US oil output at a record high; US oil supplies at 42½ year low; refinery utilization at the highest October rate since 2003

record production of crude from US wells; Strategic Petroleum Reserve is the lowest since it was initially being filled in mid-October 1982, leaving the total of all US oil supplies, including the SPR, at the lowest level since March 30th, 1984; US refinery utilization at the highest October rate since 2003, refining the most barrels in any October week since 2019

US oil prices finished higher for the first time in three weeks on increased attacks on oil tankers in the Strait of Hormuz and as the first Gulf hurricane of the year forced shut a quarter of U.S. offshore oil production…after falling 1.4% to $91.11 per barrel last week as shipments of oil thru the Strait of Hormuz increased and Saudi Arabia resumed Red Sea oil exports after repairing its damaged East-West pipeline, the contract price for the benchmark US light sweet crude for November delivery edged lower during Asian trading on Monday, as rising Middle East crude exports and and a release of oil stocks by the Group of Seven ‌nations boosted supplies, offsetting concerns about further damage to Gulf oil infrastructure amid the US-Israeli war on Iran, and was down about 1% during morning trading in London, amid rising Middle East exports and the decision by the G7 countries to release fuel from strategic reserves, but steadied as trading began in New York as the market weighed signs of Middle Eastern supply recovery and the announcement of a large-scale release from emergency reserves against a flare-up in fighting in Yemen and continued attacks on tankers in the Strait of Hormuz, before resuming its slide to settle $1.68 lower at $89.43 a barrel after crude exports from the Middle East increased and the Group of Seven nations pledged to boost supplies, even as selling was limited by ‌ongoing disruption linked to the US war with Iran….oil futures traded higher on Tuesday morning in Asia following reports that the Houthis had attacked several sites in Saudi Arabia, but were down nearly 2% by the time trading wrapped up in Europe, as supply concerns eased due to increased Middle Eastern crude exports, and due to the agreement by the Group of Seven countries to release emergency diesel and crude stockpiles, but bounced off their low during US trading and retraced the earlier losses ahead of the close, as the continuing geopolitical risk limited the market’s losses, and settled one cent higher at $89.44 a barrel as the market weighed increases in Middle Eastern crude exports and a planned ‌Group of Seven release of emergency diesel and crude stockpiles against supply concerns related to attacks by the Iran-backed Houthis in Yemen….oil prices rose during Asian trading on Wednesday as the market weighed supply constraints from a storm heading for US oil-producing regions, and as attacks by Yemen’s Houthis on Saudi targets raised disruption risks, offsetting increased increased West Asian crude supplies, but fell during morning US trading despite an EIA report that the crude draw for last week was even bigger than what API reported, and far worse than the inventory increase that was expected, and settled $1.16 lower at $88.28 a barrel after the International Energy Agency’s member states agreed to prioritize the release of diesel stocks in an effort to address surging fuel prices…oil prices rose in early Asian trading on Thursday as concerns over supplies from the Middle East persisted amid increasing attacks on shipping traffic in the Gulf and the Strait of Hormuz, and were later up more than 4% across global markets after reports suggested that the White House had asked the Pentagon to develop options to strike Iran before the midterm elections, then jumped during Thursday morning in New York amid growing supply risks as Iran was stepping up attacks on tankers in the Strait of Hormuz, and the first Hurricane of the season forced shut a quarter of U.S. offshore oil production, and settled $3.21 higher at $91.49 a barrel on continuing concerns over supply from the Middle East amid an increase in attacks on shipping in the Strait of Hormuz….oil prices fell during Asian trading on Friday, as traders weighed President Trump’s pledge not to attack Iran before the U.S. midterm elections against persistent shipping risks in the Persian Gulf and extensive offshore production shut-ins in the Gulf of Mexico, and slid further Friday morning in New York after President Trump touted “productive discussions” with Iran, and the news that China was set to resume fuel exports following a one-week suspension, but recovered to settle 36 cents higher at $91.85 a barrel as Hurricane Isaias churned toward the northern Gulf of Mexico and ‌oil companies shut more than 70% of crude production in US waters, leaving oil prices 0.8% higher for the week….

meanwhile, natural gas prices finished higher for the third time in four weeks after Hurricane Isaias idled production but left LNG exports unscathed….after falling 5.7% to $3.035 per mmBTU last week on milder weather and a quick return to normal Appalachian pipeline flows after the prior week’s outage, the price of the benchmark natural gas contract for November delivery opened a penny higher on Monday, then traded near $3.060 into the afternoon, as steady LNG exports and healthy production kept the market in check, and settled 3.1 cents higher at $3.066 per mmBTU as slipping production outweighed mild October forecasts and sagging cash prices in the East…natural gas prices were little changed in morning trading Tuesday, as traders took note of weakening weather demand, LNG strength and mixed reads on supply, then moved higher through early afternoon trading as another production estimate kept supply concerns in focus, even as the weather outlook offered mixed support for demand, and settled 4.8 cents higher at $3.114 per mmBTU on tightening supply-demand balances, despite largely unsupportive two-week weather forecasts….natural gas futures surged higher in Wednesday morning action​, as traders digested light production estimates and the potential for Tropical Storm Isaias to wreak havoc and further curb supply during the week, then rose further through midday trading as cooler longer range forecasts, expectations for a below-normal storage injection and short covering lifted the front of the curve. and settled 8.9 cents higher at a two-week high of $3.203 per mmBTU on a drop in output due to recent pipeline problems​, and ​on worries that ​t​he Gulf of Mexico storm could cause further supply disruptions…natural gas prices moved higher early Thursday as traders took stock of Hurricane Isaias approaching the Gulf Coast, and expectations for a bullish government inventory report, but slipped in trading immediately after the report, which fell short of historical norms but exceeded market expectations, and settled 3.5 cents lower at $3.168 per mmBTU as the slightly bearish government storage surprise tempered bullish sentiment in the natural gas futures market, ending a four-day run of gains for the prompt month….natural gas futures pulled back early Friday as Hurricane Isaias neared the Gulf Coast, idling offshore production but looking to mostly steer clear of the Texas-Louisiana LNG export complex, but began to recover by midday from Thursday’s storage-driven sell-off amid declining domestic production and strengthening LNG feedgas, and settled 5.2 cents higher at $3.220 per mmBTU, as offshore operators shut in 1.26 billion cubic fet of production with a powerful hurricane poised to smack the northern Gulf Coast, leaving November natural gas prices 6.1% higher for the week…

The EIA’s natural gas storage report for the week ending October 2nd indicated that the amount of working natural gas held in underground storage rose by 85 billion cubic feet to 3,500 billion cubic feet by the end of the week, which left our natural gas supplies 130 billion cubic feet, or 3.6% below the 3,630 billion cubic feet of gas that were in storage on October 2nd of last year, but 68 billion cubic feet, or 2.0% above the five-year average of 3,336 billion cubic feet of natural gas that had typically been in working storage as of the 2nd of October over the most recent five years….the 85 billion cubic foot injection into natural gas storage for the cited week was above the 80 billion cubic foot injection into storage that the market had been expecting ahead of the report, and it was more than the  77 billion cubic foot of gas that were injected into natural gas storage during the corresponding week of 2025, but less than the average 96 billion cubic foot injection into natural gas storage that had been typical at the beginning of October over the past five years…

The Latest US Oil Supply and Disposition Data from the EIA

US oil data from the US Energy Information Administration for the week ending October 2nd showed that with a big jump in our oil exports, an increase in our oil refining, and an increase in demand for oil that the EIA could not account for, we had to pull oil out of our stored crude supplies for the the twenty-first time in twenty-four weeks, and for the 43rd time in seventy-one weeks, despite a big increase in our oil imports…. Our imports of crude oil rose by an average of 1,142,000 barrels per day to average 6,840,000 barrels per day, after falling by an average of 179,000 barrels per day during the prior week, while our exports of crude oil rose by an average of 1,195,000 barrels per day to average 4,765,000 barrels per day, which, when used to offset our imports, meant that the net of our trade of oil worked out to an import average of 2,075,000 barrels of oil per day during the week ending October 2nd, an average of 53,000 fewer barrels per day than the net of our imports minus our exports during the prior week… At the same time, transfers to our oil supplies from Alaskan gas liquids, from natural gasoline, from condensate, and from unfinished oils were 16,000 barrels per day higher than the prior week at 447,000 barrels per day, while during the same week, production of crude from US wells was 24,000 barrels per day higher at a record high of 13,979,000 barrels per day.  Hence, our daily supply of oil from the net of our international trade in oil, from transfers, and from domestic well production appears to have averaged a total of 16,501,000 barrels per day during the October 2nd reporting week…

Meanwhile, US oil refineries reported they were processing an average of 16,480,000 barrels of crude per day during the week ending October 2nd, an average of 223,000 more barrels per day than the amount of oil that our refineries reported they were processing during the prior week, while over the same period, the EIA’s surveys indicated that a​n average net of 567,000 barrels of oil per day were being pulled out of the supplies of oil stored in the US… So, based on that reported & estimated data, the crude oil figures provided by the EIA appear to indicate that our total working supply of oil from storage, from net imports, from transfers, and from oilfield production during the week ending October 2nd averaged a rounded 589,000 more barrels per day than what our oil refineries reported they used during the week.  To account for the difference between the apparent supply of oil and the apparent disposition of it, the EIA just plugged a [ -589,000 ] barrel per day figure onto line 16 of the weekly U.S. Petroleum Balance Sheet, in order to make the reported data for the supply of oil and for the consumption of it balance out, a fudge factor that they label in their footnotes as “unaccounted for crude oil”, thus indicating there must have been a error or omission of that amount in the week’s oil supply & demand figures that we have just transcribed…. Since 238,000 barrels per day of demand for oil could not be accounted for in the prior week’s EIA data, that means there was a 351,000 barrel per day difference between this week’s oil balance sheet error and the EIA’s crude oil balance sheet error from a week ago, and hence the changes to supply and demand from that week to this one that are indicated by this week’s report are somehow off by that much, and are therefore pretty useless….However, since most oil traders react to the figures in these weekly EIA reports as if they were gospel, and since these weekly figures therefore often drive oil pricing and hence decisions to drill or complete oil wells, we’ll continue to report this data just as it’s published, and just as it’s watched & believed to be reasonably reliable by most everyone in the industry…(for more on how this weekly oil data is gathered, and the possible reasons for that “unaccounted for” oil supply, see this EIA explainer….also see this March 2023 twitter thread from an EIA administrator addressing these ongoing weekly errors, and what they had once hoped to do about it).

This week’s rounded 567,000 barrel per day average decrease in our overall crude oil inventories came as an average of 455,000 barrels per day were being pulled out of our commercial stocks of crude oil, while 112,000 barrels per day were being pulled out of our Strategic Petroleum Reserve, the twenty-eighth consecutive Iran war related withdrawal from the SPR, which left the SPR level at 282,983,000 barrels, the lowest since it was initially being filled in mid-October 1982……..with both commercial oil and the SPR both down sharply this week, that left the Total of all US Oil Supplies at 707,117,000 barrels, down 18.8% from 870,774,000 barrels on April 17th, and the lowest since March 30th, 1984….

Further details from the weekly Petroleum Data Overview (pdf) indicated that the 4 week average of our oil imports rose to 6,368,000 barrels per day last week, which was 4.3% more than the 6,106,000 barrel per day average that we were importing over the same four-week period last year, while the four week average of our exports rose to 4,112,000 barrels per day last week, which was 3.8% less than the 4,276,000 barrel per day average that we were exporting last year year at this time… This week’s crude oil production was reported to be 16,000 barrels per day higher at a record high of 13,979,000 barrels per day as the EIA’s estimate of the output from wells in the lower 48 states was 3,000 barrels per day higher at 13,487,000 barrels per day, while Alaska’s oil production was 21,000 barrels per day higher at 489,000 barrels per day…US crude oil production had reached a pre-pandemic high of 13,100,000 barrels per day during the week ending March 13th 2020, so this week’s reported oil production figure was 6.7% higher than that of our pre-pandemic production peak, and was also 44.1% above the pandemic low of 9,700,000 barrels per day that US oil production had fallen to during the third week of February of 2021.

US oil refineries were operating at 92.7% of their capacity while processing those 16,480,000 barrels of crude per day during the week ending October 2nd, up from 92.5% the prior week, and the highest October utilization rate since 2003….the 16,480,000 barrels of oil per day that were refined that week were ​the most barrels refined in any October week since 2019, 1.1% more than the 16,297,000 barrels of crude that were being processed daily during the week ending October 3rd of 2025, and 5.3% more than the 15,656,000 barrels that were being refined during the pre-pandemic week ending October 4th, 2019, when our refinery utilization rate was at 85.7%, which was well below the pre-pandemic normal utilization rate for this time of year, probably due to the historic flooding across southeast Texas in the wake of Tropical Storm Imelda of that year…

Even with the increase in the amount of oil that was being refined this week, gasoline output from our refineries was still lower, decreasing by 117,000 barrels per day to 9,349,000 barrels per day during the week ending October 2nd, after our refineries’ gasoline output had decreased by 124,000 barrels per day during the prior week… This week’s gasoline production was 4.1% less than the 9,753,000 barrels of gasoline that were being produced daily over the week ending October 3rd of last year, and 7.1% less than the gasoline production of 10,066,000 barrels per day seen during the prepandemic week ending October 4th, 2019….on the other hand, our refineries’ production of distillate fuels (diesel fuel and heat oil) increased by 287,000 barrels per day to 5,290,000 barrels per day, after our distillates output had decreased by 156,000 barrels per day during the prior week.  After that production increase, our distillates output was 2.3% more than the 5,169,000 barrels of distillates that were being produced daily during the week ending October 3rd of 2025, and 9.4% more than the 4,835,000 barrels of distillates that were being produced daily during the pre-pandemic week ending October 4th, 2019….

