The dollar rose against all the G10 currencies last week except for the Swiss franc. The gains were concentrated against the dollar-bloc currencies and the Scandis. Neither sterling nor the Australian dollar drew comfort from their central bank rate hikes. Nor for that matter was the Mexican peso pushed, except for the initial reaction, to the unexpected rate cut.
The US midterm election results boosted two narratives. The first was on limits to data centers. The second was a more optimistic of ending the war with Iran. The Philadelphia Stock Exchange Index fell by almost 12.5% last week. Even if one wants to link part of that to the election, note that it was the fifth consecutive weekly decline. The second narrative helps explain the nearly 13.5% drop in the December WTI contract, and it snapped a five-week rally. The 10-year US Treasury yield fell for the fifth consecutive week. It may be difficult to extend decline much further, with strong jobs report in hand, only a new shock can deter the Fed from hiking in December.
US
Drivers: The Dollar Index has become more correlated (30 days) with changes in the US 10-year yield than changes in the two-year yield (~0.47 vs. 0.33). DXY’s 30-day correlation with changes in the Philadelphia’ Stock Exchange’s Semiconductor Index has been inverted this year, except for around three weeks in late January through early February, and again briefly in late July. It has been positively correlated again over the last couple of weeks and is approaching the year’s high (~0.20). The stronger than expected October jobs report offset the decline in stocks and cracks in the AI meme and reinforced expectations of a December hike.
Data: Of course, the CPI and PPI are the week’s highlights, but are they very meaningful? The Fed’s use of the phrase “the accumulation of more information” appears partly aimed at directing the public and businesses away from putting too much emphasis on any single data point. Second, after the October jobs report, the market has little reason to second guess itself over the likelihood of a hike at the next FOMC meeting in December. Because of the government shutdown last October, the BLS will not report a year-over-year CPI this week. Headline US CPI has risen at a 4.4% annualized pace over the past three months. The core rate has risen at a 3.2% annualized clip in July-Sept period, which is a little faster than in the previous three-month period. The median forecast in Bloomberg’s survey is for a 0.3% rise in the October headline rate, which would still see the 3-month annualized pace accelerate. The median forecast for the core rate is 0.2%, which if accurate would keep the three-month annual pace steady. Producer prices are rising faster. Even with food and energy excluded PPI rose 5% year-over-year in September. The headline rate was 6.3% and new cyclical high. It was rising at a 3.1% year-over-year pace in the three months before the Middle East war disruption.
Prices: The Dollar Index was bid before the US jobs report and extended its gains after it. DXY has risen above the down trendline connecting mid- and late October highs, found near 102.20 before the weekend. A move above 102.50 signals a test on the year’s high (~103.00). Given the positioning of the momentum indicators, there may be scope toward 104.00-104.50 in the coming weeks.
EMU
Drivers: Over the past 30 sessions, changes in the euro have been less inversely correlated to changes in the US 10-year yield (~-0.33) as it is to French premium over Germany’s 10-year rates, which is around 0.48. The correlation between changes in the exchange rate and the US-German two-year rate differential is around -0.42. While changes in the euro are positive correlated with changes in Germany’s two-year yield (~0.22), it remains inversely correlated, albeit slightly, with changes in Germany’s 10-year yield (~-0.14).
Data: With the update of Q3 GDP more details will be provided. Without fresh impetus, year-over-year growth looks stuck around 1.0%-1.5%. Uninspiring. The September trade figures may be more interesting on November 13. With little fanfare, the eurozone reported a blowout deficit in August of $16.1 bln euros. It was the largest in a few years. And notably it swung the trade balance into deficit for the first eight months of the year (~5.4 bln euros) compared with a surplus of ~90.8 bln euros in the Jan-Aug period of 2025 and 104 bln euro surplus in the same 2024 period.
Prices: The euro reached nearly $1.1275 in the knee-jerk reaction to the US midterm elections. We were skeptical and anticipated it would set back ahead of the US jobs data. The jury was still out until the stronger than expected US employment data saw the euro fall to almost $1.1130 before the weekend. Recall that the euro reached almost $1.1100 in late October, which more or less coincides with the (50%) retracement of the rally from the Feb 2025 low (~$1.0140). The euro does not appear to have bottomed. A break of $1.11 could target the $1.10 area but the next retracement target is around $1.0880.
