EUR/USD since 1999: history, major cycles and what drives the euro against the dollar

A guide built on the European Central Bank's daily reference rate, 1999 to 2026 (7,107 observations), checked against the Federal Reserve's H.10 series. Updated October 5, 2026, with data to October 5. · Written by Svetoslav Grigorov

Short answer: on the ECB's official reference rate, the euro's all-time low was $0.8252 on October 26, 2000 and its all-time high $1.5990 on July 15, 2008. It fell below parity again in 2022, down to 0.9565 on September 28, and stood at 1.1204 on October 5, 2026. FRED users will see a different record high, 1.6010 on April 22, 2008, because the Fed's series is taken at noon in New York. In 27 years EUR/USD has gone through six major cycles. Interest-rate gaps, above all the 2-year gap, usually get the direction right, but explain less than half of the size of the moves.

This is a reference page, not a news story. Every figure comes from a public series. Our own calculations are marked CALCULATED, claims from sources we have read are marked VERIFIED, and plausible explanations this guide does not prove are marked HYPOTHESIS. There are no forecasts.

How to read EUR/USD

EUR/USD is the number of US dollars one euro buys. At 1.1204, €1 buys $1.1204. When the number rises, the euro strengthens and the dollar weakens; when it falls, the dollar is gaining. If you think in dollars, flip it: 1 / 1.1204 = €0.8925 per dollar. A "strong dollar" story and a "weak euro" story are the same line on the chart, read from the other side.

This guide uses the ECB euro reference rate: one value per business day, set through a concertation between central banks at around 2:10 pm Frankfurt time and published at around 4 pm. The ECB says it is for information purposes only, not a tradable price. It starts on January 4, 1999 (1.1789), the euro's first day. Series that go back to 1994 or earlier are synthetic, built from the ECU or a basket of currencies, and are not mixed in here.

All-time highs and lows: why the ECB, FRED and intraday data disagree

Search for the euro's record low and you will find 0.8252, 0.8270, 0.8225 or 0.8234; for the record high, 1.5990, "1.60" or 1.6038. They can all be right. They measure different things at different times of day.

• ECB (the figure this guide uses). CALCULATED Extremes of the 7,107 daily reference rates, set at around 2:15 pm Frankfurt time.
• FRED, noon in New York. CALCULATED FRED's DEXUSEU series comes from the Federal Reserve's H.10 release: noon buying rates in New York, certified by the New York Fed, about four hours after the ECB rate. That is where the widely quoted "1.60 on April 22, 2008" comes from: at New York noon the peak came in April (1.6010), and in July FRED never went above 1.5923. On the ECB rate, April topped out at 1.5940 (April 23, 2008) and the high came in July. So if your FRED chart and an ECB-based table disagree on the date of the record, neither is wrong.
• Intraday. The highest or lowest price traded at any moment of the day, as reported by the press and wire services: 0.8225 (Börse Online) and 0.8234 (another wire chronology) on October 26, 2000; 1.6038 on July 15, 2008, and 1.6018 on April 22, the previous record. They differ by a few pips depending on the data vendor.

CALCULATED On the 6,864 days with data in both series, New York noon differs from the ECB rate by 0.0030 dollars on average and by 0.0318 at most (December 18, 2008, a very volatile day). One figure matches no series at all: a low of 0.8115 "in January 2002" that appears in at least one technical-analysis article. In January 2002 the ECB rate never went below 0.8578, and the low for the whole of 2001 was 0.8384. This guide discards it.

The six major cycles of the euro against the dollar

CALCULATED The cycles are not textbook dates. They come from the data: turns of at least 15% between a high and a low of the daily rate, grouped into six long phases. Each one is also a dollar cycle, read the other way.

1999-2000: a weak start for the euro, a strong dollar (−30%)

CALCULATED The euro opened at 1.1789 and lost 30.0% in 22 months, to 0.8252: a pace of 17.9% a year, the fastest of the six cycles.

