February 2026Monthly reportThe dollar index barely moved while the euro and yen weakened
An external retrospective on a nearly unchanged broad dollar index, weaker euro and yen endpoints, and the timing of US economic releases.
Data window: 1 February 2026–28 February 2026
On this page
February’s broad dollar index finished close to its starting point, but the euro and Japanese yen both weakened against the dollar over the selected endpoints. The apparent contradiction disappears once the measures are separated: a basket describes a weighted combination of currencies, while a pair describes an individual exchange rate.
This external market retrospective focuses on the dollar, selected currency pairs and the US economy. It contains no analysis of PineConnector instructions for this period. The distinction between the month being measured and the publication date matters throughout: January’s CPI report arrived during February; February’s inflation and employment reports arrived in March.
The basket was nearly unchanged; the pairs were not
The Federal Reserve’s nominal broad dollar index changed by −0.07% between 30 January and 27 February 2026. That was a small net decline across the selected endpoints. Calling the broad measure “nearly unchanged” is a description of this result, rather than a claim that currency markets were inactive throughout February. Federal Reserve, H.10 broad dollar index, January–February observations.
EUR/USD fell 0.53% over the same window. The euro therefore bought fewer dollars at the endpoint. USD/JPY rose 1.11%, meaning a dollar bought more yen. Both pair movements indicate dollar strength against the other currency, despite the broad index’s small decline. Federal Reserve, H.10 euro series; H.10 yen series.
reference-rate change (%)
Fed reference observations. Higher EUR/USD means a stronger euro; higher USD/JPY means a stronger dollar against the yen. Changes exclude costs and are not broker execution returns.
Source: Board of Governors of the Federal Reserve System: H.10 nominal broad dollar index, daily; Board of Governors of the Federal Reserve System: H.10 euro daily reference rate; Board of Governors of the Federal Reserve System: H.10 Japanese yen daily reference rate Download data (CSV): Figure 1
View figure data
| Measure | Period or release date | Value |
|---|---|---|
| Broad US dollar index | 2026-01-30 to 2026-02-27 | −0.07% |
| EUR/USD | 2026-01-30 to 2026-02-27 | −0.53% |
| USD/JPY | 2026-01-30 to 2026-02-27 | +1.11% |
For readers who followed only the basket, the month could look uneventful. For someone interested specifically in converting euros or yen into dollars, that summary would omit the relevant change. February illustrates why the headline measure should match the question being asked: the broad dollar’s movement and the cost of a particular currency conversion are related subjects, but they are not the same measurement.
A weighted basket cannot stand in for every currency
The broad index combines the dollar’s exchange rates against a set of trading-partner currencies. Its weights reflect trade relationships, rather than treating each currency as an equal part of a simple average. This makes it useful for examining the dollar across a wider set of relationships. It also means that the euro and yen alone cannot explain the index’s result. Federal Reserve, H.10 technical questions and answers; dollar index definitions and weights.
Different currency movements can offset within an aggregate. February’s selected results show why that possibility matters, but they do not identify which other currencies explained the difference. That would require the remaining currency observations and the index calculation. It would be an overstatement to name a particular offsetting currency from these selected pairs alone.
The quote directions deserve equal attention. EUR/USD prices a euro in dollars; a decline means a cheaper euro in dollar terms. USD/JPY prices a dollar in yen; an increase means a more expensive dollar in yen terms. A positive sign on one and a negative sign on the other can therefore tell the same story about the dollar.
Similar endpoints do not establish a calm month
A small endpoint change does not measure volatility inside the month. February’s dollar basket could finish near its starting level without every daily move being small. The comparison uses the Federal Reserve’s New York noon reference observations, rather than a universal broker closing price or an actual retail conversion rate. Federal Reserve, About H.10.
February’s inflation news described January
The CPI release published on 13 February reported January consumer-price changes. Headline CPI rose 0.2% over January on a seasonally adjusted basis, and core CPI, excluding food and energy, rose 0.3%. These were economic figures that became public during February, even though their reference month was January. Bureau of Labor Statistics, January CPI release, 13 February 2026.
That distinction is more than a date label. Someone reconstructing the information available during February should use releases that had appeared by then. Someone asking what happened to prices during February needs the next report, which arrived in March. Combining the questions would place later information into an earlier market narrative.
There was no regularly scheduled Federal Open Market Committee policy decision in February. The committee’s meeting calendar does not follow a monthly decision cycle, so an accurate monthly report need not contain a new rate decision. Federal Reserve, FOMC meeting calendar.
An absence of a scheduled decision does not imply that currencies must remain unchanged. Exchange rates can reflect changing information and expectations between meetings. The selected evidence here does not measure those expectations or establish that the February CPI release caused the monthly pair moves. It establishes the release chronology and the currency endpoints separately.
February’s own price data showed different headline and core moves
When February CPI was released on 11 March, it showed a seasonally adjusted monthly rise of 0.3% in the all-items index and 0.2% excluding food and energy. The headline monthly increase exceeded the core increase. Energy prices rose 0.6%. BLS, February CPI release, 11 March 2026.
monthly CPI-U price change, seasonally adjusted (%)
These prices describe February; the BLS release arrived the following month. They are not facts that were available throughout the month itself.
Source: US Bureau of Labor Statistics: CPI 2026-02 original news release Download data (CSV): Figure 2
View figure data
| Measure | Period or release date | Value |
|---|---|---|
| All items | Released 2026-03-11 | +0.3% |
| Excluding food and energy | Released 2026-03-11 | +0.2% |
Headline CPI describes the full consumer-price basket. Core removes food and energy to help examine changes outside those categories. The energy observation is relevant to understanding why the broader and narrower measures can differ, but it does not by itself quantify energy’s contribution to the gap. Category weights and the full calculation would be needed for that claim.
Over the year, headline CPI was 2.4% higher and core CPI was 2.5% higher, both on an unadjusted basis. The annual measures compare with a year earlier; the monthly measures describe the latest step after adjustment for recurring seasonal patterns. They should not be substituted for one another. BLS, February CPI release and tables.
A positive inflation rate means the measured price level rose over the comparison period. It says neither that every price rose nor that every household experienced the same increase. For this retrospective, February CPI helps describe the economy during the month. Its March publication date prevents treating it as information that was already available at February’s FX endpoint.
The later jobs report added context, not a cause
The initial February employment report, published on 6 March, recorded a nonfarm payroll change of -92,000 jobs and an unemployment rate of 4.4%. The negative payroll change means the estimate showed fewer nonfarm payroll jobs than in the preceding month. These are figures from the dated initial release, which can differ from later revised estimates. BLS, February Employment Situation, 6 March 2026.
Payrolls and the unemployment rate are complementary measures. Payrolls count jobs in the establishment survey. The unemployment rate comes from the household survey and expresses unemployment relative to the labour force. A change in payrolls cannot be read as an identical change in the number of unemployed people.
February’s useful takeaway is that an average can conceal a meaningful difference in the relationship a reader cares about. The broad dollar was nearly unchanged, while the euro and yen weakened against it in the selected comparison. Read the relevant pair, check the quotation direction and keep economic release dates in order. That produces a clearer picture of the month than either a basket headline alone or a story built with information that arrived later.