PRC
Drivers: By increasing the reserve requirements for foreign currencies, the PBOC signaled an end to the one-way appreciation of the yuan. The yuan strengthened without interruption in the last dozen sessions in October. The 2% increase the reserve requirement is substantive in the sense that it will lock an estimated $20-$25 bln of liquidity at the banks, but it is also sending a clear signal. The dollar rose against the yuan for the first week in six weeks. If the PBOC has put a cap on the yuan, then the given rates, it could be an attractive funding currency.
Data: It is an important week for Chinese data. Although many participants may be skeptical of the accuracy of the economic data, they do seem to take the CPI and trade figures at face value. Both will be reported in the coming days. The general narrative, or the judgement made by what the Federal Reserve might call the “accumulation of information,” will not change. China’s consumer inflation is lower. The consensus conclusion is that this reflects weak demand. Yet, on per capita basis, consumption is rising, and investment has risen faster. Moreover, everyone might call their index the same thing (CPI), it is measured differently. For example, in China’s CPI, food has weight about 25% less in than in the US measure, and weight China gives to shelter is less than half of what the does. The trade story cannot change much. Yes, China has a large trade and still growing trade surplus. However, its trade figures belie arguments about de-globalization–in September Chinese exports were up by 36.3% year-over-year, and imports were up 31.%.
Prices: It is difficult to know the intent of Beijing, but we suspect the market will be content to job the dollar between CNH6.65 and CNH6.70. We will continue to pay close attention to the PBOC’s daily fix. It rose last week, albeit slightly, for the first time in nearly three months.
Japan
Drivers: The dollar-yen exchange rate has become more sensitive to changes in US interest rates. The correlation between changes in dollar-yen and the US 10-year yield is a little below 0.60. It reached the high for the year slightly higher in mid-June (~0.65). Recall that in early August, it was briefly inversely correlated. The correlation between the changes in the exchange rate and the US two-year yield recorded the high for the year, near 0.70 in late October. It is slightly now about 0.65. In early August, it was at a six-month low below 0.20. The dollar-yen exchange rate is inversely correlated, as one might intuit, with changes in Japan’s two-year yield, but we would note that it is modest around -0.20 and it spent most Q2 with a positive correlation. The two-year yield is a proxy for the medium-term policy outlook. Over the past week, the swaps market downgraded the chances of a BOJ hike next month to about 52% from a little more than 80% a week ago.
Data: Japan reports September trade and current account figures on November 10. Before 2015, there was a seasonal pattern of improvement in the September. It has broken down and over the past several years, it has been nearly 50/50. However, this year’s trend has been higher. In the first eight months of the year, Japan recorded an average monthly current account surplus of JPY2.37 trillion (~$15 bln), a little more than 20% higher than the average in the January-August 2025 period. Yet, Japan continues to report a trade deficit, and this is despite an exchange rate that is historically cheap on numerous metrics. Through August, Japan recorded a nearly JPY410 bln trade deficit. That said, the deficit is gradually being reduced and will likely swing back into surplus next year. The deficit in the first 8 months of 2025 was about JPY1.54 trillion and JPY2.68 trillion in the same period in 2024. On November 10, Japan also reports September household spending. Despite real wage growth this year, household consumption has not risen on a year-over-year basis since last November. The 3.1% contraction in August (year-over-year) was the smallest in three months. The correlation between the year-over-year change in retail sales and household consumption is not particularly robust (less than 0.2 over the past 36 months), but we know that retail sales slowed to 2.3% year-over-year in September from 2.9% in August. Japan reports October PPI on November 12. It fell by 0.3% in both August and September. Producer prices likely rose in October but given the base effect (0.5% in October 2025), the year-over-year pace may have moderated for the third consecutive month. It peaked in July at 7.7% year-over-year and was at 6.8% in September, a four-month low.
Prices: The dollar made a marginal new low for the week against the Japanese yen near JPY157.25 before the US jobs report. Afterwards, the greenback was bid slightly above Thursday’s high (~JPY158.50) before it consolidated and settled strongly. The dollar is butting up against the upper end of a trading range that has persisted for more than six weeks. That area also corresponds with the halfway mark of the range from the four decades high in July (~JPY164) to the September low (~JPY152.90). A break higher spur a move toward JPY160, ahead of which Japanese officials may verbally intervene.
