Currency Risk for Americans Retiring Abroad 2026
Two retirees can earn the same average return and end up in different places, because withdrawals make the order of returns matter. An American retiring into the euro area is exposed to a second ordering problem — the exchange rate — and central bank data show it has been as large as the market one.
An American retiring into the euro area faces two ordering problems at once: market sequence risk and currency risk. Central bank data show the second has been as large as the first — from October 2000 to July 2008 a dollar income lost 45.9% of its euro purchasing power, with no market decline required.
| Evidence | What it shows |
|---|---|
| GAO, 2011 report to Congress | An identical 7% average return sustained withdrawals for either 18 or 24 years, depending only on the order in which the returns arrived |
| S&P 500, 2008 | A decline of 38.5%, producing a total return of -37.0% |
| S&P 500, end-2005 to end-2009 | A cumulative total return of -2.7% across four years |
| U.S. household and non-profit net worth | Fell 20% between December 2007 and December 2008 — about $13 trillion |
| The euro, October 2000 to July 2008 | Monthly averages ranged from $0.8525 to $1.5759, so a euro cost 84.9% more in dollars at the peak. A dollar income lost 45.9% of its euro purchasing power, with no market decline required |
| The euro, July 2008 to March 2015 | Fell 31.3% against the dollar, so a dollar bought 45.7% more euros |
| The euro, January 2021 to September 2022 | Fell 18.7% in twenty months |
| ECB reference rate, 2026 | 1.1797 on 17 April, 1.1340 on 24 June, and 1.1485 on 31 July |
| Bengen, 1994 | Counsels withdrawal at no more than a four-percent rate during the early years of retirement, and identifies 3% as the level that never produced portfolio longevity under 50 years |
| The Trinity authors, peer-reviewed paper | For a 20-year payout with inflation-adjusted withdrawals and at least 75% stock, a 4% rate succeeded in 77% of periods — not the roughly 95% popular retellings attribute to it |
| Guyton and Klinger, 2006 | Initial withdrawal rates of 5.2% to 5.6% are sustainable at a 99% confidence standard for portfolios of at least 65% equities, using decision rules that include no spending increase after a year of negative total return and a 10% cut when the current withdrawal rate exceeds 120% of the initial rate |
Why order matters once money is coming out
The Government Accountability Office put the mechanism plainly in a 2011 report to Congress: if the sequence of returns in the second and third year were reversed, holding all else constant, the average annual return would be the same, yet if withdrawals are made each year, savings would be depleted sooner with the first sequence. Its accompanying illustration shows an identical 7% average return sustaining withdrawals for either 18 or 24 years depending only on order. The reason it gives is that where drawdowns begin after investments have declined, the income drawn depletes a greater proportion of the portfolio than if growth had come first.
The Society of Actuaries describes the same risk: someone saving with a long time horizon may be able to wait for prices to recover, while a retiree who needs income immediately may be forced to sell when prices are down, substantially reducing assets and future income. Where there are no cash flows, order has no effect on the ending value — arithmetic rather than a research finding, and what makes the withdrawal phase different in kind from the accumulation phase.
A caution about the vocabulary. The phrase "retirement risk zone," with its familiar five-to-ten-years-either-side bracket, does not appear in any government, intergovernmental, actuarial or peer-reviewed source located; it circulates only on marketing sites. What is independently supported is narrower: research summarised by the CFA Institute frames the risk as bad outcomes at the wrong time — not just long-term average returns but when those returns are earned — with poor returns early in decumulation more damaging than poor returns later. Work published in the Journal of Financial Planning pushes back on any fixed window at all, holding that sequence risk is always present to some degree when there are cash flows out of the portfolio.
How large the market leg has actually been
Federal figures are more useful here than illustrative scenarios. GAO records that the S&P 500 declined 38.5% in 2008, producing a total return of -37.0%, and — the more instructive number for a retiree — that from end-2005 to end-2009 the total cumulative return of the index was a 2.7% loss. That is four consecutive years in which a withdrawing portfolio funded spending out of principal with no market growth to replace it. The Economic Report of the President records the wider damage: household and non-profit net worth fell 20% between December 2007 and December 2008, about $13 trillion.