VERIFIED The New York Fed put the weakness down to expectations that euro-area growth would lag the rest of the world, continued investment outflows from the euro area, and concerns about exchange-rate policy and the pace of structural reform; portfolio and direct investment outflows reached €168.6 billion in 1999. On September 22, 2000 the ECB, the US authorities and the Bank of Japan intervened together in support of the euro: the ECB rate went from 0.8524 the day before to 0.8890.

HYPOTHESIS Interest rates tell almost none of this story: on the 10-year gap the "expected" move would be roughly zero. Capital flows and confidence in a brand-new currency probably mattered more, but that cannot be measured with these data.

2000-2008: the dollar's 48% slide against the euro (+94%)

CALCULATED From 0.8252 to 1.5990 in 7.7 years: the euro almost doubled and the dollar, measured in euros, lost 48.4%, its largest decline since 1999. The strongest years for the euro were 2002 (+19.0%) and 2003 (+20.4%). The dollar bounced in 2005 (euro −13.4%), and in 2006-2007 the euro rose again even though the 2-year rate gap favored the dollar (−1.48 and −1.44 points on average in 2005 and 2006).

VERIFIED In 2005, with the US current account deficit near 6% of GDP, Obstfeld and Rogoff argued that closing it would require a real fall in the dollar of 16% to 26% in their baseline long-run calculation, or more if the adjustment were abrupt.

HYPOTHESIS That the US external deficit drove much of the dollar's slide is plausible, but this guide does not prove it. What can be measured is that the 2-year gap lines up with only about 10 of the 25.5 log points the euro gained between September 2004 and July 2008.

2008-2016: the dollar as a haven, then a euro crisis (−35%)

CALCULATED From the 1.5990 high to 1.0364 on December 20, 2016: −35.2% over 8.4 years. It was not a straight line but a downtrend with three rebounds of 15% to 25% (2008-2009, 2010-2011 and 2012-2014). Nor did it end in 2015, as many timelines say: the April 2015 low (1.0552) was above the December 2016 low. 2008 was the most volatile year on record (14.3%).

VERIFIED In 2008 non-US banks scrambled for dollar funding, according to the Bank for International Settlements, which pushed the dollar up. The Fed cut rates to 0-0.25% on December 16, 2008 and on March 18, 2009 announced large-scale purchases of Treasuries and mortgage securities: the next day's ECB rate rose 4.1% (from 1.3130 to 1.3671). Draghi gave his "whatever it takes" speech on July 26, 2012, two days after that year's low (1.2089). The ECB took its deposit rate below zero from June 11, 2014 and on January 22, 2015 announced purchases of €60 billion a month; the next day's ECB rate fell 3.6% (from 1.1618 to 1.1198).

CALCULATED The policy divergence does show up in the data: the 2-year gap moved from +1.88 points (July 2008) to −0.84 (March 2015). That lines up with about 14 of the 32 points the euro lost over the period.

2016-2021: the euro recovers with rates against it (+19%)

CALCULATED From 1.0364 to 1.2338 (January 6, 2021) in four years, with the lowest volatility on record: 5.0% in 2019. This is the guide's counterexample. The 2-year gap hit its worst level for the euro in 2018 (−3.13 points on average), when US yields were far above European ones, and the euro still rose. At the average sensitivity, rates would "explain" a slight fall (−0.5%) over a stretch in which the euro gained 12%.

HYPOTHESIS Stronger European growth in 2017 and, in 2020, Fed rate cuts and the EU recovery fund may have played a part. This has not been checked against sources for this guide.

2021-2022: an aggressive Fed, an energy shock and parity (−22.5%)

CALCULATED From 1.2338 to 0.9565 in 21 months, with 47 days below parity. The 2-year gap went from −0.83 to −2.45 points and lines up with about 9 of the 20 points the euro lost.

VERIFIED The Fed started hiking on March 16, 2022; the ECB followed on July 21, four months later. In a CfM-CEPR survey of 41 European economists (January 2023), 56% named monetary policy differences as the main cause of the euro's weakness in 2022, and 29% the real economy.

