US Publications
Below is a list of our US Publications for the last 5 months. If you are looking for reports older than 5 months please email info@pantheonmacro.com, or contact your account rep
Please use the filters on the right to search for a specific date or topic.
Datanotes Daily Monitor Samuel Tombs
- The first ADP payroll estimate is among the worst indicators of both initial and benchmarked payroll data.
- The final data line up better, but only because ADP re-weights its data after benchmarking by the BLS.
- The Treasury’s method for inferring the CPI without BLS data implies a 0.36% monthly rise in October.
- The manufacturing sector has seen little benefit from the new tariffs so far this year…
- …Recent gains in output have been limited to a few industries that dance to the beat of their own drum…
- …Industrial policies have a role to play in reviving USmanufacturing, but tariffs are a blunt tool.
- We calculate tariffs have lifted core PCE inflation by 0.4pp, below Mr. Powell’s “five to six tenths” estimate.
- Pass-through, however, is probably just over half complete, and services inflation will fall next year.
- The looming suspension of SNAP benefits could hit GDP by 0.2% if paused through the end of Q4.
- Chair Powell has jolted markets by saying a December easing is “not a foregone conclusion, far from it”...
- ...But most hiring indicators still point to near-stagnant payrolls; post-shutdown data will spur more easing.
- October’s regional Fed surveys point to flat employment demand and slower wage growth ahead.
- Tariff revenues continue to underwhelm; the ending of the de minimis exemption has been uneventful.
- Accordingly, we are shaving 0.1pp off our forecast for the peak in core PCE inflation in December.
- Charts implying a dramatic rise in “different cell” imputation overstate the decline in data quality.
- The year-to-date change in Homebase’s measure of employment is almost identical to last year...
- ...But this also was true in the summer, when payrolls slowed decisively; we track other indicators instead.
- Canada CPI data point to risk of a big increase in US food at home prices in September.
- The regional Fed and PMI surveys are no better at forecasting GDP than just extrapolating the trend.
- Durables goods spending by consumers is reasonably well signalled by the UoM confidence survey.
- Airline passenger and hotel occupancy data are useful for forecasting that segment of spending only.
- The weakening dollar means that DXY is no longer overshooting its long-term link with Treasury yields.
- ...But further fiscal easing and politicization of the Fed are key downside risks for the dollar in 2026.
- Housing inflation likely has further to fall, given the renewed drop in rental growth in recent months.
- Homebase data point to steady employment growth, and WARN data indicate layoffs remain low...
- ...But Indeed job postings are falling at a faster pace, and Empire State hiring intentions have weakened.
- High mortgage rates and consumers’ low confidence imply higher homebuilder optimism won’t last.
- Corporate balance sheets look healthy in aggregate; private credit is a small and stable part of the picture.
- Mortgage refinancing is continuing to reverse its mid-September surge; expect low levels next year too.
- The Empire State survey signals renewed impetus in factory gate inflation; fingers crossed it’s an outlier.
- We expect a 0.4% rise in the headline CPI—below the 0.5% priced into swaps—and a 0.3% core print.
- Core goods prices likely were boosted again in September by the tariffs, including new vehicle prices.
- Residual seasonality will lift services prices, but the rebound in airline fares is over, and rent is cooling.
- September’s payroll report likely will be released about three working days after the shutdown ends.
- October payrolls will be unaffected by the shutdown, but the unemployment rate will be lifted by 0.2pp.
- The rotation of the regional Fed voters implies a slight hawkish shift in the FOMC early next year.
- AI capex—net of tech imports—lifted H1 GDP growth by an annualized rate of around 0.3pp.
- The boost to spending due to the wealth effect from surging tech stocks likely has been similar.
- That suggests to us that weaker growth is more likely than a recession if the AI boom turns to bust.
- The NY Fed survey suggests the mood among consumers was souring again even before the shutdown.
- The weak labor market and further upward pressure on inflation from tariffs are the most likely culprits.
- Alternative indicators of payrolls are even worse guides to the final estimates than the initial prints.
- Indicators from Revelio, QuickBooks and Paychex are all essentially useless guides to official payrolls.
- Combining NFIB, Conference Board and regional Fed survey data is the only way to beat the consensus.
- We look for a 75K rise in September private payrolls, above these surveys, due to residual seasonality.
- The impact of AI on labor demand so far looks small, even for the most at-risk occupations.
- The payroll slowdown this year has far more to do with trade and immigration policies.
- Auto sales are set to weaken, as an EV tax credit expires and tariffs start to push up prices.
Drops in the openings-to-unemployment ratio and quits signals slower wage growth ahead.
- The government shutdown will hold up key data releases and likely will drag on economic growth.
- Another 25bp easing from the Fed at its next meeting seems like prudent risk-management.
- The effective tariff rate has now crept up to just 12%, and a further climb is likely in the next few months.
- JOLTS openings ticked up slightly in August, but the underlying trend in labor demand still looks weak.
- Conference Board’s labor market numbers point to stagnant payrolls and higher unemployment.
- The shifting balance in the labor market points to weaker underlying wage growth ahead.
- Reliable surveys point to September payrolls rising at a similarly slow pace as the past couple months.
- Seasonal problems signal a jump in hospitality jobs, but federal policies likely weighed on education jobs.
- The unemployment rate likely crept up, while a calendar quirk probably dampened average earnings.