Even with this week’s decrease in our gasoline production, our supplies of gasoline in storage at the end of the week rose for the eighth time in thirty-four weeks, increasing by 382,000 barrels to 204,744,000 barrels during the week ending October 2nd, after our gasoline inventories had decreased by 1,684,000 barrels to a twelve year low during the prior week.  Our gasoline supplies rose this week even though the amount of gasoline supplied to US users rose by 80,000 barrels per day to 8,769,000  barrels per day, because our exports of gasoline fell by 101,000 barrels per day to 831,000 barrels per day, and because our imports of gasoline rose by 12,000 barrels per day to 512,000 barrels per day.. After fifty-six gasoline inventory withdrawals over the past eighty-five weeks, our gasoline supplies were 6.5% lower than last October 3rd’s gasoline inventories of 219,093,000 barrels, and about 6% below the five year average of our gasoline supplies for this time of year…

Even after this week’s increase in distillates production, our supplies of distillates fell for the sixteenth time in thirty-four weeks, decreasing by 42,000 barrels to 105,138,000 barrels during the week ending October 2nd, after our distillates supplies had decreased by 2,251,000 barrels during the prior week… Our distillates supplies fell by less this week because the amount of distillates supplied to US markets, an indicator of domestic demand, fell by 298,000 barrels per day to 3,650,000 barrels per day, even as our exports of distillates rose by 235,000 barrels per day to 1,764,000 barrels per day, while our imports of distillates fell by 35,000 barrels per day to 118,000 barrels per day… After 31 withdrawals from distillates inventories over the past 64 weeks, our distillates supplies at the end of the week were 15.7% below the 124,665,000 barrels of distillates that we had in storage on October 3rd of 2025, and were about 12% below the five year average of our distillates inventories for this time of the year…

Finally, after the big jump in our oil exports, our commercial supplies of crude oil in storage fell for the 17th time in twenty-six weeks, and for the 27th time over the past year, decreasing by 3,186,000 barrels over the week, from 427,320,000 barrels on September 25th to 424,134,000 barrels on October 2nd, after our commercial crude supplies had increased by 922,000 barrels over the prior week….After this week’s decrease, our commercial crude oil inventories were still about 1% above recent five-year average of commercial oil supplies for this time of year, and they were about 19% above the average of our available crude oil stocks as of the first weekend of October over the 5 years at the beginning of the past decade, with the difference between those comparisons arising because it wasn’t until early 2015 that our oil inventories had first topped 400 million barrels. After our commercial crude oil inventories had jumped to record highs during the Covid lockdowns in the Spring of 2020, then jumped again after February 2021’s winter storm Uri froze off US Gulf Coast refining, but then fell sharply due to increased exports to Europe following the onset of the Ukraine war, only to jump again following the Christmas 2022 refinery freeze-offs, changes in our commercial crude inventories had been less extreme up until the onset of the Iran war, when they were initially built up to a three year high by mid-April, before falling to the lowest in nearly eight years by late July…This week’s decrease was the largest in six weeks, but as of October 2nd our commercial crude inventories were still 0.9% above the 420,261,000 barrels of oil we had in commercial storage on October 3rd of 2025, and 0.3% more than the 422,741 ,000 barrels of oil that we had in storage on September 27th of 2024, but fractionally less than the 424,239,000 barrels of oil we had left in commercial storage on October 6th of 2023…

This Week’s Rig Count

The US rig count increased by five over the week ending October 9th, as the number of rigs targeting oil was up by six, while the count of rigs targeting natural gas was down by one, and miscellaneous rigs were unchanged…for a quick snapshot of this week’s rig count, we are again including below a screenshot of the rig count summary table from Baker Hughes…in the table below, the first column shows the active rig count as of October 9th, the second column shows the change in the number of working rigs between last week’s count (October 2nd) and this week’s (October 9th) count, the third column shows last week’s October 2nd active rig count, the 4th column shows the change between the number of rigs running on Friday and the number running on the Friday of the same week of a year ago, and the 5th column shows the number of rigs that were drilling at the end of that reporting period a year ago, which in this week’s case was Friday, the 10th of October, 2025…

++

3rd estimate of Q2 GDP and annual revision; September’s jobs report, August’s income and outlays, construction spending, factory inventories, and JOLTS

The major economic releases of last week were the 3rd estimate of 2nd quarter GDP from the Bureau of Economic Analysis, which included an annual revision covering the prior five years, and the August report on Personal Income and Spending, also from the BEA, which includes two months of data on personal consumption expenditures and hence will account for more than 47% of 3rd quarter GDP, the Employment Situation Summary for September and the Job Openings and Labor Turnover Survey (JOLTS) for August, both from the Bureau of Labor Statistics, and the August report on Construction Spending (pdf) and the Full Report on Manufacturers’ Shipments, Inventories and Orders for August, both from the Census Bureau…in addition, this week also saw the last regional Fed manufacturing survey for September: the Dallas Fed Texas Manufacturing Outlook Survey, which covers Texas and adjacent western Louisiana and southeastern New Mexico, reported its general business activity edged down to +9.8 in September from +11.6 in August, indicating a smaller plurality of Texas businesses are experiencing improving metrics than in the prior month..

The week’s major privately issued reports included the ADP Employment Report for September, wherein the private payroll processor reported a 90,000 increase in private jobs in September; the September report on light vehicle sales from Omida, aka Wards Automotive, (the source of the BEA’s data) which estimated that vehicles sold at a 16.39 million annual rate in September, down from the 16.80 million annual rate in August, and down from the 16.72 million annual sales rate in September of last year, the Case-Shiller house price indexes for July from S&P Case-Shiller, who reported that their national home price index based on relative May, June and July home sales prices was 1.9% higher than their price index for the same three months of a year earlier, and the September Manufacturing Report On Business from the Institute for Supply Management, which reported that the manufacturing PMI (Purchasing Managers Index) slipped to 54.5% in September from 54.6% in August, indicating that a slightly smaller plurality of manufacturing purchasing managers continued to report improving metrics for the key facets of their business in September….

Employers Added 29,000 Jobs in September; Unemployment Rate, Employment Rate, and Participation Rate All Rose

The Employment Situation Summary for September reported a much smaller increase in payroll jobs than what had been expected, but that the employment rate and the labor force participation rose, and that the unemployment rate rose largely on upward rounding…estimates extrapolated from the establishment survey data projected that employers added a seasonally adjusted 29,000 jobs in September, after the previously estimated payroll job change for July was revised down from last month’s revised increase of 21,000 jobs to a decrease of 10,000, while the payroll jobs increase for August was revised from 162,000 down to 133,000, revisions which mean that this report thus shows 31,000 fewer payroll jobs than last month’s, in contrast to the 84,000 job increase forecast by the financial media….the unadjusted data, meanwhile, shows that there were actually 333,000 more payroll jobs in September, largely due to job increases relating to the beginning of the school year, and that the seasonal adjustment brought the headline jobs number down to a level where that normal September job increase was negated…

Seasonally adjusted job increases were spread through the goods producing and the service sectors, while modest job losses were seen in government, information, and financial activities…since the BLS summary of the job gains by sector is clear and as detailed as our usual synopsis, we’ll just quote from that summary here:

• Total nonfarm payroll employment changed little in September (+29,000), following an average monthly gain of 45,000 over the prior 12 months. Employment in all major industries changed little over the month. (See table B-1.)
• Health care employment continued its upward trend in September (+17,000), but at a slower pace than the average monthly gain over the prior 12 months (+33,000). In September, employment continued to trend up in ambulatory health care services (+13,000) and in hospitals (+12,000), while nursing and residential care facilities lost jobs (-9,000).
• Employment in construction changed little in September (+11,000). The industry had added an average of 10,000 jobs per month over the prior 12 months. In September, employment in nonresidential specialty trade contractors continued to trend up (+12,000).
• Manufacturing employment was little changed in September (+9,000) but is up by 72,000 since a recent low in December 2025. Over the month, employment increased in plastics and rubber products manufacturing (+5,000) and in machinery manufacturing (+5,000).
• In September, financial activities employment was little changed (-7,000). Employment in financial activities is down by 129,000 since a recent peak in May 2025, with most of the job loss in insurance carriers and related activities (-90,000).
• Employment also showed little change over the month in other major industries, including mining, quarrying, and oil and gas extraction; wholesale trade; retail trade; transportation and warehousing; information; professional and business services; social assistance; leisure and hospitality; other services; and government..

The establishment survey also showed that average hourly pay for all employees rose by 5 cents an hour to $37.81 an hour, after it had increased by a revised 11 cents an hour in August; at the same time, the average hourly earnings of production and nonsupervisory employees increased by 7 cents to $32.60 an hour….employers also reported that the average workweek for all private payroll employees was unchanged at 34.4 hours in September, while hours for production and non-supervisory personnel was unchanged at 33.8 hours…at the same time, the manufacturing workweek was unchanged at 40.6 hours, while average factory overtime was unchanged at 3.0 hours…

Meanwhile, the seasonally adjusted extrapolation from the September household survey indicated that the number of those who would self-report being employed rose by an estimated 406,000 to 163,152,000, while the similarly estimated number of those who would qualify as being unemployed and looking for work rose by 78,000 to 7,109,000; and hence the civilian labor force increased by a rounded 485,000…since the working age population had grown by 139,000 over the same period, that meant the number of employment aged individuals who were not in the labor force fell by a rounded net of 346,000 to 105,292,000….the increase in the total labor force was enough to raise the labor force participation rate, from 61.6% in August to 61.8 in September, even as it was still down from 62.5% in September of last year…at the same time, the relatively larger increase in number employed was enough to boost the employment to population ratio, which we could think of as an employment rate, from 59.1% in August to 59.2% in September…meanwhile, with the modest increase in the number unemployed, the unemployment rate rose to 4.2% from 4.1%, which was a function or rounding more than anything, since the actual increase was from 4.14% to 4.17%….meanwhile, the number of the employed who reported they were forced to accept just part time work rose by 111,000 to 4,501,000 in September, but even so, the alternative measure of unemployment, U-6, which includes those “employed part time for economic reasons”, fell from 7.7% in August to 7.6% in September, the lowest U6 unemployment rate since January 2025, largely due to a decrease in discouraged workers and a 236,000 drop in “marginally attached” workers….

Like most reports from the Bureau of Labor Statistics, the employment situation press release itself is easy to read and understand, so you can get more details on these two reports from there…note that almost every paragraph in that release points to one or more of the tables that are linked to on the bottom of the release, and those tables are also on a separate html page here that you can open it alongside of the press release to avoid the need to scroll up and down the page..

Third Estimate of 2nd Quarter GDP & Revisions From 2021 to Present

The Third Estimate of our 2nd Quarter GDP from the Bureau of Economic Analysis released on Thursday included an annual revision, which in this year’s case revised GDP data from the first quarter of 2021 through the first quarter of 2026, resulting in revisions to GDP, GDP by industry, gross domestic income, and related components…the updated estimates show that real GDP increased at an average annual rate of 3.4 percent over the five year period from 2020 to 2025, an upward revision from the 3.3% growth rate for that period that was previously published…for 2021, the change in real GDP was revised from 6.2 percent to 6.3 percent, reflecting a downward revision to imports (which are a subtraction in the calculation of GDP) and upward revisions to state and local government spending and to consumer spending….for 2022, the change in real GDP was revised down from 2.5 percent to 2.4 percent, reflecting downward revisions to consumer spending and to state and local government spending…meanwhile, the change in real GDP for 2023 was unrevised at 2.9 percent, as upward revisions to state and local government spending and to nonresidential fixed investment were offset by downward revisions to consumer spending on services and to private inventory investment….for 2024, the change in real GDP was revised from 2.8 percent to 3.0 percent, primarily reflecting upward revisions to nonresidential fixed investment and state and local government spending…lastly, for 2025, the change in real GDP was revised from 2.1 percent to 2.3 percent, reflecting upward revisions to nonresidential fixed investment in structures and state and local government spending, which were partly offset by an upward revision to imports…

The growth rate of the first quarter of 2026, which had last been reported at 2.1% when we reviewed the 3rd estimate of 1st quarter GDP three months ago, was revised to a 2.5% growth rate after this revision, as upward revisions to consumer consumption of goods and services and to exports more than offset downward revisions to fixed private investment, to inventories, and to government…More importantly, the price index for gross domestic purchases for the first quarter of 2026 was revised from 3.6% to 3.2%, reflecting a change in the methodology used to compute inflation, which thus underpinned the rounded 0.4 percentage point upward revision to real 1st quarter growth, suggesting there was little change in the nominal GDP components before the inflation adjustment was applied…checking that, we find that the current dollar GDP growth rate for the first quarter was actually statistically unchanged at 5.76%..

Revisions like those to the GDP readings of the past five years, and to the 3rd estimate of 1st quarter GDP, published as “final” just three months ago, should caution you to take even this 3rd estimate of 2nd quarter growth, which was released on Wednesday, with a grain of salt…the Third Estimate of our 2nd Quarter GDP from the Bureau of Economic Analysis indicated that our real output of goods and services grew at a 2.2% annual rate in the 2nd quarter, revised from the 1.5% growth rate that was reported in the second estimate last month, as growth of personal consumption expenditures for both goods and services, fixed private investment, inventories, net exports, the federal government, and state and local government were all upwardly revised….In current dollars, our second quarter GDP grew at a 8.49% annual rate, increasing from a revised $31,906.3 billion annual rate in the 1st quarter to a $32,563.0 billion annual rate in the 2nd quarter, with the headline 2.2% annualized rate of increase in real output arrived at after annualized GDP inflation adjustments averaging 6.1% were computed from the price changes of the GDP components and applied to their current dollar change…that so-called “GDP deflator” was revised down from the 6.4% reported a month ago, which thus  accounted for 0.3% of the upward revision to GDP…

As we review this month’s revisions to the 2nd quarter data, remember that this press release reports all quarter over quarter percentage changes at an annual rate, which means that they’re expressed as a change that’s compounded by 4 times from the change that actually occurred from one 3 month period to the next, and that the prefix “real” is used to indicate that each change has been adjusted for inflation using price changes now chained from 2017, and then that all percentage changes in this report are calculated from those 2017 dollar figures, which are then used as quantity indexes, rather than as any reality based dollar amounts….for our purposes, all the data that we’ll use in reporting the changes here comes directly from here comes directly from National Income and Product Accounts Data Tables.…specifically, we’ll be referencing Table 1.1.5. Gross Domestic Product, which shows the quarterly current dollar values for each of the major GDP components since the first quarter of 2024, Table 1.1.7. Percent Change From Preceding Period in Prices for Gross Domestic Product, which shows the change in the price indexes for each of the major GDP components over the same period;  Table 1.5.2. Contributions to Percent Change in Real Gross Domestic Product, Expanded Detail, which shows the contribution of each of the components to the GDP change for those quarters; and Table 1.5.6. Real Gross Domestic Product, Expanded Detail, Chained Dollars, which shows the inflation adjusted value of each of the GDP components…

Growth of real personal consumption expenditures (PCE), used to compute the largest component of GDP, was revised from the 3.4% growth rate reported last month to a growth rate of 3.8% in this 3rd estimate…that growth rate figure was arrived at by deflating the 8.97% growth rate in the dollar amount of consumer spending with the PCE price index, which indicated dollar weighted consumer inflation grew at a 5.0% annual rate in the 2nd quarter, which was revised from the 5.3% PCE inflation rate reported a month ago…..real (inflation adjusted) consumption of durable goods grew at a 7.4% annual rate, revised from the 5.7% growth rate reported in the 2nd estimate, and added 0.52 percentage points to GDP, as real consumption of motor vehicles and parts grew at an 11.3% rate and accounted for half of the growth in durable goods….at the same time, real consumption of nondurable goods by individuals grew at a 3.1% annual rate, revised from the 3.5% increase reported in the 2nd estimate, and added 0.41 percentage points to the 2nd quarter’s economic growth, as growth in consumption of food, clothing and footwear, and ‘other’ non-durables offset a modest decrease in real consumption of gasoline……meanwhile, real consumption of services grew at a 3.4% annual rate, revised from the 3.1% growth rate reported last month, and added 1.57 percentage points to the final GDP tally, as a 7.0% growth rate in real consumption of health care accounted for more than 50% of the growth in services…

Meanwhile, seasonally adjusted real gross private domestic investment grew at a 4.6% annual rate in the 2nd quarter, revised from the 2.7% growth estimate reported last month, as real private fixed investment grew at a 7.7% rate, revised up from the 7.0% growth rate reported in the second estimate, while the shrinkage of real inventories was somewhat smaller than previously estimated….the change in real investment in non-residential structures was revised from shrinking at a 1.8% rate to growing at an 0.1% rate, while real investment in equipment grew at a 13.4% rate, revised from the 13.6% growth rate previously reported…at the same time, the quarter’s investment in intellectual property products was revised from growth at a 8.8% rate to growth at a 9.2% rate, while the growth rate of residential investment was revised from 1.3% to 2.8% annually….after those revisions, the increase in investment in non-residential structures added 0.00 percentage points to the 2nd quarter’s growth rate, the increase in investment in equipment added 0.74 percentage points to the quarter’s growth, the increase in investment in intellectual property added 0.51 percentage points, while the increase in investment in residential structures added 0.10 percentage points to the 2nd quarter’s GDP growth…