HYPOTHESIS The rest of the fall fits the energy shock after the invasion of Ukraine, which raised Europe's import bill, and safe-haven demand for the dollar. That is a majority view, not a result measured here.

2022-2026: out of parity, and a dollar that did not act as a haven in 2025 (+25%)

CALCULATED From 0.9565 to 1.1974 (January 28, 2026). 2025 (+13.1%) was the euro's biggest yearly gain since 2017. The 2-year gap lines up with about 5 of the 16 points of the rise.

VERIFIED According to the ECB, after the US tariff announcement of April 2025 the dollar fell while Treasury yields rose, the opposite of what happened in 2008 and in 2020.

HYPOTHESIS This may mean the dollar has lost part of its safe-haven role. It is an open hypothesis, not a proven change of regime.

CALCULATED 2026, still open. Since the January high the euro has fallen 6.4%, to 1.1204 on October 5, also the low of the year (−4.65% since the end of 2025; the dollar is up 4.87% in euro terms). The ECB raised its deposit rate to 2.25% on June 17 and to 2.50% on September 16, but the US 2-year yield rose more (from 3.47% on average in February to 4.65% in September) and the gap widened in the dollar's favor, from −1.45 to −1.56 points. What matters is the relative gap, not whether one central bank is hiking. Until there is a 15% turn, no new cycle is called.

What drives EUR/USD

Interest rates: why the 2-year gap matters and the 3-month gap does not

CALCULATED We compared monthly moves in EUR/USD with moves in three rate gaps between the euro area and the United States. What each central bank pays today (3 months) has almost nothing to do with the euro. What the market expects from the Fed and the ECB over the next couple of years (2 years) is the closest match:

CALCULATED Applying the average sensitivity (5.5% of EUR/USD per point of gap) to each stretch shows how much of the move "fits" rates. It is an illustration, not a causal estimate:

The short version: rates get the direction, not the size. The gap has the right sign in four of the five stretches but explains only 34% to 45% of the size, and nothing in 2015-2021. By stretch, the correlation ranges from 0.25 (2015-2021) to 0.54 (2008-2015 and 2022-2026). Correlation is not causation: rates and the currency can both be moving because of a third factor, such as growth or market stress. The academic evidence partly backs this up: Engel (2016) finds that currencies with higher real interest rates tend to be stronger, by more than standard theory predicts.

The dollar as a safe haven: 2008, 2020 and the April 2025 exception

In crises the dollar has usually acted as a haven. In 2008 that came from banks' need for dollar funding; in March 2020 the euro hit its low of the year (1.0707 on March 20), and the ECB counts 2020 among the episodes in which the dollar behaved as a haven. April 2025 broke the pattern: the dollar fell during a risk-off episode, as described in cycle 6. One episode does not make a new rule, but it is the main open question for the dollar side of EUR/USD.

Growth, energy and the US external deficit

HYPOTHESIS What rates do not explain is consistent with three slower forces: growth expectations (1999-2000, 2017), Europe's energy bill (2022) and the US current account deficit (2002-2008). All three appear in the sources cited, but none has been measured here as rigorously as rates.

Why nobody can forecast exchange rates

Meese and Rogoff (1983) showed that structural exchange-rate models did not forecast better, at horizons of one to twelve months, than assuming tomorrow's rate will equal today's. That is why this guide uses rates to understand the past, not to call the future.

EUR/USD year by year, 1999 to 2026

CALCULATED Close on the last business day of each year, change on the previous close, and the year's high and low, all on the ECB reference rate. The euro rose more than 10% in six years (2002, 2003, 2006, 2007, 2017 and 2025) and fell more than 10% in four (1999, 2005, 2014 and 2015).

1999: change from the first value (January 4, 1999). 2026: to October 5. Volatility over the whole series is 9.3% a year; by year it ranges from 5.0% (2019 and 2026 so far) to 14.3% (2008).