UK
Drivers: With a couple of exceptions, since early April the rolling 30-day correlation between changes in sterling and the two- and 10-year UK yields have been inverse. The inverse correlations are now about -0.15 and -0.40, respectively. Sterling is a little inversely correlated with changes in the US two- and 10-year rates (-0.15 and -0.28). The EU and UK summit is on November 20, and regardless of the eventual destination, the path of travel is clear, relations with Europe will deepen and broaden again. While the UK will continue to insist on maintaining sterling, the correlation between the euro and sterling high–since late January, the 30-day correlation has not been below 0.75. It is now a little above 0.90. The 100-day correlation is slightly below 0.85. It has not been below 0.80 since the end of Q1 26.
Data: The UK reports Q3 GDP on November 12. The British economy is seen to be gradually slowing. It grew 0.6% in Q1 and 0.5% in Q2. It is expected to have grown by 0.4% in Q3 and projected to slow to 0.2% in Q4 (median forecasts in Bloomberg’s survey). The monthly GDP reports do not add up to the quarterly figure. In Q2, the cumulative monthly growth was 0.3% and in Q1 it was 1.0%. Still, for more context, the UK economy grew 0.4% in July and 0.1% in August. September details will also be provided. After last week’s hike, the market has about a 55% chance of a follow-up hike at the next meeting on December 17, which seems somewhat high. Subjectively, we suspect it may be closer to a 1-in-3 proposition.
Prices: Sterling briefly traded above $1.35 last week for the first time since mid-September. Perhaps sterling fell victim to buy the rumor of a rate hike and sell the fact when it was delivered. Sterling settled lower on Thursday, when the BOE hiked, and took a bigger leg down ahead of weekend, slightly below $1.3345. Sterling settled below the trendline connecting the mid- and late-October lows, found ~$1.3380 on Monday. Nearby support may be around $1.3320, the (50%) retracement of sterling’s bounce from the mid-October low, near $1.3135. However, more telling will be whether $1.3275 area, the low from late October holds.
Canada
Drivers: The typical driver of the exchange rate has slackened. The rolling 30-day correlation of the changes in the Dollar Index and the greenback against the Canadian dollar has eased from 0.80 in the second half September, the high for the year, to around 0.70 now. In terms of US interest rates, the exchange rate is more correlated over the past 30 sessions with the US 10-year yield and two-year yield have converged near 0.40. The correlation between the exchange rate and the two-year differential is also around 0.40. The 30-day correlation of the exchange rate and changes in the US 10-year premium over Canada has eased form over 0.60 in mid-September to a little below 0.45 now. The swaps market was already downgrading the chances of a December rate hike before the disappointing jobs data ahead of the weekend. The odds fell to about 37% from a little more than 50% a week ago, which still strikes us a bit elevated.
Data: Canada’s economic reports feature September building permits, manufacturing sales, and wholesale sales, which are not typically market movers.
Prices: On the last trading day in October, the US dollar fell to about CAD1.3940, which met the (61.8%) retracement target of its rally from late August (~CAD1.3730) to the mid-October high (e~CAD1.4300). It made a marginal new low in the immediate post-US mid-term election sell-off but recovered smartly and reached almost CAD1.4140 ahead of the weekend after the diverging employment reports. The next target is in the CAD1.4160 area, but we suspect there may be potential to retest the October high, given the positioning of the momentum indicators and a greater sense the widest policy rate gap in 30 years may persist longer than anticipated.
Australia
Drivers: Over the past 30 sessions, the inverse correlation between changes in the Australian dollar and the Dollar Index has moderated around -0.25, the least since November 2025. It bottomed near -085 in May. The 30-day correlation between the exchange rate and the US two-year yield has turned positive (~0.25) for the first time since late March. It bottomed in June around -0.82. Meanwhile, the Aussie’s correlation with changes Australia’s two-year yield has risen to nearly 0.40, the most this year, while the correlation with changes in Australia’s 10-year yield has jumped to almost 0.33 from a small inversion around the middle of October. It peaked this year in August near 0.45. The Aussie’s 30-day correlation with changes in gold peaked a little below 0.90 in June and has trended lower and is now near zero. It was last inverted in July 2025, albeit briefly.