At the same time, the second quarter’s change in real private inventories was revised from the previously reported $91.1 billion decrease in inflation adjusted dollars to indicate inventories shrunk at an inflation adjusted $63.4 billion rate…that came after inventories had shrunk at an inflation adjusted $31.9 billion in the 1st quarter, revised from the $28.1 billion 1st quarter inventory contraction previously reported, and hence the $31.5 billion negative change in real inventory growth from that of the 1st quarter subtracted 0.53 percentage points from the 2nd quarter’s growth rate, revised from the 0.72 percentage point subtraction shown in the second estimate, when the quarterly change in real inventory growth was shown at -$63.0 billion….however, since growth in inventories would indicate that more of the goods produced during the quarter would have been left in storage or “sitting on the shelf”, the $31.5 billion decrease in their growth conversely means real final sales of GDP were greater by that amount, and therefore the BEA found that real final sales of GDP grew at a 2.75% rate in the 2nd quarter, revised from the 2.2% rate of increase in real final sales originally shown in the second estimate…

The previously reported increase in real exports was somewhat greater after this estimate, while the previously reported increase in real imports was revised higher by less at the same time, and as a result, the negative impact of our imbalanced foreign trade on GDP was a bit smaller than in the second estimate….our real exports grew at a 5.0% rate, revised from the 4.5% growth rate reported in the second estimate, and since exports are added to GDP because they are part of our production that was not consumed or added to investment in our country and hence not included in other GDP components, that export growth added 0.56 percentage points to 2nd quarter GDP, revised from the  addition of 0.50 percentage points shown in the second estimate….meanwhile, the previously reported 12.5% increase in our real imports was revised to a 12.6% increase , and since imports subtract from GDP because they represent either consumption or investment added to another GDP component that shouldn’t have been because was not produced here, that import increase subtracted 1.66 percentage points from the 2nd quarter’s growth rate, revised from the 1.64 percentage point subtraction due t0 imports shown in the second estimate…..thus our deteriorating trade balance subtracted a net 1.10 percentage points from 2nd quarter GDP, revised from the rounded 1.14 percentage point subtraction that had been indicated in the second estimate…

Finally, there were also revisions to real government consumption and investment in this 3rd estimate, up for both the Federal government and for state and local government, as the entire government sector shrunk at an 0.1% rate, revised from the 1.0% contraction rate previously reported…real federal government consumption and investment was seen to have shrunk at a 2.4% annual rate, revised from the 4.2% contraction rate shown in the 2nd estimate, as real federal outlays for defense were revised to show growth at a 5.3% rate, revised from the 2.4% growth rate previously reported, and added 0.19 percentage points to 2nd quarter GDP, while all other federal consumption and investment shrunk at a 12.7% rate, revised from the 13.2% contraction rate previously reported, and subtracted 0.34 percentage points from 2nd quarter GDP….meanwhile, real state and local consumption and investment grew at a 1.2% rate over the quarter, which was revised from the 0.9% growth rate reported in the 2nd estimate, and added 0.13 percentage points to 2nd quarter GDP growth, as state and local investment shrunk at a 0.9% rate and subtracted 0.02 percentage points from the 0.15 addition from state and local consumption expenditures…note that government outlays for social insurance are not included in this GDP component; rather, they are included within personal consumption expenditures only when such funds are spent on goods or services, thus indicating there was an increase in the output of those goods or services…

August Personal Income up 0.2%, Spending Up 0.9%; PCE Prices Up 0.3%; Savings Rate at a 44 Month Low; Two Months PCE Would Add 199 Basis Points to Q3 GDP

Like the GDP report, the Income and Outlays report for August also went through an annual revision, with its revisions also from the first quarter of 2021 through the first quarter of 2026 for all of the metrics it reports, including personal consumption expenditures (PCE), the personal income and disposable personal income data, our savings and savings rate, and the PCE price index, the inflation gauge the Fed targets….A summary of those revisions and comparisons to previously published data is provided by an Information on 2026 Annual Updates page that accompanies this month’s report...since all the revisions made to personal consumption expenditures had already been incorporated into the GDP revisions that we have just reviewed, at this time we’ll only consider those revisions from recent months that are relevant to putting the August and potential 3rd quarter changes in perspective…

Also like the GDP report, all the dollar values reported by this release are seasonally adjusted and at an annual rate, ie, they tell us how much national income, spending, and savings would change over a year if August’s adjusted income and spending were extrapolated over an entire year…however, the percentage changes are computed monthly, from one month’s annualized figure to the next, and in this case of this month’s report they give us the percentage change in each annualized metric from July to August…..as they now try to explain in the opening line of the news release for this report, “Personal income increased $66.6 billion (0.2 percent at a monthly rate) in August“, which means that the annualized figure for seasonally adjusted personal income in August, $27,626.9 billion, was $66.6 billion, or more than 0.2% greater than the annualized personal income figure of $27,560.3 billion for July; the actual, unadjusted change in personal income from July to August, which would be on the order of one-twelfth of that size, is not given…similarly, annualized disposable personal income, which is income after taxes, rose by almost 0.3%, from an annual rate of $24,154.6 billion in July to an annual rate of $24,223.2 billion in August….the monthly contributors to the annualized $66.6 billion increase in personal income, which can be can be viewed in Table 2.6. of the National Income and Product Accounts, are therefore also annualized…in August, the major reasons for the $66.6 billion annual rate of increase in personal income were an annualized $36.6 billion increase in wages and salaries, and an annualized $22.9 billion increase in government social benefits to person, led by increases in Social security and Medicare benefits, and an $5.8 billion annualized increase in proprietors’ income, of which farmers saw a $3.7 billion annualized increase….

For personal consumption expenditures (PCE), BEA reports that they increased at a $190.8 billion annual rate, or by almost 0.9 percent, as the annual rate of PCE rose from $22,117.7 billion in July to $22,308.5 billion in August; that was after the July PCE figure was revised down from the originally reported $22,250.4 billion to $22,117.7 billion annually, while prior months were revised as well, revisions which were already included in the concurrent 3rd estimate of 2nd quarter GDP…..total personal outlays for August, which includes interest payments and personal transfer payments in addition to PCE, rose by an annualized $190.7 billion to $23,232.9 billion annually, which left total personal savings, which is disposable personal income less total outlays, at a $990.2 billion annual rate in August, down from the revised $1,112.4 billion annualized personal savings in July… as a result, the personal saving rate, which is personal savings as a percentage of disposable personal income, fell from 4.6% in July to 4.1% in August, the lowest personal savings rate since November 2022…

Before personal consumption expenditures can be used in the GDP computation, they must first be adjusted for inflation to give us the real change in consumption, and hence the real change in goods and services that were produced for that consumption….the BEA does that by computing a price index for personal consumption expenditures, which is a chained price index based on 2017 prices = 100, and which is included in NIPA Table 2.8.4 for this report…….that index rose from 131.172 in July to 131.579 in August, a month over month inflation rate that’s statistically 0.3102796%, which BEA reports as a 0.3% increase, following the rounded 0.1% increase in the PCE price index they reported for July…applying that August inflation adjustment to the nominal change in August spending left real PCE up by 0.550669%, or by a rounded 0.6% in August, after a real PCE increase of 0.1% in July …note that when those price indexes are applied to a given month’s annualized PCE in current dollars, it yields that month’s annualized real PCE in chained 2017 dollars, which are the means that the BEA uses to compare one month’s or one quarter’s real goods and services produced to another….the BEA’s result for that is shown in NIPA Table 2.8.6, where we see that August’s chained dollar personal consumption total works out to 16,955.3 billion annually, 0.550334% more than July’s 16,862.5     billion, a difference in real PCE that the BEA reports as +0.6%…

However, to estimate the impact of the change in real PCE on the change in GDP, month over month changes such as that don’t help us much, since GDP is reported quarterly…thus we have to compare July and August’s real PCE to the the real PCE of the 3 months of the second quarter….while this report reports real PCE for each of those months separately, we can get their annualized average in NIPA Table 1.5.6 for the 2nd quarter GDP report, where we find that the annualized real PCE for the 2nd quarter was represented by 16,788.7 billion in chained 2017 dollars….then, by averaging the annualized chained 2017 dollar figures for July and August, 16,862.5 billion and 16,955.3 billion respectively, we get an equivalent annualized PCE for the two months of the 3rd quarter that we have data for so far….when we compare that average of 16908.9 billion to the 2nd quarter real PCE representation of of 16,788.7 billion, we find that 3rd quarter real PCE has grown at a 2.89% annual rate for the two months of the 3rd quarter that we have…{note the math we’ve used to get that annual growth rate: (((16,955.3 + 16,862.5) / 2) / 16,788.7 ) ^ 4 = 1.02894734 }…that’s a pace that would add 1.99 percentage points to the growth rate of the 3rd quarter, should there be no change in September’s real PCE from that July & August average…

Job Openings and Layoffs Fell in August, Hiring Rose, and Job Quitting was Little Changed

The Job Openings and Labor Turnover Survey (JOLTS) report for August from the Bureau of Labor Statistics estimated that seasonally adjusted job openings fell by 256,000, from 7,335,000 in July to 7,079,000 openings in August, after July job openings were revised 64,000 higher, from the 7,271,000 reported last month to 7,335,000 now….August jobs openings were still 2.3% higher than the 6,919,000 job openings reported for August a year ago, as the job opening ratio expressed as a percentage of the employed fell from 4.4% in July to 4.3% in August, but was still up from 4.2% in August a year ago…the largest percentage decrease in job openings was the 45,000 decrease to 50,000 job openings in real estate and rental and leasing, while job openings information sector increased by 45,000 to 123,000 (see table 1 for more details)…like most BLS releases, the press release for this report is easy to understand and also refers us to the associated table for the data cited, which are linked at the end of the release…

The JOLTS release also reports on labor turnover, which consists of hires and job separations, which in turn is further divided into layoffs and discharges, those who quit their jobs, and ‘other separations’, which includes retirements and deaths….in August, seasonally adjusted new hires totaled 5,192,000, up by 46,000 from the revised 5,146,000 who were hired or rehired in July, as the hiring rate as a percentage of all employed rose from 3.2% in July to 3.3% in August, and was also up from the 3.2% hiring rate in August a year earlier (details of hiring by sector since March are in table 2)….meanwhile, total separations fell by 58,000, from 5,128,000 in July to 5,070,000 in August, while the separations rate as a percentage of the employed was unchanged at 3.2% in August, which was still down from the separations rate of 3.3% in August a year ago (see details in table 3)…subtracting the 5,070,000 total separations from the total hires of 5,192,000 would imply an increase of 122,000 jobs in August, a little less than the revised payroll job increase of 133,000 jobs in August reported by the September establishment survey later in the week, but still within the expected +/-110,000 margin of error in these incomplete survey extrapolations…..

Breaking down the seasonally adjusted job separations, the BLS finds that 3,066,000 of us voluntarily quit our jobs in August, down by 23,000 from the revised 3,089,000 who quit their jobs in July, while the quits rate, widely watched as an indicator of worker confidence, was unchanged at 1.9% of total employment, which was still down from the quits rate of 2.0% a year earlier (see details in table 4)….in addition to those who quit, 1,641,000 employees were either laid off, fired or otherwise discharged in August, down by 61,000 from the revised 1,702,000 who were discharged in July, while the discharges rate fell from 1.1% of all those who were employed during July to 1.0% of the employed in August, which was also also down from the 1.2% discharges rate in August of a year earlier….meanwhile, other separations, which includes retirements and deaths, were at 263,000 in August, up from 237,000 in July, for an ‘other separations rate’ of 0.2%, same as in July and as in August of last year….both seasonally adjusted and unadjusted details by industry and by region on hires and job separations, and on job quits and discharges can be accessed using the links to tables at the bottom of the press release…

Construction Spending Rose 0.9% in August, with July Implies 94 Basis Point Hit to 3rd Quarter GDP

The August report on construction spending (pdf) from the Census Bureau estimated that our seasonally adjusted construction spending construction spending for the month was at an annual rate of $2,203.1 billion, which was 0.9 percent (±1.0 percent)* above the revised annualized estimate of $2,184.5 billion in construction spending in July, but was 4.1 percent (±1.6 percent) above the estimated annualized level of construction spending of August of last year….for the first eight months of this year, construction spending amounted to $1,428.5 billion, which was 7.6 percent (±1.2 percent) above the $1,327.0 billion spent over the same period in 2023…

July’s construction spending was originally reported at a $2,167.7 billion annual rate, so it has been revised almost 0.8% higher, to a $2,184.5 billion annual rate, while June’s construction spending was revised 0.9% higher, from the $2,167.7 billion annual rate reported last month to a $2,187.35 billion rate now…that annualized $19.6 billion upward revision to June spending would mean that 2nd quarter GDP was underestimated by about 0.13 percentage points, give or take, depending on inflation adjustments for the components revised…..however, the 2nd quarter’s GDP will not be revised to reflect that underestimation until the annual revision of next summer…

A further breakdown of the different subsets of construction spending is provided in a Census summary, which precedes the detailed spreadsheets:

• Private Construction:  Spending on private construction was at a seasonally adjusted annual rate of $1,655.3 billion, 1.1 percent (±0.5 percent) above the revised July estimate of $1,637.7 billion. Residential construction was at a seasonally adjusted annual rate of $882.3 billion in August, 1.1 percent (±1.3 percent)* above the revised July estimate of $872.7 billion. Nonresidential construction was at a seasonally adjusted annual rate of $773.0 billion in August, 1.0 percent (±0.5 percent) above the revised July estimate of $765.0 billion.
• Public Construction: In August, the estimated seasonally adjusted annual rate of public construction spending was $547.8 billion, 0.2 percent (±2.0 percent)* above the revised July estimate of $546.8 billion. Educational construction was at a seasonally adjusted annual rate of $113.1 billion, 0.1 percent (±1.5 percent)* above the revised July estimate of $112.9 billion. Highway construction was at a seasonally adjusted annual rate of $150.6 billion, 0.1 percent (±5.6 percent)* above the revised July estimate of $150.5 billion.

This construction spending report is used as source data for 3 subcomponents of GDP; investment in private non-residential structures, investment in residential structures, and government investment outlays, for both state and local and Federal governments…. however, gauging the impact of revised July and August construction spending as reported here on 3rd quarter GDP is difficult because all figures given in this report are in nominal dollars at an annual rate, while data used to compute the change in GDP must be adjusted for changes in price…accurately adjusting construction  spending for price changes is not easy either, because the National Income and Product Accounts Handbook, Chapter 6 (pdf), lists a multitude of BLS and privately published deflators that are used by the BEA for the various components of non-residential investment, such as the Engineering News Record construction cost index for utilities’ construction spending….so in lieu of trying to find and adjust for all of those obscure price indices, we’ve opted to just use the producer price index for final demand construction as an inexact shortcut to make the price adjustment needed in order to come up with a ballpark estimate.

That producer price index showed that aggregate construction costs were unchanged in August, after rising 2.2% in July, rising 0.1% in June and rising 0.1%  from April to May…on that basis, we can estimate that construction costs for August were roughly 2.2% more than those of June, roughly 2.3% more than those of May, and roughly 2.4% more than those of April, while obviously unchanged from those of July…we then use those percentages to inflate the lower cost spending figures for each of the 2nd quarter months, which, for comparison purposes, is arithmetically the same as adjusting higher cost July & August construction spending downward…annualized construction spending in millions of dollars for the second quarter months is shown at $2,187,350 for June, $2,168,396 for May, and $2,168,231 for April in this report, while it was at $2,184,474 million for July and $2,203,129 million for August….thus to compare July and August’s inflation adjusted construction spending to that of the second quarter, our formula becomes: ((2,203,129 + 2,184,474 * 1.000 ) / 2 ) / (( 2,187,350 * 1.022 + 2,168,396 * 1.023 + 2,168,231 * 1.024 ) / 3) = 0.98612455465, meaning real construction over July and August averaged 1.39% lower than that of the 2nd quarter….hence, that means that after crudely adjusting for inflation, we can estimate that real construction for the 3rd quarter fell at a 5.44% annual rate from that of the 2nd quarter….put another way, that’s a decrease at a $30,868 billion annual rate, which suggests that if September should show no improvement, the decrease in real construction would subtract a net of about 0.47 percentage points from 3rd quarter GDP across those components that it is source data for..