The withdrawal-rate literature is usually paraphrased into something stronger than it says. William Bengen’s 1994 Journal of Financial Planning paper counsels withdrawal "at no more than a four-percent rate during the early years of retirement," while identifying 3% as the level that never produced portfolio longevity under 50 years. The Trinity authors’ subsequent peer-reviewed paper reports that for a 20-year payout with inflation-adjusted withdrawals and at least 75% stock, a 4% rate succeeded in 77% of periods — not the ~95% popular retellings attribute to it.
What is specific to Americans abroad
An American retiring in Portugal, Spain or France generally spends in euros while holding assets and receiving Social Security in dollars. That layers a second sequence on top of the market one, and Federal Reserve H.10 data show it has been the larger of the two over some spans.
On monthly averages, the euro has ranged since its introduction from 0.8525 dollars in October 2000 to 1.5759 dollars in July 2008 — a euro cost 84.9% more in dollars at the peak than at the trough. Read from the dollar-holder’s side, in October 2000 a dollar bought €1.173; in July 2008 it bought €0.635. An American whose income was in dollars lost 45.9% of their euro purchasing power over those eight years, with no market decline required.
It runs the other way too. From July 2008 to March 2015 the euro fell 31.3% against the dollar, so a dollar bought 45.7% more euros; from January 2021 to September 2022 it fell 18.7% in twenty months. Even within one recent quarter the ECB reference rate moved from 1.1797 on 17 April 2026 to 1.1340 on 24 June 2026, and stood at 1.1485 on 31 July 2026.
A bad market year and an adverse currency year arriving together, early in withdrawals, is the specific combination a cross-border plan is exposed to and a domestic one is not.
It is also worth saying what is not known. Searches of the OECD, BIS, ECB, IMF, World Bank, NBER, SSA and the Center for Retirement Research found no research on households that hold assets in one currency and spend in another; what exists concerns currency exposure in pension funds as institutions, a different question. The exchange-rate history above requires no intermediary’s opinion, but the household-level version of the problem appears not to have been studied.
On the two standard responses
One of the usual answers has independent evidence behind it and one does not.
Variable withdrawal rules are supported. Guyton and Klinger’s 2006 Journal of Financial Planning paper concludes that initial withdrawal rates of 5.2% to 5.6% are sustainable at a 99% confidence standard for portfolios of at least 65% equities, using decision rules that include no spending increase after a year of negative total return and a 10% cut when the current withdrawal rate exceeds 120% of the initial rate.
The cash-buffer or bucket approach is not. The one piece of independent academic work located on the question — Javier Estrada of IESE Business School, published in the Journal of Investing — finds that simple static strategies, which by definition involve periodic rebalancing, clearly outperform bucket strategies on four separate measures, because static strategies sell what has become relatively expensive and buy what has become relatively cheap, while bucket strategies forgo the second half of that. No government, intergovernmental, actuarial or peer-reviewed source supporting cash buffers on a total-return basis was located; every affirmative source found was a firm selling a service or a product. The defensible case for the technique is behavioural rather than mathematical — and the behaviour it addresses is documented: work published by the Center for Retirement Research on 2008-09 found that investors did in fact sell low.
What the evidence does and does not settle
The sequence of returns cannot be chosen, and neither can the exchange rate. What the record establishes is the scale of both, and that for a dollar-funded household spending euros they are two separate exposures rather than one. What it does not establish is which response fits a given household — that turns on the ratio of essential to discretionary spending, how much of the income is dollar-denominated and fixed, the timing of the first withdrawal relative to the move, and a tolerance for varying spending that no dataset contains.