Parity: how often and when

CALCULATED Of the 7,107 reference rates the ECB has published since 1999, 753 (10.6%) were below one dollar: 706 between January 27, 2000 and December 5, 2002 (every one in 2001), and 47 in 2022, between August 23 and November 10. None since.

VERIFIED In the market the euro touched $1.0000 on July 12, 2022 and dipped to 0.9998 on July 13, as reported at the time. The ECB rate did not cross parity in July: its low that month was 1.0005 (July 14). The first ECB rate below 1 came on August 23 (on FRED, a day earlier). Another case for sticking to one series.

What it means for you

CALCULATED For anyone who earns or invests in dollars, EUR/USD decides what euro assets are worth back home. €10,000 was worth $11,750 at the end of 2025 (1.1750) and $11,204 on October 5, 2026 (1.1204): 4.65% less, with nothing changing but the exchange rate. Over the euro's history, one dollar has bought between €0.6254 (July 2008) and €1.2118 (October 2000), almost twice as much. The same arithmetic applies to a trip to Europe.

The mirror image is a European who owns US assets. The guide to the S&P 500 in euros measures that effect year by year: in 2025 the index returned 17.78% in dollars and 4.14% in euros. This guide does not recommend hedging or not hedging currency risk; it shows what has happened.

Methodology, sources and limits

Series. ECB euro reference rate (EXR.D.USD.EUR.SP00.A), US dollars per euro, January 4, 1999 to October 5, 2026: 7,107 values. There is no rate on weekends or TARGET closing days, so "days" in this guide are days with a published rate. Cross-checked against FRED (DEXUSEU, from the Fed's H.10 release). Annual averages and year-end closes match the ECB's official annual series.

Calculations. Extremes: highest and lowest daily value. Yearly change: year-end close over the previous one, minus 1. Volatility: standard deviation of daily log changes, annualized. Cycles: turns of at least 15% between daily extremes. Rate gaps: monthly averages, euro area minus United States; 3-month: Euribor minus Treasury bills; 2-year: ECB euro-area AAA yield curve (from September 2004) minus Treasuries; 10-year: German Bund minus Treasuries. Correlations and sensitivities: monthly changes.

Limits. The 2-year gap only exists from 2004 and uses AAA-rated bonds, not just Germany's. Monthly averages smooth out extreme days. Intraday figures come from the press and wire services and are quoted as context. Correlations do not prove causes. The data describe the past: exchange rates cannot be forecast reliably.

For day-to-day moves, see our EUR/USD analysis. Another major currency against the dollar, with its own rate story, is covered in the USD/JPY guide, and the dollar's link with real yields in the gold guide. More pairs are in the currencies section; technical terms are in the glossary (in Spanish), and the site's criteria in the methodology (in Spanish).

Sources consulted on October 5, 2026: ECB: US dollar/euro reference rate, daily, ECB: euro foreign exchange reference rates, FRED: US dollars to euro (DEXUSEU), Federal Reserve: about the H.10 release, ECB: euro-area AAA 2-year yield, US Treasury: daily par yield curve rates, ECB: 3-month Euribor, FRED: 3-month Treasury bill, ECB: deposit facility rate, New York Fed: foreign exchange operations, January-March 2000, ECB: intervention, September 22, 2000, Obstfeld and Rogoff (2005), BIS: the US dollar shortage in global banking, Fed: December 16, 2008, Fed: March 18, 2009, Draghi: July 26, 2012, ECB: January 22, 2015, Fed: March 16, 2022, ECB: July 21, 2022, CfM-CEPR survey, January 2023, ECB: what safe haven after the April US tariff announcement?, Engel (2016), Meese and Rogoff (1983), summary, Börse Online: intraday, Fox News: euro timeline, Telemetro: 2008 record, Boston Globe: 2022 parity.

Educational notice: this guide is analysis based on public data, may be wrong, and is not financial advice. It describes the past: it is not a forecast for the euro or the dollar, nor a recommendation to buy, sell or exchange any currency. The decision and the risk are always the reader's own.