Factory Shipments unchanged in August, Factory Inventories Rose 0.5%

The August Full Report on Manufacturers’ Shipments, Inventories, & Orders (pdf) from the Census Bureau reported that the seasonally adjusted value of new orders for manufactured goods rose by $0.7 billion or 0.1 percent to $663.5 billion in August, following an increase of 0.8% to $662.8 billion in July, which was revised from the 0.9% increase to $663.6 billion in new goods orders for July reported last month….however, since the Census Bureau does not even collect data on new orders for non durable goods for this widely watched “factory orders report”, both the “new orders” and “unfilled orders” sections of this report are really only accurate as revised updates to the August advance report on durable goods that we reported on last week…on those revisions, the Census Bureau’s own summary, which precedes their detailed spreadsheet of the metrics included in this report, is quite complete, so we’ll just quote directly from that summary here:

• Summary: New orders for manufactured goods in August, up two consecutive months, increased $0.7 billion or 0.1 percent to $663.5 billion, the U.S. Census Bureau reported today. This followed a 0.8 percent July increase. Shipments, down following eight consecutive monthly increases, decreased less than $0.1 billion or virtually unchanged to $658.6 billion. This followed a 0.8 percent July increase. Unfilled orders, up twenty-five of the last twenty-six months, increased $9.9 billion or 0.6 percent to $1,609.6 billion. This followed a 0.6 percent July increase. The unfilled orders-to-shipments ratio was 6.87, up from 6.81 in July. Inventories, up eleven consecutive months, increased $5.1 billion or 0.5 percent to $972.4 billion. This followed a 0.4 percent July increase. The inventories-to-shipments ratio was 1.48, up from 1.47 in July.
• New Orders for manufactured durable goods in August, down following two consecutive monthly increases, decreased $0.2 billion or 0.1 percent to $338.5 billion, down from the previously published virtually unchanged decrease. This followed a 0.9 percent July increase. Transportation equipment, down three of the last four months, drove the decrease, $0.8 billion or 0.7 percent to $114.1 billion. New orders for manufactured nondurable goods increased $0.9 billion or 0.3 percent to $325.0 billion.
• Shipments of manufactured durable goods in August, down following eight consecutive monthly increases, decreased $0.9 billion or 0.3 percent to $333.6 billion, down from the previously published 0.2 percent decrease. This followed a 0.9 percent July increase. Transportation equipment, also down following eight consecutive monthly increases, drove the decrease, $2.0 billion or 1.8 percent to $109.4 billion. Shipments of manufactured nondurable goods, up eight of the last nine months, increased $0.9 billion or 0.3 percent to $325.0 billion. This followed a 0.7 percent July increase. Petroleum and coal products, up seven of the last eight months, drove the increase, $0.9 billion or 1.4 percent to $70.7 billion.
• Unfilled Orders for manufactured durable goods in August, up twenty-five of the last twenty-six months, increased $9.9 billion or 0.6 percent to $1,609.6 billion, unchanged from the previously published increase. This followed a 0.6 percent July increase. Transportation equipment, up twelve of the last thirteen months, led the increase, $4.7 billion or 0.5 percent to $1,009.9 billion.
• Inventories of manufactured durable goods in August, up eleven consecutive months, increased $3.2 billion or 0.5 percent to $608.3 billion, unchanged from the previously published increase. This followed a 0.5 percent July increase. Transportation equipment, up ten of the last eleven months, led the increase, $1.2 billion or 0.6 percent to $192.7 billion. Inventories of manufactured nondurable goods, up six of the last seven months, increased $1.8 billion or 0.5 percent to $364.0 billion. This followed a 0.3 percent July increase. Petroleum and coal products, up two consecutive months, led the increase, $1.2 billion or 2.5 percent to $49.0 billion. By stage of fabrication, August materials and supplies increased 0.6 percent in durable goods and 0.9 percent in nondurable goods. Work in process increased 0.8 percent in durable goods and 1.1 percent in nondurable goods. Finished goods increased 0.1 percent in durable goods and were virtually unchanged in nondurable goods

To gauge the effect of August factory inventories on 3rd quarter GDP, they must first be adjusted for changes in price with appropriate components of the producer price index…by stage of fabrication, the value of finished goods inventories was 0.1% higher at $330,046 million; the value of work in process inventories was 0.9% higher at $276,346 million, and the value of materials and supplies inventories was 0.7% higher at $365,990 million.…meanwhile, the producer price index for August indicated that prices for finished goods were on average 1.1% higher, that prices for intermediate processed goods were 1.8% higher, and that prices for unprocessed goods were 1.1% higher….assuming average valuations will be similar for like inventories, we could thus estimate that August’s real finished goods inventories decreased by 1.0%, that real inventories of intermediate processed goods decreased by 0.9%, and that real raw material inventory inventories were about 0.4% lower…while on net it the appears that August’s real inventories were somewhat lower, they follow July’s factory inventory change, when real inventories were higher by somewhat more…since real NIPA factory inventories were down sharply in the 2nd quarter, accounting for about three-fourths of the quarter’s inventory decrease, that two months of factory inventory data appears to indicate a small real increase in July factory inventories would therefore mean that the difference between the 2nd quarter decrease and the July increase would be added to the 3rd quarter’s real growth in GDP…

>> (the above is the synopsis that accompanied my regular sunday morning news links emailing, which in turn was mostlyselected from my weekly blog poston theglobal glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are chosen from the aforementioned GGO posts,contact me…)  

US oil production at a record high, SPR is lowest since October 1982; gasoline supplies are lowest since November 2014

US oil prices fell for a second week after rising six out of the prior seven, as shipments of oil thru the Strait of Hormuz increased and Saudi Arabia resumed Red Sea oil exports after repairing its damaged East-West pipeline….after falling 3.8% to $92.41 a barrel last week on hopes that negotiations to end the war in Iran would proceed with Iran’s president in New York for a United Nations General Assembly meeting, the contract price for the benchmark US light sweet crude for October delivery jumped more than 2% during Asian trading on Monday after US President Trump rejected an Iranian peace proposal aimed at resolving the conflict and reopening the Strait of Hormuz, and was up $2.51 to trade near $106.83 a barrel as markets opened in the U.S., after Trump’s rejection of the Iranian peace proposal came amid reports he was considering new military strikes on the country, but pulled back from session highs later in the session after Saudi Arabia reportedly ramped up its critical East-West pipeline after repairs, and settled just 19 cents higher at $92.60 a barrel amid the stalemate in U.S.-Iran peace talks, while Trump said that U.S. negotiators were expected to engage in more talks later in the week…oil prices continued to rise in Asian trading Tuesday amid fading hopes for the reopening of the Strait of Hormuz and expectations that the US Federal Reserve would raise interest rates again next month, but softened Tuesday morning in New York as signs of higher Middle Eastern oil supply outweighed concerns over stalling U.S.-Iranian negotiations, and then tumbled to settle $3.22 lower at $89.38 a barrel after trade sources said that Saudi Arabia resumed oil loadings from its Red Sea port of Yanbu after restarting operations on the East-West Pipeline…oil prices rose during trading in Asia on Wednesday after US President Trump rejected reports that he was prepared to ease sanctions on Iran, and were still up during trading on the ICE Futures Europe exchange in London, then rose to a session high during morning trading in New York after the EIA reported that both gasoline and distillate inventories fell, leaving Midwest gasoline supplies at their lowest on record, and settled $1.04 higher at $90.42 a barrel amid the continuing stalemate in U.S.-Iran peace talks…oil prices rose around 2% during Asian trading on Thursday after China suspended exports of oil products to most destinations, raising concerns about tighter fuel supplies worldwide, but were mixed in a volatile morning session Thursday in New York as a tightening global fuels market outweighed resurging crude exports from the Middle East, and settled $2.45 higher at $92.87 a barrel on the Chinese export ban and on the news that the U.S. was sending a third aircraft carrier strike group and up to 10,000 additional troops to the Middle East… oil prices edged higher during Asian trading on Friday as China halted fuel exports, while the US weighed additional military deployments to the Middle East, raising concerns over a potential escalation in tensions with Iran and further disruption to regional energy supplies, but fell $2 during European trading after European leaders agreed on Friday to a request by US President Trump to release diesel from their reserves to lower prices and reduce the need to import fuel from America, and were down more than $3 Friday morning in New York after the Elysée Palace proposed a joint EU-wide release of crude oil and diesel from strategic reserves, and settled $1.76 or 1.9% lower at $91.11 per barrel after the Group of Seven nations announced the release of diesel and crude stocks to ease surging fuel prices, while Saudi Arabia planned an offensive against Iran-backed Houthi militants in Yemen, which left oil prices 1.4% lower for the week…

meanwhile, natural gas prices finished lower for the first time in three weeks on milder weather and a quick return to normal Appalachian pipeline flows after last week’s outage…after rising 9.8% to $3.196 per mmBTU last week on falling production from domestic wells, another smaller than normal injection of gas into storage, and a leak-related shutdown of the Mountaineer XPress pipeline in West Virginia, the price of the benchmark natural gas contract for October delivery opened 10.8 cents lower on its last day of trading on Monday, and trended lower into the afternoon after an early rally failed, as traders assessed lower demand and the October contract’s settlement, and expired 19.6 cents lower at $3.000 per mmBTU as restored Mountaineer XPress supplies and fading shoulder season demand weighed on the front of the curve. while the price of the benchmark natural gas contract for November delivery settled 11.9 cents lower at $3.106  per mmBTU….now the quoted front month, that benchmark natural gas contract for November opened 4.6 cents lower on Tuesday and trended lower thereafter, as mild forecasts kept the pressure on, and settled 9.5 cents lower at $3.011 per mmBTU after a volatile session, pulled down by mild early October weather forecasts and restored Appalachian pipeline flows…that November natural gas price started Wednesday 0.8 cents lower and dipped to an intraday low of $2.966 by 9:55AM, as traders monitored production levels amid largely comfortable fall temperatures, then rose into the afternoon to settle 1.5 cents higher at $3.026 per mmBTU, as traders weighed thinning production against weak shoulder season demand ahead of Thursday’s government inventory report…November natural gas opened 1.1  lower on Thursday and hovered near $3.005 ahead of the weekly report, then drifted lower into the afternoon amid bearish short-term weather forecasts after failing to react to the as-expected injection, and settled 5.9 cents lower at $2.967 per mmBTU, as an in-line storage report failed to counter pressure from mild autumn weather and soft seasonal demand….natural gas futures started Friday’s session trading gingerly, amid largely stagnant – and bearish – demand-side fundamentals, then probed higher during late morning trading on diverse regional weather patterns and possible supply constraints, and settled 6.8 cents higher at $3.035 per mmBTU as an unplanned outage on the 30-inch diameter Tetco pipeline in Kentucky and approaching cooler weather provided support against a still-loose fundamental backdrop….natural gas prices thus ended 5.0% lower for the week, while the benchmark natural gas contract for November delivery, which had ended the prior week at $3.225 per mmBTU, finished 5.9% lower….

The EIA’s natural gas storage report for the week ending September 25th indicated that the amount of working natural gas held in underground storage rose by 64 billion cubic feet to 3,415 billion cubic feet by the end of the week, which left our natural gas supplies 138 billion cubic feet, or 3.9% below the 3,553 billion cubic feet of gas that were in storage on September 25th of last year, but 79 billion cubic feet, or 2.4% above the five-year average of 3,336 billion cubic feet of natural gas that had typically been in working storage as of the 25th of September over the most recent five years….the 64 billion cubic foot injection into natural gas storage for the cited week was close to the 63 billion cubic foot injection into storage that the market had been expecting ahead of the report, while it was more than the 56 billion cubic foot of gas that were injected into natural gas storage during the corresponding week of 2025, but less than the average 80 billion cubic foot injection into natural gas storage that had been typical for the last week in September over the past five years…

The Latest US Oil Supply and Disposition Data from the EIA

US oil data from the US Energy Information Administration for the week ending September 25th showed that after a big seasonal drop in our oil refining more than offset an increase in our oil exports, we had surplus oil left to add to our stored crude supplies for the the third time in twenty-three weeks, and for the 28th time in seventy weeks, despite an increase in demand for oil that the EIA could not account for…. Our imports of crude oil fell by an average of 179,000 barrels per day to 5,698,000 barrels per day, after falling by an average of 1,181,000 barrels per day during the prior week, while our exports of crude oil rose by an average of 289,000 barrels per day to average 3,570,000 barrels per day, which, when used to offset our imports, meant that the net of our trade of oil worked out to an import average of 2,128,000 barrels of oil per day during the week ending September 25th, an average of 468,000 fewer barrels per day than the net of our imports minus our exports during the prior week… At the same time, transfers to our oil supplies from Alaskan gas liquids, from natural gasoline, from condensate, and from unfinished oils were 4,000 barrels per day higher than the prior week at 431,000 barrels per day, while during the same week, production of crude from US wells was 16,000 barrels per day higher at a record high of 13,965,000 barrels per day.  Hence, our daily supply of oil from the net of our international trade in oil, from transfers, and from domestic well production appears to have averaged a total of 16,514,000 barrels per day during the September 25th reporting week…

Meanwhile, US oil refineries reported they were processing an average of 16,257,000 barrels of crude per day during the week ending September 25th, an average of 554,000 fewer barrels per day than the amount of oil that our refineries reported they were processing during the prior week, while over the same period, the EIA’s surveys indicated that a net of 20,000 barrels of oil per day were being added to the supplies of oil stored in the US… So, based on that reported & estimated data, the crude oil figures provided by the EIA appear to indicate that our total working supply of oil from net imports, from transfers, and from oilfield production during the week ending September 25th averaged a rounded 238,000 more barrels per day than what was added to storage plus our oil refineries reported they used during the week.  To account for the difference between the apparent supply of oil and the apparent disposition of it, the EIA just plugged a [ -238,000 ] barrel per day figure onto line 16 of the weekly U.S. Petroleum Balance Sheet, in order to make the reported data for the supply of oil and for the consumption of it balance out, a fudge factor that they label in their footnotes as “unaccounted for crude oil”, thus indicating there must have been a error or omission of that amount in the week’s oil supply & demand figures that we have just transcribed…. Since 215,000 barrels per day of oil supplies could not be accounted for in the prior week’s EIA data, that means there was a 453,000 barrel per day difference between this week’s oil balance sheet error and the EIA’s crude oil balance sheet error from a week ago, and hence the changes to supply and demand from that week to this one that are indicated by this week’s report are somehow off by that much, and are therefore pretty useless….However, since most oil traders react to the figures in these weekly EIA reports as if they were gospel, and since these weekly figures therefore often drive oil pricing and hence decisions to drill or complete oil wells, we’ll continue to report this data just as it’s published, and just as it’s watched & believed to be reasonably reliable by most everyone in the industry…(for more on how this weekly oil data is gathered, and the possible reasons for that “unaccounted for” oil supply, see this EIA explainer….also see this March 2023 twitter thread from an EIA administrator addressing these ongoing weekly errors, and what they had once hoped to do about it).