Data: After last week’s rate hike, Australia’s economic calendar turns light. A couple of bank confidence surveys and some data on home loan and investor loan values. Melbourne Institute Inflation Expectations (trimmed mean) will be reported on November 12. It was at 4.7% in July, the lowest since January, and rose to 4.9% in August and September, before it rose to 5.3% in October. The futures market sees about a 2/3 chance that last week’s hike ends the tightening cycle.
Prices: The Australian dollar was tagged for about 2.2% last week, its worst week since Liberation Day week in April 2025. It would appear that the macro developments were supportive; the central bank hiked for the fifth time this year, the final composite PMI was revised up (though still below the 50 boom/bust level), household spending was more resilient than expected, and the September trade surplus was nearly three times larger than expected. The Aussie was sold to $0.6855 ahead of the weekend, its lowest since the end of Q1 26. The low set in March was closer to $0.6835, but the downside risk may extend toward $0.6760. The technical caveat is that the Aussie settled well below the lower Bollinger band (~$0.6900).
Mexico
Drivers: The Mexican peso has become somewhat unmoored to its usual drivers. Even the usually robust correlation with the JP Morgan Emerging Market Currency Index has slackened. It was the weakest since January (slightly below 0.60). The correlation of the dollar’s changes against the peso and changes in the Dollar Index is near 0.20, which is also the weakest since January. The exchange rate’s correlation with changes in the US two-year yield is slightly inverses for the first time since March. It peaked in June above 0.80. The correlation of the changes in the exchange rate and Mexico’s two-year yield is near zero. It peaked in April, near 0.70. The exchange rate’s inverse correlation with changes in the S&P 500 has also moderated. It is now near -0.25, around the least in nine months. It was around -0.85 in April.
Data: Mexico’s central bank surprised most market participants, including ourselves, by delivering a quarter-point rate cut last week. It did sound dovish at its previous meeting, but many seemed to play it down by putting in the context the central bank trying to secure some degrees of freedom from the Federal Reserve. The 3-2 vote seemed to be spun as a one-off adjustment. Its forward guidance was for an extended pause. That means that this week’s October CPI is unlikely to have much impact on the central bank’s reaction function. Still, the risk is that the October CPI accelerates a little from September 3.45% and 3.75% readings for the headline and core, respectively. Prices did rise faster in the first half of October. The economy is not growing to potential and the 0.9% quarter-over-quarter expansion after 1.4% in Q2 represents a near-term peak. The median forecast in Bloomberg’s survey sees growth slowing to 0.2% this quarter before recovering to 0.5% in Q1 27. September industrial production will be reported on November 11. Through August, it rose by an average of 0.25 compared with an average of a 0.03% contraction in the first eight months of 2025 and -0.10 average in the first eight months of 2024. Yet, this may overstate the demand in the domestic economy. Mexico’s exports have risen by about 36% year-over-year in the first nine months of the year. Yet, Mexico’s auto exports have actually declined slightly this year. Through September vehicle exports have slowed to 2.53 mln from 2.57 mln. Mexico’s vehicle production has also slowed. Through September, Mexico produced 2.95 mln vehicles (exports the vast majority) compared with 3.03 mln in the Jan-Sept 2025 period. A few countries’ ratio of vehicle export to production rival Mexico’s (Canada, Spain, and Slovakia). China’s sheer size is source of its disruption. It exports a more modest 20% (compared with ~80%-90%) of its vehicle production or around 6 mln vehicles (and rising).
Prices: The dollar traded between a little below MXN18.53 and MXN18.87 last week, moving in broad intraday ranges. It has been in this range for about two-and-a-half weeks. The technical question is whether this is a topping pattern or a nesting before another leg up. We had previously thought there was potential into the MXN19.00-19.0750 area. A convincing break of the MXN18.50 area, which would likely coincide with the five-day moving average fall below the 20-day moving average for the first time since mid-September, would suggest a dollar high is in place. The median forecast in Bloomberg’s latest survey is for the dollar to end the year at MXN18.3450 (the indicative forward is around MXN18.7050). We suspect the dollar can be even lower (~MXN18.00-MXN18.20).
Disclaimer
