked visa float, crystallised FX loss, a home market that moved | Often is the trigger (money ran low); the liquid sum shrinks as it is needed. |
| The health to travel | Escort $3–35k / stretcher $15–75k / air ambulance $50–120k+ (Asia-to-home past $200k); fit-to-fly gate | Cost rises with severity; past a threshold, unavailable at any price. |
| The entitlement to land into | NHS depends on 'ordinarily resident'; severed by relocation; 150% of cost for anyone deemed not resident; Medicare territorial | Switched off by leaving; re-entry is assessed, not automatic — the charge bites a hedged or ambiguous return. |
| A home that still receives you | House sold, tenancy gone, the home network dispersed by age | Erodes monotonically with time abroad, independent of the trigger. |
| The capacity to invoke it | Anosognosia ~60% MCI / ~81% Alzheimer's; the acute-bereavement window | Removed by the trigger itself, in the cognitive and bereavement cases. |
Source: Preconditions synthesised from gh1–gh5; medical-transport bands aggregated across providers (SkyCare, Medical Air Service, Allianz), NHS/Medicare entitlement per GOV.UK / CMS — indicative, not quotes · checked 2026-05-24
Read the right-hand column as a single object. Every precondition either is the trigger, moves the wrong way as the trigger strengthens, or is removed by the trigger directly. None of them is independent of the antecedent the way the fire-alarm stairs are independent of the fire. Plot any of them against severity and they all slope the same way. The availability of the return option is a decreasing function of exactly the variable that would call it. It is a door that narrows in proportion to how much you need to go through it — that co-movement is the artefact. It is not summarisable as “returning is hard”; it is the specific claim that the option is anti-correlated with its own use, and that the correlation is structural rather than incidental.
This is also why the documented return rate concentrates where it does. Return is exercised early and well, by the cohort for whom every precondition still holds, or it is frequently not exercised at the moment of the trigger at all. The clause is not a constant. It is a window, and the window is closing along the same axis that opens the need.
Why the sentence still feels safe
The conditional is not stupid, and the people who say it are not naive. It feels safe for a precise structural reason. At the moment it is spoken, early in the project, while the person is well, solvent, capacitated, and still housed at both ends, every one of the five preconditions is satisfied. The conditional is true at the point of utterance. The error is not in evaluating it — it is in inferring from one true evaluation that the truth value is constant. The sentence reads its own truth off the easiest point on the curve and treats that reading as the whole curve. It is the geographic-cure reasoning in miniature: a state that is real at departure, assumed to be permanent, drawn on years later when it has quietly decayed to fiction. The same decay runs through paradise as a place you visit: a condition sampled at its best point and projected flat.
The narrow case where it holds
The clause is not universally false, and saying so would be its own kind of dishonesty. It holds for one configuration, named precisely. The mover who is still early, still well, still well-funded, still capacitated, still housed at the home end, and who exercises the option deliberately and ahead of the trigger, treating it as a funded and maintained asset rather than an assumed constant. For that person the conditional is true because they invoke it at the point on the curve where all five preconditions still hold, not at the point where the trigger has already consumed them. Repatriation cover sized to the air-ambulance band, the home entitlement deliberately preserved or its re-qualification gap explicitly budgeted, a specific home destination kept open, and, for the cognitive and bereavement cases, a named decision-maker and trigger agreed in advance, against the day the person cannot decide for themselves. That is no longer the sentence. That is a plan, with a funded mechanism and a named executor. The sentence has neither — which is the entire difference between them.
The cold close
“I’ll just go home if it goes wrong” is a conditional whose consequent is reliably available only while its antecedent is false. Each precondition of going home is consumed by, anti-correlated with, or directly removed by the triggers that would invoke the clause: the money is tightest when the money is the problem, the journey is dearest and may be clinically barred when the health is the problem, the home entitlement was switched off by leaving, the home itself dispersed while you were away, and in the cognitive and bereavement cases the faculty that would say “now” is the faculty the trigger takes. The option is most available exactly when it is least needed and least available exactly when it is called.
It is not, on the evidence of its own structure, a plan. It is reassurance that is true at the moment it is spoken and is taken for a constant because the speaker is standing on the one point of the curve where it holds. Whether to convert it into an actual plan (funded, dated, with the home end kept open and a decider named) is a decision available only early, while the preconditions still hold, and not at the point the curve has already closed. That is the whole of what the structure says. It is a description of a conditional and the point on its curve at which it is read. It is not counsel, and nothing in it is asserted as a probability for any person.
This article is analysis, not advice; it does not diagnose or instruct. Verify any health, legal, financial, or immigration specifics with a licensed professional in the relevant jurisdiction.
Questions
How much does it actually cost to get home if you are seriously ill abroad?
Not a return airfare. By aggregated medical-repatriation pricing, a commercial medical escort for a stable but unwell patient runs roughly $3,000–$25,000, rising to $15,000–$35,000 for a long-haul international airline escort. A commercial stretcher booking (multiple seats removed) is about $15,000–$75,000. A dedicated air ambulance for an unstable patient is $12,000–$25,000 short-haul and $50,000–$120,000-plus for long-haul intercontinental repatriation. The patient must also be declared fit to fly by both a medical director and the airline, which is a clinical gate, not a payment.
Does the NHS or Medicare just resume when I move back?
Not cleanly. UK NHS entitlement to free secondary (hospital) care depends on being "ordinarily resident" and ends on permanent relocation. A citizen returning to resume properly settled residence will most likely qualify from the date of arrival, but the test is assessed rather than automatic, and anyone a trust treats as not ordinarily resident is charged 150% of the cost of treatment — the live exposure for a hedged or ambiguous return. US Medicare is territorial: it does not lapse but pays essentially nothing where an expat actually lives. The home system the sentence relies on was severed by the act of leaving it.
Why is the return option described as anti-correlated with its own use?
Because each precondition of "go home" — liquid money, fitness to travel, a home entitlement to land into, a home that still receives you, the capacity to decide — is consumed by, or moves opposite to, the very triggers that would invoke the clause. As the trigger strengthens (money runs low, a health event, cognitive decline, bereavement), every precondition weakens together. The option is most available exactly when it is least needed, and least available exactly when it is called.
Is there any case where the clause does hold?
Yes, and it is named here rather than erased. The clause holds for the early, well, well-funded, still-capacitated, still-housed mover who exercises it deliberately and ahead of the trigger — as a funded, maintained asset, not an assumed constant. Documented expatriate return concentrates at exactly those early-and-well points. The failure is not that return is impossible; it is that the clause is read off the wrong point on its own curve.
What would make "I'll go home if it goes wrong" an actual plan rather than reassurance?
It would have to be costed and pre-funded as a real asset (repatriation cover or a ring-fenced sum sized to the air-ambulance band, not an airfare), the home entitlement deliberately preserved or its re-qualification gap budgeted, a specific home destination kept open rather than assumed, and a trigger and decision-maker named in advance for the case where the person cannot decide. A plan has a funded mechanism and a named executor. The sentence has neither.