This week’s rounded 20,000 barrel per day average increase in our overall crude oil inventories came as an average of 132,000 barrels per day were being added to our commercial stocks of crude oil, while 112,000 barrels per day were being pulled out of our Strategic Petroleum Reserve, the twenty-seventh consecutive Iran war related withdrawal from the SPR, which left the SPR level at 283,767,000 barrels, the lowest since it was initially being filled in October 1982….Further details from the weekly Petroleum Status Report (pdf) indicated that the 4 week average of our oil imports fell to 6,364,000 barrels per day last week, which was still 4.8% more than the 6,073,000 barrel per day average that we were importing over the same four-week period last year, while the four week average of our exports fell to 3,775,000 barrels per day last week, which was 7.1% less than the 4,064,000 barrel per day average that we were exporting last year year at this time… This week’s crude oil production was reported to be 16,000 barrels per day higher at a record high of 13,965,000 barrels per day as the EIA’s estimate of the output from wells in the lower 48 states was 3,000 barrels per day higher at 13,484,000 barrels per day, while Alaska’s oil production was 13,000 barrels per day higher at 468,000 barrels per day…US crude oil production had reached a pre-pandemic high of 13,100,000 barrels per day during the week ending March 13th 2020, so this week’s reported oil production figure was 6.5% higher than that of our pre-pandemic production peak, and was also 43.9% above the pandemic low of 9,700,000 barrels per day that US oil production had fallen to during the third week of February of 2021.

US oil refineries were operating at 92.5% of their capacity while processing those 16,257,000 barrels of crude per day during the week ending September 25th, down from 94.0% the prior week, but still high for mid-September….the 16,257,000 barrels of oil per day that were refined that week were 0.6% more than the 16,168,000 barrels of crude that were being processed daily during the week ending September 26th of 2025, and 1.5% more than the 16,017,000 barrels that were being refined during the pre-pandemic week ending September 27th, 2019, when our refinery utilization rate was at 86.4%, which was a bit below the pre-pandemic normal utilization rate for this time of year…

With the decrease in the amount of oil that was being refined this week, gasoline output from our refineries was also lower, decreasing by 124,000 barrels per day to 9,466,000 barrels per day during the week ending September 25th, after our refineries’ gasoline output had decreased by 54,000 barrels per day during the prior week… This week’s gasoline production was 1.3% more than the 9,344,000 barrels of gasoline that were being produced daily over the week ending September 26th of last year, but 7.3% less than the gasoline production of 10,081,000 barrels per day seen during the prepandemic week ending September 27th, 2019….at the same time, our refineries’ production of distillate fuels (diesel fuel and heat oil) decreased by 156,000 barrels per day to 5,003,000 barrels per day, after our distillates output had decreased by 69,000 barrels per day during the prior week.  Even after those production decreases, our distillates output was 0.9% more than the 4,959,000 barrels of distillates that were being produced daily during the week ending September 26th of 2025, and 3.9% more than the 4,813,000 barrels of distillates that were being produced daily during the pre-pandemic week ending September 27th, 2019….

With this week’s decrease in our gasoline production, our supplies of gasoline in storage at the end of the week fell for the twenty-sixth time in thirty-three weeks, decreasing by 1,684,000 barrels to 204,362,000 barrels during the week ending September 25th, the lowest since November 7th 2014, after our gasoline inventories had decreased by 1,686,000 barrels during the prior week.  Our gasoline supplies fell again this week even though the amount of gasoline supplied to US users fell by 158,000 barrels per day to 8,689,000  barrels per day, because our exports of gasoline rose by 94,000 barrels per day to 838,000 barrels per day, while our imports of gasoline rose by 99,000 barrels per day to 500,000 barrels per day.. After fifty-six gasoline inventory withdrawals over the past eighty-four weeks, our gasoline supplies were 7.4% lower than last September 26th’s gasoline inventories of 220,694,000 barrels, and about 7% below the five year average of our gasoline supplies for this time of year…

After this week’s decrease in distillates production, our supplies of distillates fell for the fifteenth time in thirty-three weeks, decreasing by 2,251,000 barrels to 105,180,000 barrels during the week ending September 25th, after our distillates supplies had decreased by 428,000 barrels during the prior week… Our distillates supplies fell by more this week even as the amount of distillates supplied to US markets, an indicator of domestic demand, fell by 27,000 barrels per day to 3,948,000 barrels per day, because our exports of distillates rose by 198,000 barrels per day to 1,529,000 barrels per day, while our imports of distillates rose by 68,000 barrels per day to 153,000 barrels per day… After 30 withdrawals from distillates inventories over the past 63 weeks, our distillates supplies at the end of the week were 14.9% below the 123,577,000 barrels of distillates that we had in storage on September 26th of 2025, and were about 14% below the five year average of our distillates inventories for this time of the year…

Finally, after the seasonal slowdown in our oil refining, our commercial supplies of crude oil in storage rose for the 11th time in twenty-six weeks, and for the 27th time over the past year, increasing by 922,000 barrels over the week, from 426,398,000 barrels on September 18th to 427,320,000 barrels on September 25th, after our commercial crude supplies had increased by 2,969,000 barrels over the prior week….After this week’s increase, our commercial crude oil inventories were about 2% above recent five-year average of commercial oil supplies for this time of year, and they were about 20.1% above the average of our available crude oil stocks as of the fourth weekend of September over the 5 years at the beginning of the past decade, with the difference between those comparisons arising because it wasn’t until early 2015 that our oil inventories had first topped 400 million barrels. After our commercial crude oil inventories had jumped to record highs during the Covid lockdowns in the Spring of 2020, then jumped again after February 2021’s winter storm Uri froze off US Gulf Coast refining, but then fell sharply due to increased exports to Europe following the onset of the Ukraine war, only to jump again following the Christmas 2022 refinery freeze-offs, changes in our commercial crude inventories had been less extreme up until the onset of the Iran war, when they were initially built up to a three year high by mid-April, before falling to the lowest in nearly eight years by late July…This week’s increase was the second following three decreases, and as of September 25th our commercial crude inventories were 2.6% above the 416,546,000 barrels of oil we had in commercial storage on September 26th of 2025, and were 2.5% more than the 416,931,000 barrels of oil that we had in storage on September 27th of 2024, and 3.2% more than the 414,063,000 barrels of oil we had left in commercial storage on September 29th of 2023…

This Week’s Rig Count

The US rig count fell by one over the week ending October 2nd, as the number of rigs targeting oil was up by one, but the count of rigs targeting natural gas was down by two, while miscellaneous rigs were unchanged…for a quick snapshot of this week’s rig count, we are again including below a screenshot of the rig count summary table from Baker Hughes…in the table below, the first column shows the active rig count as of October 2nd, the second column shows the change in the number of working rigs between last week’s count (September 25th) and this week’s (October 2nd) count, the third column shows last week’s September 25th active rig count, the 4th column shows the change between the number of rigs running on Friday and the number running on the Friday of the same week of a year ago, and the 5th column shows the number of rigs that were drilling at the end of that reporting period a year ago, which in this week’s case was Friday, the 3rd of October, 2025…

++

August’s durable goods and new home sales

There were just two widely watched economic reports released this week: the August advance report on durable goods and the August report on new home sales, both from the Census bureau…meanwhile, the week also saw the release of the Chicago Fed National Activity Index (CFNAI) for August, a weighted composite index of 85 different economic metrics, which fell to to –0.04 in August from +0.08 in July, after the July index was revised up from –0.08 to +0.08; that left the more often cited 3 month moving average of the index at +0.01 in August, up from a revised –0.01 in July, with those figures for both months so close to zero that both would indicate that national economic activity has been near the historical trend over the summer months..

The week also saw the release of two more regional Fed manufacturing surveys for September: the Kansas City Fed manufacturing survey, covering western Missouri, Colorado, Kansas, Nebraska, Oklahoma, Wyoming and northern New Mexico, reported its broadest composite index rose to +14 in September, up from +10 in August and +9 in July, meaning a larger plurality of that region’s manufacturers reported improving business metrics this month than during the prior two, while the Richmond Fed Survey of Manufacturing Activity, covering an area that includes Virginia, Maryland, the Carolinas, the District of Columbia and West Virginia, reported its broadest composite index fell to −2 in September from +4 in August and from +5 in July, which means that a small plurality of that region’s manufacturers reported deteriorating business metrics during September, after a small plurality had reported improvement during the prior two months…

August Durable Goods: New Orders Fell < 0.1%, Shipments Fell 0.2%, Inventories Rose 0.5%

The Advance Report on Durable Goods Manufacturers’ Shipments, Inventories and Orders for August (pdf) from the Census Bureau reported that the value of the widely watched new orders for manufactured durable goods decreased by $0.1 billion to $338.6 billion, the first decrease in three months, following a 0.9% increase to $338.7 billion in July, which was revised from the 1.1% increase to $339.3 billion reported for July’s new durables orders a month ago….after an 8.5% jump in the value of April’s orders, new orders are now running 7.7% above those of a year ago, up a bit from the 7.6% year-to date increase we saw in this report last month…

As is usually the case, the volatile monthly change in the value of new orders for transportation equipment drove this month’s headline decrease, as the value of new transportation equipment orders fell $0.7 billion or 0.6 percent to $114.1 billion, on a 4.3% decrease to $18,825 million in new orders for commercial aircraft and a 0.6% decrease to $73,347 million in new orders for motor vehicles and parts….excluding new orders for transportation equipment, other new orders were up 0.3% in August, while new orders for nondefense capital goods excluding aircraft, a proxy for equipment investment, rose 1.6% to $87,629 million, after rising 0.6% in July…

Meanwhile, the seasonally adjusted value of August’s shipments of durable goods, which will be inputs into various components of 3rd quarter GDP after adjusting for changes in prices, fell for the first time in nine months, decreasing by $0.7 billion or 0.2 percent to $333.8 billion, after the value of July’s shipments was revised from $334.7 billion to $334.56 million, thus revising the previously reported 1.0% increase in July shipments to a 0.9% increase from June….a decrease of $2.0 billion or 1.8 percent to $109.5 billion in the value of shipments of transportation equipment led the August decrease, as the value of shipments of commercial aircraft fell 8.6% to $16,267 million, and the value of shipments of motor vehicles and parts fell 0.7% to $73,240 million…on the other hand, the value of shipments of durable goods other than transportation equipment rose 0.6% to $224,360 million, led by a 1.5% jump in shipments of computers and related products… As a result, the value of shipments of nondefense capital goods excluding aircraft rose 0.6% to $84,998 in August, after rising 1.4% in July, changes that will be reflected in 3rd quarter GDP equipment investment figures, after adjusting for changes in prices since the 2nd quarter ….

At the same time, the value of seasonally adjusted inventories of durable goods, also a major GDP contributor, rose for the eleventh  consecutive month, increasing by $3.0 billion or 0.5 percent to $608.1 billion, after the value of July’s inventories was revised from $604.4 billion to $605.054 billion, which was still a 0.5% increase from June…a $1.1 billion or 0.6 percent increase to $192.7 billion in the value of inventories of transportation equipment led the August inventory increase, while the value of inventories of other than transportation equipment rose 0.5% to $415.4 billion….

Finally, the value of unfilled orders for manufactured durable goods, which is probably a better measure of industry conditions than the widely watched but volatile new orders, rose for the twenty-fifth time in twenty-six months, increasing by $9.8 billion or 0.6 percent to $1,609.4 billion, after the value of July’s unfilled orders was revised from the $1,599.9 billion reported last month to $1,599.6 billion, which is still a 0.6% increase from June…a $4.7 billion or 0.5 percent increase to $1,009.8 billion in the value of unfilled orders for transportation equipment led the August increase, while unfilled orders other than those for transportation equipment were 0.9% higher at $599.6 billion…compared to a year earlier, the unfilled order book for durable goods is now 8.8% above the level of last August, as the value of unfilled orders for transportation equipment is 9.8% above its year ago level, on a 11.9% increase in the value of the backlog of orders for defense aircraft and a 9.9% increase in the value of the backlog of orders for civilian aircraft and parts…

August New Home Sales Highest in 8 Months; Average Sales Price Lowest in Two Years

The Census report on New Residential Sales for August (pdf) estimated that new single family homes were selling at a seasonally adjusted pace of 684,000 new homes a year, which was 6.4 percent (±19.5 percent)* above the revised July rate of 643,000 new home sales a year, but was 2.0 percent (±15.7 percent)* below the estimated annual rate that new homes were selling at in August of last year….the asterisks indicate that based on their small sampling, Census could not be certain whether August new home sales rose or fell from July, or even from those of a year ago, with the figures in parenthesis representing the 90% confidence range for reported data in this report, which has the largest margin of error and is subject to the largest revisions of any census construction series….hence, these initial new home sales reports are not very reliable and often see significant revisions… with this report, sales new single family homes in July were revised from the annual rate of 678,000 reported last month down to a 643,000 annual rate, while home sales in June, initially reported at an annual rate of 628,000 and revised to a 678,000 a year rate last month, were revised to a 672,000 a year rate with this report, and while May’s annualized home sale rate, initially reported at a 580,000 rate and revised from a 618,000 to a 630,000 rate last month, were revised up to a 636,000 rate with this release…

The annual rates of sales reported here are seasonally adjusted after extrapolation from the estimates of canvassing Census field reps, which indicated that approximately 57,000 new single family homes sold in August, up from the estimated 53,000 new homes that sold in July but down from the 58,000 that sold in June….the raw figures from Census field agents further allowed for estimates that the median sales price of new houses sold in August was $393,700, up from the estimated median sales price of $392,200 in July, but down from the median sales price of $417,900 in August a year ago, and that the average August new home sales price was at a two  year low of $478,700, down from the $526,400 average sales price in July, and down from the average sales price of $525,100 average in August a year ago….a seasonally adjusted estimate of 483,000 new single family houses remained for sale at the end of August, which represents a 8.5 month supply at the August sales rate, down from the revised 9.0 month supply of unsold homes in July, which was originally reported as a 9.6 month supply…. 