Sources
- U.S. Government Accountability Office — GAO-11-400, Retirement Income: Ensuring Income throughout Retirement Requires Difficult Choices, June 2011 (the reversed-sequence demonstration; S&P 500 2008 and 2005-2009 figures) — https://www.gao.gov/assets/a319390.html — checked 2 August 2026
- Board of Governors of the Federal Reserve System — H.10 / FRED series EXUSEU, U.S. dollars to euro spot exchange rate, monthly, 1999-2026 — https://fred.stlouisfed.org/data/EXUSEU.txt — checked 2 August 2026
- European Central Bank — Euro foreign exchange reference rates, U.S. dollar — https://www.ecb.europa.eu/stats/policy_and_exchange_rates/euro_reference_exchange_rates/html/eurofxref-graph-usd.en.html — checked 2 August 2026
- Society of Actuaries — Managing Post-Retirement Risks: Strategies for a Secure Retirement, 2025 — https://www.soa.org/globalassets/assets/files/resources/research-report/2020/post-retirement-strategies-secure-chart.pdf — checked 2 August 2026
- CFA Institute Research and Policy Center — Managing Sequence Risk (summarising Clare, Seaton, Smith and Thomas, Financial Analysts Journal, 2017) — https://rpc.cfainstitute.org/research/financial-analysts-journal/2017/managing-sequence-risk — checked 2 August 2026
- William P. Bengen — Determining Withdrawal Rates Using Historical Data, Journal of Financial Planning, October 1994 — https://www.financialplanningassociation.org/sites/default/files/2021-04/MAR04%20Determining%20Withdrawal%20Rates%20Using%20Historical%20Data.pdf — checked 2 August 2026
- Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz — Sustainable Withdrawal Rates From Your Retirement Portfolio, Financial Counseling and Planning, 10(1), 1999 — https://www.afcpe.org/wp-content/uploads/2018/10/vol1014.pdf — checked 2 August 2026
- Jonathan T. Guyton and William J. Klinger — Decision Rules and Maximum Initial Withdrawal Rates, Journal of Financial Planning, March 2006 — https://www.financialplanningassociation.org/sites/default/files/2021-11/2006%20-%20Guyton%20and%20Klinger%20-%20Decision%20Rules%20and%20SWR%20(1).PDF — checked 2 August 2026
- Larry R. Frank Sr. and David M. Blanchett — The Dynamic Implications of Sequence Risk on a Distribution Portfolio, Journal of Financial Planning, June 2010 — https://www.financialplanningassociation.org/sites/default/files/2021-10/JUN10%20JFP%20Frank%20and%20Blanchett%20PDF.pdf — checked 2 August 2026
- Javier Estrada, IESE Business School — The Bucket Approach for Retirement: A Suboptimal Behavioral Trick?, Journal of Investing, 28(5), 2019 — https://blog.iese.edu/jestrada/files/2019/01/BucketApproach.pdf — checked 2 August 2026
- Executive Office of the President — Economic Report of the President 2010, chapter 2 (S&P 500 second half of 2008; household net worth) — https://www.govinfo.gov/content/pkg/ERP-2010/html/ERP-2010-other-3.htm — checked 2 August 2026
- Social Security Administration — Annual Statistical Supplement 2025, Table 5.J (beneficiaries in foreign countries) — https://www.ssa.gov/policy/docs/statcomps/supplement/2025/5j.html — checked 2 August 2026
Figures and rates cited were current as of August 2026 and are subject to change.
Common questions
- Why does a bad market year early in retirement matter more than a bad one later?
- Because withdrawals during a decline deplete a greater proportion of the portfolio than if growth had come first. The GAO told Congress that reversing the returns of the second and third year leaves the average annual return unchanged, yet with annual withdrawals savings are depleted sooner — its illustration shows an identical 7% average return lasting either 18 or 24 years. Without cash flows, order has no effect.
- Is the 4% rule really a 95% success rate?
- No. The Trinity authors’ peer-reviewed paper reports that for a 20-year payout with inflation-adjusted withdrawals and at least 75% stock, a 4% rate succeeded in 77% of periods — not the ~95% popular retellings attribute to it. William Bengen’s 1994 paper put the figure at no more than a four-percent rate during the early years of retirement, identifying 3% as the level that never produced longevity under 50 years.