(the above is the synopsis that accompanied my regular Sunday morning news links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are chosen from the aforementioned GGO posts, contact me…)

highest natural gas close in 33 weeks; distillates imports at 52 week low; US rig count at 27 month high; SPR near 44 year low

highest natural gas closing price since February 6th after force majeure on the leaking Mountaineer XPress pipeline in WV; US Strategic Petroleum Reserve at its lowest since October 1982; distillates imports at a 52 week low;  US rig count at a 27 month high

US oil prices fell for ​t​he second time in eight weeks on hopes that negotiations to end the war in Iran would proceed with Iran’s president in New York for a United Nations General Assembly meeting….after inching up 0.2% to $100.30 a barrel last week after drone damage to Saudi Arabia’s East-West pipeline led to a suspension of their exports through the Red Sea, the contract price for the benchmark US light sweet crude for October delivery fell to its lowest level in more than a week during Asian trading early on Monday, as traders hoped diplomatic efforts over the Iran war would gain momentum this week amid a United Nations meeting in New York, and ​w​as down 3.4% during early trading in New York, after Trump reportedly told a news channel that he would “probably” be open to a sit-down meeting with Iranian President Masoud Pezeshkian, and settled $4.52 or 4.5% lower at $95.78 a barrel as traders hoped for a breakthrough in peace talks this week during the United Nations meeting….that October oil contract price rose slightly on its last day of trading as markets opened in Asia on Tuesday, as traders watched for developments in possible US-Iran talks on the sidelines of the United Nations General Assembly meetings in New York, and was up by more than 1.1% as markets started trading in London, as traders continued assessing the situation in the Middle East and the prospects for its settlement. but turned sharply lower during early trade in New York as markets awaited hints at diplomacy from U.N. general assembly speeches by U.S. President Donald Trump and his Iranian counterpart Masoud Pezeshkian that could signal a reopening of the Strait of Hormuz to oil and other energy shipments, and expired $1.19 lower at $94.59 a barrel amid the prospects increased oil flow from the Middle East, after Iran signaled it could reopen the Strait of Hormuz and Saudi Arabia was set to resume its oil exports from its Red Sea port of Yanbu, while the price of the benchmark US oil contract for November delivery settled $1.85 lower at $90.52 a barrel…with markets now citing the price of that benchmark US light sweet crude for November delivery, prices fell across global markets on Wednesday after US President Trump said American and Iranian representatives had held “very good” and productive discussions on the sidelines of the United Nations General Assembly in New York, but reversed initial losses during early US trading amid persistent Middle East supply risks, as traders weighed the prospect of improved Saudi export flows and U.S.-Iran diplomacy against the potential for further disruptions, and were later up nearly 5% amid concerns that the Middle East conflict could widen, even after U.S. and Iranian officials held their first talks since June, as signals suggested little willingness to reach a ceasefire, and settled $1.64 higher at $92.16 a barrel as traders evaluated Iranian President Masoud Pezeshkian’s vow never to surrender, after US President Trump warned he could “annihilate” Iran….oil prices fell during early Asian trading on Thursday, as Iran said it remained open to diplomacy to end the war with the United States, but then jumped about 4% to a one-week high during US trading after Yemen’s Houthis fired missiles at Saudi Arabia and diplomatic talks between the US and Iran showed little sign of progress, and settled $2.45 higher at $94.61 a barrel as the Houthi missile attack on Saudi Arabia revived fears of supply disruptions, while reports of US-Iran talks on reopening the Strait of Hormuz subsequently tempered the gains…oil prices edged lower during Asian trading on Friday as markets weighed conflicting factors, including the possibility of a ceasefire between the United States and Iran and ongoing Houthi attacks on Saudi Arabia, then slid further Friday morning in New York on revived direct negotiations between the U.S. Iranian diplomats in New York for the first time since June, sparking market optimism about a deal that could reopen the Strait of Hormuz, and settled $2.20 lower at $92.41 a barrel after Iran’s foreign minister offered to reopen the Strait of Hormuz in seven days if the US meets his terms….oil prices thus finished 7.9% lower for the week, while the price of the benchmark US oil contract for November delivery, which had ended the prior week at $96.08 a barrel, finished this one 3.8% lower..

meanwhile, natural gas prices finished higher for the sixth time in seven weeks on falling production from domestic wells as another smaller than normal injection of gas into storage further reduced the surplus inventory….after rising 2.9% to $2.912 per mmBTU last week on an increase in LNG demand and on forecasts for warmer than normal forecasts through early October, the price of the benchmark natural gas contract for October delivery opened 6.1 cents lower on Monday, after losing ground in pre-market trading, as cooler temperatures took hold and scheduled LNG maintenance limited exports, then ​d​ropped again after a brief early rally to settle 7.6 cents lower at $2.836 per mmBTU amid healthy supply readings and forecasts for benign weather through the final stretch of September and into early October….the front-month natural gas contract opened 1.1 cents lower on Tuesday, but rose ​through the morning as production levels from US wells fell to nearly​ a two-month low, and settled 12.9 cents higher at $2.965 per mmBTU on bargain buying and the bullish natural gas market fundamentals of easing production and a shrinking storage inventory surplus…natural gas prices opened 7.4 cents higher on Wednesday and rose to an eleven-week intraday high of $3.042 ahead of 9:05 AM, then fell back to reach an intraday low of $2.959 at 10:15 AM, as traders struggled to find a balance between cooling weather, lower production, and short covering, then arced higher once again to settled 5.8 cents higher at $3.023 per mmBTU as short covering and colder overnight forecasts outweighed the weakest power burn since Labor Day….natural gas prices started 2.2 cents lower on Thursday and traded near $3.005 through the day’s opening hour, then spiked upwards as the historically bullish weekly storage report hit the wire and hit a 14-month intraday high of $3.317 at 1:45 PM, before closing 27.4 cents higher $3.297 per mmBTU, the biggest one-day gain since January, on a seasonally lean storage print, lighter production and hints of heating demand, and on a natural gas leak on the Mountaineer XPress pipeline in West Virginia…natural gas futures retreated early Friday from the three-month high of Thursday, as traders weighed how long a force majeure on the leaking Mountaineer XPress (MXP) pipeline could hold back Lower 48 production, and were down by 14.5 cents by midday, as the pipeline leak in the heart of the Appalachian natural gas patch exacerbated typical expiration week volatility, as nearly 1.8 million Dth/d of firm takeaway capacity was expected to remain offline through the weekend, and settled 10.1 cents lower at $3.196 per mmBTU, as traders took profits following a pronounced rally that sent prices to their highest levels since the peak of summer, leaving October natural gas prices up 9.8% on the week…

The EIA’s natural gas storage report for the week ending September 18th indicated that the amount of working natural gas held in underground storage rose by 53 billion cubic feet to 3,351 billion cubic feet by the end of the week, which left our natural gas supplies 146 billion cubic feet, or 4.2% below the 3,497 billion cubic feet of gas that were in storage on September 18th of last year, but 95 billion cubic feet, or 2.9% above the five-year average of 3,256 billion cubic feet of natural gas that had typically been in working storage as of the 18th of September over the most recent five years….the 53 billion cubic foot injection into natural gas storage for the cited week was a little less than the 56 billion cubic foot injection into storage that the market had been expecting ahead of the report, while it was considerably less than the 77 billion cubic foot of gas that were injected into natural gas storage during the corresponding week of 2025, and also less than the average 76 billion cubic foot injection into natural gas storage that had been typical for the ​t​hird week in September over the past five years…

The Latest US Oil Supply and Disposition Data from the EIA

US oil data from the US Energy Information Administration for the week ending September 18th showed that after a big drop in our oil exports and a big decrease in our oil refining, we had surplus oil left to add to our stored crude supplies for the ​j​ust the second time in twenty-two weeks, and for the 27th time in sixty-nine weeks, with a decrease in the domestic supply of oil the EIA could not account for also contributing…. Our imports of crude oil fell by an average of 1,181,000 barrels per day to 5,877,000 barrels per day, after rising by an average of 234,000 barrels per day during the prior week, while our exports of crude oil fell by an average of 1,550,000 barrels per day to average 3,281,000 barrels per day, which, when used to offset our imports, meant that the net of our trade of oil worked out to an import average of 2,596,000 barrels of oil per day during the week ending September 11th, an average of 369,000 more barrels per day than the net of our imports minus our exports during the prior week… At the same time, transfers to our oil supplies from Alaskan gas liquids, from natural gasoline, from condensate, and from unfinished oils were 6,000 barrels per day lower than the prior week at 428,000 barrels per day, while during the same week, production of crude from US wells was 5,000 barrels per day lower at 13,939,000 barrels per day.  Hence, our daily supply of oil from the net of our international trade in oil, from transfers, and from domestic well production appears to have averaged a total of 16,963,000 barrels per day during the September 18th reporting week…

Meanwhile, US oil refineries reported they were processing an average of 16,811,000 barrels of crude per day during the week ending September 11th, an average of 519,000 fewer barrels per day than the amount of oil that our refineries reported they were processing during the prior week, while over the same period, the EIA’s surveys indicated that a net of 366,000 barrels of oil per day were being added to the supplies of oil stored in the US… So, based on that reported & estimated data, the crude oil figures provided by the EIA appear to indicate that our total working supply of oil from net imports, from transfers, and from oilfield production during the week ending September 18th averaged a rounded ​2​15,000 fewer barrels per day than what was added to storage plus our oil refineries reported they used during the week.  To account for the difference between the apparent supply of oil and the apparent disposition of it, the EIA just plugged a [ +215,000 ] barrel per day figure onto line 16 of the weekly U.S. Petroleum Balance Sheet, in order to make the reported data for the supply of oil and for the consumption of it balance out, a fudge factor that they label in their footnotes as “unaccounted for crude oil”, thus indicating there must have been a error or omission of that amount in the week’s oil supply & demand figures that we have just transcribed…. Since 577,000 barrels per day of oil supplies could not be accounted for in the prior week’s EIA data, that means there was a 362,000 barrel per day difference between this week’s oil balance sheet error and the EIA’s crude oil balance sheet error from a week ago, and hence the changes to supply and demand from that week to this one that are indicated by this week’s report are somehow off by that much, and therefore aren’t very useful….However, since most oil traders react to the figures in these weekly EIA reports as if they were gospel, and since these weekly figures therefore often drive oil pricing and hence decisions to drill or complete oil wells, we’ll continue to report this data just as it’s published, and just as it’s watched & believed to be reasonably reliable by most everyone in the industry…(for more on how this weekly oil data is gathered, and the possible reasons for that “unaccounted for” oil supply, see this EIA explainer….also see this March 2023 twitter thread from an EIA administrator addressing these ongoing weekly errors, and what they had once hoped to do about it).

This week’s rounded 366,000 barrel per day average increase in our overall crude oil inventories came as an average of 424,000 barrels per day were added to our commercial stocks of crude oil, while 58,000 barrels per day were being pulled out of our Strategic Petroleum Reserve, the twenty-sixth consecutive Iran war related withdrawal from the SPR, which left the SPR level at 284,552,000 barrels, the lowest since it was initially being filled in October 1982….Further details from the weekly Petroleum Status Report (pdf) indicated that the 4 week average of our oil imports fell to 6,632,000 barrels per day last week, which was 5.3% more than the 6,300,000 barrel per day average that we were importing over the same four-week period last year, while the four week average of our exports fell to 4,003,000 barrels per day last week, which was 2.3% less than the 4,098,000 barrel per day average that we were exporting last year year at this time… This week’s crude oil production was reported to be 5,000 barrels per day lower at 13,939,000 barrels per day as the EIA’s estimate of the output from wells in the lower 48 states was 3,000 barrels per day higher at 13,484,000 barrels per day, while Alaska’s oil production was 8,000 barrels per day lower at 463,000 barrels per day…US crude oil production had reached a pre-pandemic high of 13,100,000 barrels per day during the week ending March 13th 2020, so this week’s reported oil production figure was 6.4% higher than that of our pre-pandemic production peak, and was also 43.7% above the pandemic low of 9,700,000 barrels per day that US oil production had fallen to during the third week of February of 2021.

US oil refineries were operating at 94.0% of their capacity while processing those 16,811,000 barrels of crude per day during the week ending September 18th, down from 96.8% the prior week, but still higher than normal for mid-September….the 16,811,000 barrels of oil per day that were refined that week were 2.0% more than the 16,476,000 barrels of crude that were being processed daily during the week ending September 19th of 2025, and 1.8% more than the 16,513,000 barrels that were being refined during the pre-pandemic week ending September 20th, 2019, when our refinery utilization rate was at 89.8%, which was a bit below the pre-pandemic normal utilization rate for this time of year…

With the decrease in the amount of oil that was being refined this week, gasoline output from our refineries was also lower, decreasing by 54,000 barrels per day to 9,590,000 barrels per day during the week ending September 18th, after our refineries’ gasoline output had increased by 366,000 barrels per day during the prior week… This week’s gasoline production was 1.2% less than the 9,707,000 barrels of gasoline that were being produced daily over the week ending September 19th of last year, and 6.3% les than the gasoline production of 10,240,000 barrels per day seen during the prepandemic week ending September 20th, 2019….at the same time, our refineries’ production of distillate fuels (diesel fuel and heat oil) decreased by 68,000 barrels per day to 5,159,000 barrels per day, after our distillates output had decreased by 121,000 barrels per day during the prior week.  Even after those production decreases, our distillates output was 3.5% more than the 4,984,000 barrels of distillates that were being produced daily during the week ending September 19th of 2025, and 3.2% more than the 5,000,000 barrels of distillates that were being produced daily during the pre-pandemic week ending September 20th, 2019….

With this week’s decrease in our gasoline production, our supplies of gasoline in storage at the end of the week fell for the twenty-fifth time in thirty-two weeks, decreasing by 1,686,000 barrels to 206,046,000 barrels during the week ending September 18th, after our gasoline inventories had increased by 794,000 barrels to a 42 week low during the prior week.  Our gasoline supplies fell this week because the amount of gasoline supplied to US users rose by 49,000 barrels per day to 8,847000 barrels per day, and because our imports of gasoline fell by 136,000 barrels per day to 401,000 barrels per day, while our exports of gasoline fell by 117,000 barrels per day to 838,000 barrels per day… After fifty-five gasoline inventory withdrawals over the past eighty-three weeks, our gasoline supplies were 4.9% lower than last September 19th’s gasoline inventories of 216,569,000 barrels, and about 6% below the five year average of our gasoline supplies for this time of year…

After this week’s decrease in distillates production, our supplies of distillates fell for the fourteenth time in thirty-two weeks, decreasing by 428,000 barrels to 107,431,000 barrels during the week ending September 18th, after our distillates supplies had increased by 1,585,000 barrels during the prior week… Our distillates supplies fell this week because the amount of distillates supplied to US markets, an indicator of domestic demand, rose by 474,000 barrels per day to 3,975,000 barrels per day, and even as our exports of distillates fell by 283,000 barrels per day to 1,331,000 barrels per day, while our imports of distillates fell by 29,000 barrels per day to a ​fifty two week low of 85,000 barrels per day… After 29 withdrawals from distillates inventories over the past 62 weeks, our distillates supplies at the end of the week were 12.7% lower than the 122,999,000 barrels of distillates that we had in storage on September 19th of 2025, and were about 12% below the five year average of our distillates inventories for this time of the year…

Finally, after the drop in our oil exports and the slowdown in our refining, our commercial supplies of crude oil in storage rose for the 11th time in twenty-six weeks, and for the 27th time over the past year, increasing by 2,969,000 barrels over the week, from 423,429,000 barrels on September 11th to 426,398,000 barrels on September 18th, after our commercial crude supplies had decreased by 640,000 barrels over the prior week….After this week’s increase, our commercial crude oil inventories were about 2% above recent five-year average of commercial oil supplies for this time of year, and they were about 30% above the average of our available crude oil stocks as of the fourth weekend of September over the 5 years at the beginning of the past decade, with the difference between those comparisons arising because it wasn’t until early 2015 that our oil inventories had first topped 400 million barrels. After our commercial crude oil inventories had jumped to record highs during the Covid lockdowns in the Spring of 2020, then jumped again after February 2021’s winter storm Uri froze off US Gulf Coast refining, but then fell sharply due to increased exports to Europe following the onset of the Ukraine war, only to jump again following the Christmas 2022 refinery freeze-offs, changes in our commercial crude inventories had been less extreme up until the onset of the Iran war, when they were initially built up to a three year high by mid-April, before falling to the lowest in nearly eight years by late July…This week’s increase was the first in four weeks, and as of September 18th our commercial crude inventories were 2.8% above the 414,754,000 barrels of oil we had in commercial storage on September 19th of 2025, and were 3.2% more than the 413,042,000 barrels of oil that we had in storage on September 20th of 2024, and 2.4% more than the 416,287,000 barrels of oil we had left in commercial storage on September 22nd of 2023…

This Week’s Rig Count

The US rig count increased by four ​to a 28 month high over the week ending September 25th, as the number of rigs targeting oil was up by three, the count of rigs targeting natural gas was up by one, and miscellaneous rigs were unchanged…for a quick snapshot of this week’s rig count, we are again including below a screenshot of the rig count summary table from Baker Hughes…in the table below, the first column shows the active rig count as of September 25th, the second column shows the change in the number of working rigs between last week’s count (September 18th) and this week’s (September 25th) count, the third column shows last week’s September 18th active rig count, the 4th column shows the change between the number of rigs running on Friday and the number running on the Friday of the same week of a year ago, and the 5th column shows the number of rigs that were drilling at the end of that reporting period a year ago, which in this week’s case was Friday, the 26th of September, 2025…

++

August’s retail sales, industrial production, and new housing construction; July’s business inventories

Widely watched economic reports released last week included the Retail Sales Report for August and the Business Sales and Inventories report for July, both from the Census bureau, the August Industrial Production and Capacity Utilization report from the Fed, and the August report on New Residential Construction, also from the Census bureau.…in addition, on Friday the Bureau of Labor Statistics released the Regional and State Employment and Unemployment Report for August, a report which breaks down the two employment surveys from the monthly national jobs report by state and by region….while the text of that report provides a useful summary of this data, the serious statistics aggregation can be found in the tables linked at the end of the report, where one can find the civilian labor force data and the change in payrolls by industry for each of the 50 states, the District of Columbia, Puerto Rico, and the Virgin Islands…

This week also saw the release of the first two regional Fed manufacturing surveys for September: the Empire State Manufacturing Survey from the New York Fed, which covers New York, northern New Jersey, a county in Connecticut, and Puerto Rico, reported their headline general business conditions index fell from a four year high of +20.6 in August to +7.6 in September, meaning that a smaller plurality of Second District manufacturers reported improvement in business conditions in September than a month earlier, while the Philadelphia Fed Manufacturing Survey for September, covering most of Pennsylvania, southern New Jersey, and Delaware, reported their broadest diffusion index of manufacturing conditions fell from from +47.4 in August. to +37.8 in September, which they explain was because 45.3% of the respondents reported an increase in general business activity, while 7.5% reported a decrease from August to September….

Retail Sales Rose 1.2% in August after July’s Sales Were Revised 0.1% Higher

Seasonally adjusted retail sales rose 1.2% in August after retail sales for July were revised 0.1% higher…the Advance Retail Sales Report for August (pdf) from the Census Bureau estimated that our seasonally adjusted retail and food services sales totaled $773.9 billion during the month, which was up 1.2 percent (±0.4%) from July’s revised sales of $764.5 billion, and 6.0 percent (±0.5 percent) higher than the adjusted sales in August of last year….estimated unadjusted sales, extrapolated from a survey of a small sampling of retailers, indicated actual sales were virtually unchanged, rising from $784,603 million in July to $784,846 million in August, while they were up 5.3% from the $745,232 million of sales in August of last year…

July’s seasonally adjusted sales were revised more than 0.1% higher, from the $763.6 billion reported last month to $764.5 billion, while June sales were revised nearly 0.1% higher, from $768.1 billion to $768.6 billion, and as a result of that small difference, the rounded June to July change was revised from down 0.6 percent (±0.4 percent) to down 0.5 percent (±0.2 percent)…the upward revision to June’s sales means that the annualized rate of nominal 2nd quarter sales was roughly $1.9 billion higher than was previously reported, which would probably add 0.02 or 0.03 percentage points to the 2nd quarter GDP growth rate when the 3rd GDP estimate is published next week…

Included below is a copy of the table of the monthly and yearly percentage changes in retail sales by business type taken from the August Census Marts pdf….the first double column below gives us the seasonally adjusted percentage change in sales for each type of retail business from July to August in the first sub-column, and then the year over year percentage change for those businesses since last August in the 2nd column; the second pair of columns gives us the revision of last month’s July advance monthly estimates (now called “preliminary”) as revised in this report, likewise for each business type, with the June to July change under “Jun 2026 (r)revised”, and the revised July 2025 to July 2026 percentage change for each business type in the last column shown…if you want to see which July sales were revised, our copy of the table of last month’s advance July sale estimates, before this month’s revision, is here….

To compute August’s real personal consumption of goods data for national accounts from this August retail sales report, the BEA will initially use the corresponding price changes from the August consumer price index, which we reviewed last week, to adjust nominal dollar sales for inflation….to estimate what they will find, we’ll start by pulling out the usually volatile sales of gasoline from the other totals….from the third line on the above table, we can see that August retail sales excluding the 3.1% increase in sales at gas stations were up by 1.1%….then, by subtracting the figures representing the 0.4% increase in grocery & beverage store sales and the 1.2% increase in food services sales from the figures behind that percentage, we find that core retail sales were up by 1.17% month over month…since the August CPI report showed that the the composite price index of all goods less food and energy goods was 0.1% higher in August, we can thus figure that real retail sales excluding food and energy, or real core PCE, will show an increase of about 1.07% for the month…however, the actual adjustment in national accounts for each of the types of sales shown above will vary by the change in the related price index…for instance, while nominal sales at motor vehicle and parts dealers were 0.6% higher in August, the price index for transportation commodities other than fuel was was 0.3% higher, as average prices for new vehicles rose 0.3%, which means that real sales of automobiles only rose by about 0.3%…

In addition to figuring those core retail sales, to make a complete estimate of August real PCE, we should also adjust food and energy retail sales for their price changes separately, just as the BEA will do…the August CPI report showed that the food price index rose 0.1% in August, as the price index for food purchased for use at home was unchanged, while average restaurant prices were 0.3% higher….with no change in grocery prices, the 0.4% nominal increase in grocery & beverage store sales nominal sales would be a real 0.4% increase…on the other hand, 1.2% increase in sales at bars and restaurants, once adjusted for 0.3% higher prices, suggests that real sales at bars and restaurants were up by about 0.9% for the month…meanwhile, while nominal sales at gas stations were up 3.1%, there was a 4.1% increase in the price index for gasoline, which would suggest real sales of gasoline were almost 1.0% lower, with a caveat that gasoline stations do sell more than gasoline, products which should not be adjusted with gasoline prices, so the decrease in real sales at gas stations was likely a bit less than our blunt estimate….by appropriately weighing and averaging the real sales changes that we have thus estimated back together, and excluding food services, we can then estimate that the income and outlays report for August will show that real personal consumption of goods rose by around 0.8% in August, after falling by a revised 0.5% in July and after rising by a revised 0.7% in June, and rising by 0.3% in May, but after falling 0.2% April, after rising by 0.9% in March, rising by 0.8% in February and falling by 0.6% in January….at the same time, 0.9% increase in real sales at bars and restaurants would have a modest positive impact on July’s real personal consumption of services…

Industrial Production Unchanged in August as Increased Utility Output Offsets Reduced Manufacturing

The Fed’s August report on Industrial production and Capacity Utilization indicated that industrial production was unchanged in August, after rising by a revised 0.2% in both June and July, and is now up 1.4% from a year ago….the industrial production index, with the benchmark now set for average 2017 production to equal to 100.0, rose to 103.1 in August from 103.0 in July, after July’s index was revised but unchanged from the 103.0 that was reported for July a month ago…at the same time, the June reading for the IP index was revised but unchanged at 102.8, the May reading for the IP index was revised from 102.5 up to 102.6. the April reading for the index was also revised from 102.5 to 102.6, and the March reading for the index was revised down from 101.8 to 101.7. ..

The manufacturing index, which accounts for more than 75% of the total IP index, fell by 0.3%, from 98.43 in July to 98.16  in August, after the June reading for the manufacturing index was revised but unchanged at 98.3, the May reading for the manufacturing index was revised from the 98.0 published last month to 98.1, the April reading for the manufacturing index was also revised from 98.0 to 98.1, and the March manufacturing index was revised up from 97.3 to 97.4….meanwhile, the mining index, which includes oil and gas well drilling, rose 0.1% to 122.7 in August, after the July mining index was revised up from 122.4 to 122.5, leaving the mining index up by 0.3% from a year ago….finally, the utility index, which often fluctuates due to above or below normal temperatures, was rose by 1.8%, from 111.1 in July to 113.0 in August, after the July utility index was revised up from 110.7 to 111.1, and after the June utility index was revised from 110.2 to 110.5…with this August’s air conditioning demand much stronger than that of last year, the utility index is now 6.2% above its year ago reading of 106.4..

This report also provides capacity utilization figures, which are expressed as the percentage of our plant and equipment that was in use during the month, and which indicated that seasonally adjusted capacity utilization for total industry was unchanged at 76.3% in August, with July’s utilization figure unrevised from the 76.3% reported a month ago…capacity utilization of NAICS durable goods production facilities was fell from 76.3% in July to 75.8% in August, while capacity utilization for non-durables producers fell from 75.7% in July to 75.6% in August, after July’s non-durables utilization was revised up from 75.6%…meanwhile, capacity utilization for the mining sector rose to 86.3% in August from 86.2% in July, which was originally reported as 86.1%, while utilities were operating at 71.3% of capacity during August, up from their 70.2% of capacity during July, which was revised up from 70.0%…for more details on capacity utilization by type of manufacturer, see Table 7: Capacity Utilization: Manufacturing, Mining, and Utilities, which shows the historical capacity utilization figures for a dozen types of durable goods manufacturers, 8 classifications of non-durable manufacturers, mining, utilities, and capacity utilization for a handful of other special categories….

Business Sales Rose 0.3% in July, Business Inventories Rose 0.8%

After the release of the August retail sales report, the Census Bureau released the composite Manufacturing and Trade Inventories and Sales report for July (pdf), which incorporates the revised July retail data from that August retail report and the earlier published July factory data and last week’s July wholesale trade report to give us a consolidated picture of the business contribution to the economy for that month….according to the Census Bureau, total manufacturer’s and trade sales were estimated to be valued at a seasonally adjusted $2,120.7 billion in July, up 0.3 percent (±0.2%) from June’s revised sales, and 8.9 percent (±0.3 percent) above July’s sales of a year earlier…note that June’s total sales were concurrently revised up from the originally reported $2,111.3 billion to $2,111.66 billion with this report, which is now shown as down 1.0% from May, revised from the 1.1% decrease reported last month….manufacturer’s sales were up 0.8% to $658,833 million in July, while retail trade sales, which exclude restaurant & bar sales from the revised July retail sales reported earlier, fell 0.7% to $660,638 million, while wholesale sales rose 0.8% to $801,278 million..

Meanwhile, total manufacturer’s and trade inventories, a major component of GDP, were estimated to be valued at a seasonally adjusted $2,764.7 billion at the end of July, up 0.8% (±0.1%) from June, and 3.8 percent (±0.5 percent) higher than in July a year earlier…the value of end of June inventories was revised from the $2,740.2 billion reported last month to $2,742.5 billion, which is now up 0.1% from June…seasonally adjusted inventories of manufacturers were estimated to be valued at $966,871 million, 0.4% higher than in June, inventories of retailers were valued at $838,983 million, 0.8% more than in June, while inventories of wholesalers were valued at $958,854 million at the end of July, up 1.3% from June…

With the original release of the factory inventory data two weeks ago, we judged that the real change in July factory inventories would result in a significant boost to the growth rate of 3rd quarter GDP, because they’d be modestly positive vs the sharply negative real factory inventories in the second quarter; then, with the release of the wholesale inventory last week, we figured that since real July wholesale inventories were also higher, coming after a large 2nd quarter decrease, they’d also be sharply positive for 3rd quarter GDP….meanwhile, since the producer price index for July showed that prices for finished goods were on average 0.4% lower, that means that real retail inventories were roughly 1.2% higher for the month….since real retail inventories saw a modest increase in the second quarter, whether the third quarter’s increase in real inventories adds or subtracts from GDP will depend on which quarter’s inventory increase is ultimately higher…

New Housing Starts and Building Permits Fell in August

The August report on New Residential Construction (pdf) from the Census Bureau indicated that their widely watched estimate of new housing units that were started during the month was at a seasonally adjusted annual rate of 1,275,000, which was 2.6 percent (±12.0 percent)* below July’s revised July estimated annual rate of 1,309,000 housing unit starts, and was 1.2 percent (±10.8 percent)* below last August’s pace of 1,291,000 housing starts a year…the asterisks indicate that the Census does not have sufficient data to determine whether housing starts actually rose or fell from July, or even from August of last year, while the figures in parenthesis indicate the most likely range of the changes indicated; in other words, August’s housing starts could have been up by 9.4% or down by as much as 14.6% from those of July, with a 10% chance that the actual change could have even been outside of that wide range…in this report, the annual rate for July housing starts was revised from the 1,239,000 reported last month to 1,309,000, while June’s housing starts, which were first reported at a 1,427,000 annual rate, were revised from last month’s initial revised figure of a 1,415,000 annual rate to an annual rate of 1,439,000 with this report….

Those annual rates of housing starts reported here were extrapolated from a survey of a small percentage of US building permit offices visited by Census field agents, which estimated that 110,500 housing units were started in August, down from the 118,800 units started in July, and down from 135,300 housing starts in June…of those housing units started in August, an estimated 80,800 were single family homes and 28,600 were units in structures with more than 5 units, up from the revised 77,300 single family starts in July, but down from the 40,500 units started in structures with more than 5 units in July…

The monthly data on new building permits, with a smaller margin of error, is probably a better monthly indicator of new housing construction trends than the volatile and often revised housing starts data…in August, Census estimated new building permits were being issued for a seasonally adjusted annual rate of 1,394,000 housing units, which was 2.7 percent below the revised July rate of 1,433,000 permits, but was 3.5 percent above the rate of building permit issuance in August a year earlier…the annual rate for housing permits issued in July was revised to 1,433,000 units from the annual rate of 1,443,000 reported last month…

Again, these annual estimates for new permits reported here were extrapolated from the unadjusted estimates of canvassing census agents, which showed permits for roughly 117,400 housing units were issued in August, down from the revised estimate of 128,900 new permits issued in July…the August permits included 74,000 permits for single family homes, down from 80,600 in July, and 39,200 permits for housing units in apartment buildings with 5 or more units, down from 43,200 such multifamily permits a month earlier…

(the above is the synopsis that accompanied my regular Sunday morning news links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are chosen from the aforementioned GGO posts, contact me…)

US oil supplies the lowest since March 1984; SPR at 43½ year low; highest Autumn refinery utilization rate since 1998

Strategic Petroleum Reserve at the lowest level since it was being filled in November 1982; total of all US oil inventories, including SPR, are the lowest since March 30th, 1984; US refinery utilization rate is highest for this time of year since 1998