- How much can the euro-dollar rate change what my US retirement income buys in Europe?
- Over one eight-year stretch it moved more than the market did. On Federal Reserve monthly averages the euro ranged from 0.8525 dollars in October 2000 to 1.5759 dollars in July 2008. A dollar bought €1.173 in October 2000 and €0.635 in July 2008 — a 45.9% loss of euro purchasing power, with no market decline required. It has run the other way too.
- Do cash buckets actually protect a retirement portfolio from sequence risk?
- Not on a total-return basis, on the evidence located. Javier Estrada of IESE Business School, published in the Journal of Investing, finds that simple static strategies involving periodic rebalancing clearly outperform bucket strategies on four separate measures, because static strategies sell what has become relatively expensive and buy what has become relatively cheap. The defensible case for buckets is behavioural rather than mathematical.
- Is there really a retirement risk zone five to ten years either side of retiring?
- That phrase does not appear in any government, intergovernmental, actuarial or peer-reviewed source located — it circulates on marketing sites. What is independently supported is narrower: the CFA Institute frames the risk as poor returns early in decumulation being more damaging than poor returns later, and sequence risk is present wherever cash flows leave the portfolio.
- Does the order of returns matter while I am still saving rather than withdrawing?
- Where there are no cash flows, order has no effect on the ending value — arithmetic rather than a research finding, and what makes the withdrawal phase different in kind. The Society of Actuaries frames it the same way: a saver with a long horizon may be able to wait for prices to recover, while a retiree needing income may be forced to sell when prices are down.
- How bad was 2008 for someone already drawing on their portfolio?
- GAO records that the S&P 500 declined 38.5% in 2008, producing a total return of -37.0%, and that from end-2005 to end-2009 the total cumulative return of the index was a 2.7% loss — four consecutive years in which a withdrawing portfolio funded spending out of principal.
- Is there any research supporting withdrawal rules that flex with the market?
- Yes. Guyton and Klinger’s 2006 Journal of Financial Planning paper concludes that initial withdrawal rates of 5.2% to 5.6% are sustainable at a 99% confidence standard for portfolios of at least 65% equities, using decision rules that include no spending increase after a year of negative total return and a 10% cut when the current withdrawal rate exceeds 120% of the initial rate.
- Has the euro ever moved in favour of someone holding dollars?
- Yes, and by large margins. From July 2008 to March 2015 the euro fell 31.3% against the dollar, so a dollar bought 45.7% more euros; from January 2021 to September 2022 it fell 18.7% in twenty months. Even within one recent quarter the ECB reference rate moved from 1.1797 on 17 April 2026 to 1.1340 on 24 June 2026.
- Is there research on people who hold assets in dollars but spend in euros?
- None was located. Searches of the OECD, BIS, ECB, IMF, World Bank, NBER, SSA and the Center for Retirement Research found no research on households that hold assets in one currency and spend in another; what exists concerns currency exposure in pension funds as institutions, a different question.
- What makes sequence risk different for an American retiring in Europe?
- A second ordering problem sits on top of the market one. An American retiring in Portugal, Spain or France generally spends in euros while holding assets and receiving Social Security in dollars, and Federal Reserve H.10 data show the currency leg has been the larger of the two over some spans. A bad market year and an adverse currency year arriving together is the specific combination.
A note on financial planning content
Financial planning topics discussed by US Expat are educational and general in nature. They should not be treated as personalized investment, financial, or retirement advice unless provided through the appropriate advisory relationship and documentation. Where regulated advisory services are referenced, they are offered separately and are subject to applicable jurisdiction, registration, and compliance requirements.
The information provided by US Expat is for educational and informational purposes only. It should not be treated as personalized tax, legal, immigration, investment, insurance, or financial advice. U.S. citizens abroad should consult qualified professionals who understand their specific facts and circumstances. US Expat is not a law firm, tax filing firm, or immigration law firm.
Rules and thresholds for U.S. citizens abroad change. Pages are reviewed periodically; confirm current details with a qualified professional before acting.