US oil prices managed to increase for a sixth time in seven weeks after drone damage to Saudi Arabia’s East-West pipeline led to a suspension of their exports through the Red Sea…after rising 9.4% to $100.05 a barrel last week after the US “destroyed” at least eight Iranian oil tankers in the Persian Gulf and Gulf of Oman, Iran fired advanced missiles at U.S. warships, and Houthi attacks on Saudi oil facilities forced them to shut down, the contract price for the benchmark US light sweet crude for October delivery jumped more than $3 per barrel as the market opened in Asia on Monday following more strikes on Saudi Arabia and ships in the Persian Gulf on Sunday. and were later up more than 4% on global markets after Arab states in the Gulf called off a meeting with Iran planned for Monday, while Yemen’s Iran-backed Houthis launched a new attack on Saudi Arabia, further jeopardizing global oil supplies. and similarly jumped Monday morning in New York after ​a drone strikes on Friday forced shut Saudi Arabia’s East-West pipeline, which had been transporting 4 million bpd of crude oil to export terminals at the Red Sea, bypassing Iran’s blockade of the Strait of Hormuz, but pulled back ​from ​that four month high to settle $1.34 higher at $101.39 a barrel, as the market weighed the widening conflict in the Middle East against the possibility of a diplomatic resolution​, after President Trump stated that Iran wanted to reach a deal with the U.S…oil prices continued to rally early in Asian trading on Tuesday, as the Iran-aligned Houthis in Yemen expanded their control over the west coast of the Red Sea, while the attacks on a key Saudi onshore oil pipeline triggered concerns about additional disruptions to the already severely disrupted oil supply from the Middle East, but retreated from session highs while remaining above the prior session’s close during morning trading in New York, as escalating Houthi offensive actions in Saudi Arabia continued to command a formidable geopolitical risk premium across energy markets, then bounced off their morning low to settle $4.44 higher at $105.83 per barrel on the news that oil loadings at Saudi Arabia’s Red Sea port of Yanbu had been suspended following the shut in of its East-West Pipeline….oil prices fell on as markets opened in Asia on Wednesday after data from the American Petroleum Institute​ (API) showed that US crude oil, gasoline and distillate inventories all unexpectedly rose last week, and continued to trade lower across global markets after Saudi Arabia began offering additional crude exports through Oman, easing some fears that Saudi exports might fall more sharply, and held those losses in early US trading even after the EIA reported US crude inventories fell a bit for a 3rd week in a row, drastically different from the 7.1​ million barrel build that the API ​had reported, then tumbled to settle $3.40 lower at $102.43 a barrel on the first Fed interest rate hike in three years​, and on reports that Saudi Arabia could restore half of its East-West pipeline capacity ​d​amaged by militia attacks within days …oil prices fell more than 1% in early Asian trading on Thursday, largely on reports that Saudi Arabia was offering additional crude shipments through Oman, easing concerns over potential supply disruptions in the Middle East. and were down more than 3% across global markets, as concerns over immediate supply disruptions in the Middle East eased, and continued to retreat Thursday morning in New York on easing supply concerns amid reports of Saudi Arabia establishing workarounds to the recent supply disruption from the Red Sea, but later bounced off its low and retraced its earlier losses to trade back towards its high, ahead of settling 52 cents lower at $101.91 a barrel as reports of additional Saudi crude cargoes leaving Oman’s Sohar port eased supply concerns….oil prices fell for a third consecutive day in Asia on Friday amid Saudi Arabia’s efforts to resume exports after the “East-West” oil pipeline was damaged, but were mixed in early US trading as reports that the Saudi East-West pipeline could soon restart at half capacity weighed on prices​, while reports that Saudi Aramco would suspend term contracts to Europe next month dampened those bearish effects, but settled $1.61 lower at $100.30 a barrel after China, acting on a request from Saudi Arabia, asked Iran to limit attacks by Houthi rebels on Saudi oil infrastructure that had threatened a second oil export route from the Middle ‌East, but still managed to eke out a 0.2% gain on the week…

meanwhile, natural gas prices finished higher for the fifth time in six weeks on an increase in LNG demand and on forecasts for warmer than normal forecasts through early October…after falling 4.8% to $2.831 per mmBTU last week as traders shifted their focus to the Autumn “shoulder season”, when little demand for either cooling or heating w​a​s expected, the price of the benchmark natural gas contract for October delivery opened 4.1 cents higher on Monday, then rebounded from an early intraday low, as cooling demand in the South and sustained LNG demand provided support, and continued rising to settle 6.5 cents higher at $2.896 per mmBTU​, as hotter weekend forecasts pushed September cooling demand to within striking distance of the all-time record, while near-record production and an approaching late month cooldown capped gains…the Octob​e​r natural gas contract opened 4.4 cents higher on Tuesday and rose to an intraday high of $2.952 ahead of 10:00 AM, as traders continued to appr​e​ciate strong cooling and LNG demand against stout storage levels and impending should-month weather, and hung on to settle 2.3 cents higher at $2.919 per mmBTU as record-challenging heat across the South propped up demand, even as softer LNG feedgas and a bearish shift in forecasts beyond the weekend kept the rally in check….natural gas prices opened 4.0 cents higher on Wednesday and hit an intraday high of $2.968 within minutes of the opening bell, then backed off as bearish shoulder season sentiment proved to be dominant, and settled 2.8 cents lower at $2.891 per mmBTU as fading shoulder season demand weighed on the nearby contracts despite a final burst of late-summer heat….the October contract opened slightly lower on Thursday, but quickly rose to trade near $2.940 ahead of the weekly storage report, and peaked at an intraday high of $2.959 as the bullish-leaning storage report hit the wire, but then faded to settle 1.0 cent higher at $2.901 per mmBTU after a smaller-than-expected government storage injection reinforced signs that the US inventory cushion is tightening….natural gas futures reversed sharply higher early Friday, erasing a slide driven by cooler late-September forecasts. as LNG feedgas demand rebounded on Cameron LNG’s return, but became rangebound at midday as ebbing weather demand countered LNG momentum​, before settling 1.1 cents higher at $2.912 per mmBTU, as bullish forecasts for more demand next week and an increase in daily flows to LNG export plants offset a bearish increase in output and ample amounts of gas in storage, and thus finished 2.9% higher for the week..

The EIA’s natural gas storage report for the week ending September 11th indicated that the amount of working natural gas held in underground storage rose by 44 billion cubic feet to 3,298 billion cubic feet by the end of the week, which left our natural gas supplies 122 billion cubic feet, or 3.6% below the 3,420 billion cubic feet of gas that were in storage on September 11th of last year, but 118 billion cubic feet, or 3.7% above the five-year average of 3,180 billion cubic feet of natural gas that had typically been in working storage as of the 11th of September over the most recent five years….the 44 billion cubic foot injection into natural gas storage for the cited week was a little less than the 47 billion cubic foot injection into storage that the market had been expecting ahead of the report, ​a​nd it was much less than the 87 billion cubic foot of gas that were injected into natural gas storage during the corresponding week of 2025, and also less than the average 74 billion cubic foot injection into natural gas storage that had been typical for the first week in September over the past five years…

The Latest US Oil Supply and Disposition Data from the EIA

US oil data from the US Energy Information Administration for the week ending September 11th showed that even after an increase in our oil imports and decrease in our oil refining, a big jump in our oil exports meant we still had to pull oil out of our stored crude supplies for the twentieth time in twenty-one weeks, and for the 42nd time in sixty-eight weeks, despite an increase in the domestic supply of oil the EIA could not account for…. Our imports of crude oil rose by an average of 234,000 barrels per day to 7,058,000 barrels per day, after rising by an average of 53,000 barrels per day during the prior week, while our exports of crude oil rose by an average of 1,414,000 barrels per day to average 4,831,000 barrels per day, which, when used to offset our imports, meant that the net of our trade of oil worked out to an import average of 2,227,000 barrels of oil per day during the week ending September 11th, an average of 1,180,000 fewer barrels per day than the net of our imports minus our exports during the prior week… At the same time, transfers to our oil supplies from Alaskan gas liquids, from natural gasoline, from condensate, and from unfinished oils were 454,000 barrels per day lower than the prior week at 433,000 barrels per day, while during the same week, production of crude from US wells was 3,000 barrels per day lower at 13,944,000 barrels per day.  Hence, our daily supply of oil from the net of our international trade in oil, from transfers, and from domestic well production appears to have averaged a total of 16,604,000 barrels per day during the September 11th reporting week…

Meanwhile, US oil refineries reported they were processing an average of 17,330,000 barrels of crude per day during the week ending September 11th, an average of 256,000 fewer barrels per day than the amount of oil that our refineries reported they were processing during the prior week, while over the same period, the EIA’s surveys indicated that a total of 149,000 barrels of oil per day were being pulled from the supplies of oil stored in the US… So, based on that reported & estimated data, the crude oil figures provided by the EIA appear to indicate that our total working supply of oil from storage, from net imports, from transfers, and from oilfield production during the week ending September 11th averaged a rounded 577,000 fewer barrels per day than what our oil refineries reported they used during the week.  To account for the difference between the apparent supply of oil and the apparent disposition of it, the EIA just plugged a [ +577,000 ] barrel per day figure onto line 16 of the weekly U.S. Petroleum Balance Sheet, in order to make the reported data for the supply of oil and for the consumption of it balance out, a fudge factor that they label in their footnotes as “unaccounted for crude oil”, thus indicating there must have been a error or omission of that amount in the week’s oil supply & demand figures that we have just transcribed…. Moreover, since 888,000 barrels per day of demand for could not be accounted for in the prior week’s EIA data, that means there was a 1,465,000 barrel per day difference between this week’s oil balance sheet error and the EIA’s crude oil balance sheet error from a week ago, and hence the changes to supply and demand from that week to this one that are indicated by this week’s report are somehow off by that much, and therefore are complete nonsense….However, since most oil traders react to the figures in these weekly EIA reports as if they were gospel, and since these weekly figures therefore often drive oil pricing and hence decisions to drill or complete oil wells, we’ll continue to report this data just as it’s published, and just as it’s watched & believed to be reasonably reliable by most everyone in the industry…(for more on how this weekly oil data is gathered, and the possible reasons for that “unaccounted for” oil supply, see this EIA explainer….also see this March 2023 twitter thread from an EIA administrator addressing these ongoing weekly errors, and what they had once hoped to do about it).

This week’s rounded 149,000 barrel per day average decrease in our overall crude oil inventories came as an average of 91,000 barrels per day were being pulled out of our commercial stocks of crude oil, while 58,000 barrels per day were being pulled out of our Strategic Petroleum Reserve, the twenty-fifth consecutive Iran war related withdrawal from the SPR, but the smallest yet, which left the SPR level at 284,957,000 barrels, the lowest since it was initially being filled in November 1982….with both commercial oil and the SPR both down again this week, that left the Total of all US Oil Supplies at 708,386,000 barrels, down 18.6% from 870,774,000 barrels on April 17th, and the lowest since March 30th, 1984….

Further details from the weekly Petroleum Status Report (pdf) indicated that the 4 week average of our oil imports rose to 6,703,000 barrels per day last week, which was 7.5% more than the 6,235,000 barrel per day average that we were importing over the same four-week period last year, while the four week average of our exports rose to 4,131,000 barrels per day last week, which was 5.1% more than the 3,929,000 barrel per day average that we were exporting last year year at this time… This week’s crude oil production was reported to be 3,000 barrels per day lower at 13,944,000 barrels per day as the EIA’s estimate of the output from wells in the lower 48 states was 9000 barrels per day lower at 13,481,000 barrels per day, while Alaska’s oil production was 6,000 barrels per day higher at 463,000 barrels per day…US crude oil production had reached a pre-pandemic high of 13,100,000 barrels per day during the week ending March 13th 2020, so this week’s reported oil production figure was 6.4% higher than that of our pre-pandemic production peak, and was also 43.8% above the pandemic low of 9,700,000 barrels per day that US oil production had fallen to during the third week of February of 2021.

US oil refineries were operating at 96.8% of their capacity while processing those 17,330,000 barrels of crude per day during the week ending September 11th, down from 97.8% the prior week, but still the highest September refinery utilization rate since 1998….the 17,330,000 barrels of oil per day that were refined that week were 5.5% more than the 16,424,000 barrels of crude that were being processed daily during the week ending September 12th of 2025, and 3.7% more than the 16,707,000 barrels that were being refined during the pre-pandemic week ending September 13th, 2019, when our refinery utilization rate was at 91.2%, which was a bit below the pre-pandemic normal utilization rate for this time of year…

Even with the decrease in the amount of oil that was being refined this week, gasoline output from our refineries was somewhat higher, increasing by 336,000 barrels per day to 9,644,000 barrels per day during the week ending September 11th, after our refineries’ gasoline output had decreased by 587,000 barrels per day during the prior week… This week’s gasoline production was 2.5% more than the​ 9​,407,000 barrels of gasoline that were being produced daily over the week ending September 12th of last year, and 2.0% more than the gasoline production of 9,451,000 barrels per day seen during the prepandemic week ending September 13th, 2019….on the other hand, our refineries’ production of distillate fuels (diesel fuel and heat oil) decreased by 121,000 barrels per day to 5,227,000 barrels per day, after our distillates output had increased by 222,000 barrels per day during the prior week.  Even after that production decrease, our distillates output was 5.5% more than the 4,955,000 barrels of distillates that were being produced daily during the week ending September 12th of 2025, and 2.3% more than the 5,109,000 barrels of distillates that were being produced daily during the pre-pandemic week ending September 13th, 2019….

With this week’s increase in our gasoline production, our supplies of gasoline in storage at the end of the week rose for the seventh time in thirty-one weeks, increasing by 794,000 barrels to 207,732,000 barrels during the week ending September 11th, after our gasoline inventories had increased by 1,269,000 barrels to a 42 week low during the prior week.  Our gasoline supplies rose by less this week because the amount of gasoline supplied to US users rose by 247,000 barrels per day to 8,798,000 barrels per day, and because our exports of gasoline rose by 83,000 barrels per day to 955,000 barrels per day while our imports of gasoline rose by 73,000 barrels per day to 537,000 barrels per day,… But after fifty-four gasoline inventory withdrawals over the past eighty-two weeks, our gasoline supplies were 4.6% lower than last September 12th’s gasoline inventories of 217,650,000 barrels, and about 5% below the five year average of our gasoline supplies for this time of year…

After this week’s decrease in distillates production, our supplies of distillates rose for the eighteenth time in thirty-one weeks, increasing by 1,585,000 barrels to 107,859,000 barrels during the week ending September 11th, after our distillates supplies had increased by 2,087,000 barrels during the prior week… Our distillates supplies rose again this week as the amount of distillates supplied to US markets, an indicator of domestic demand, fell by 177,000 barrels per day to 3,501,000 barrels per day, and even as our exports of distillates rose 58,000 barrels per day to 1,614,000 barrels per day, while our imports of distillates fell by 71,000 barrels per day to 114,000 barrels per day… After 28 withdrawals from distillates inventories over the past 61 weeks, our distillates supplies at the end of the week were 13.5% lower than the  124,684,000 barrels of distillates that we had in storage on September 12th of 2025, and were about 13% below the five year average of our distillates inventories for this time of the year…

Finally, after the jump in our oil exports, our commercial supplies of crude oil in storage fell for the 15th time in twenty-six weeks, and for the 26th time over the past year, decreasing by 640,000 barrels over the week, from 424,069,000 barrels on September 4th to 423,429,000 barrels on September 11th, after our commercial crude supplies had decreased by 391,000 barrels over the prior week….After this week’s decrease, our commercial crude oil inventories were about 1% above recent five-year average of commercial oil supplies for this time of year, while they were still about 28% above the average of our available crude oil stocks as of the second weekend of September over the 5 years at the beginning of the past decade, with the difference between those comparisons arising because it wasn’t until early 2015 that our oil inventories had first topped 400 million barrels. After our commercial crude oil inventories had jumped to record highs during the Covid lockdowns in the Spring of 2020, then jumped again after February 2021’s winter storm Uri froze off US Gulf Coast refining, but then fell sharply due to increased exports to Europe following the onset of the Ukraine war, only to jump again following the Christmas 2022 refinery freeze-offs, changes in our commercial crude inventories had been less extreme up until the onset of the Iran war, when they were initially built up to a three year high by mid-April, before falling to the lowest in nearly eight years by late July…This week’s decrease was the third consecutive after four increases, and as of September 11th our commercial crude inventories were 1.9% above the 415,361,000 barrels of oil we had in commercial storage on September 12th of 2025, and were 1.4% more than the 417,513,000 barrels of oil that we had in storage on September 13th of 2024, and 1.1% more than the 418,456,000 barrels of oil we had left in commercial storage on September 15th of 2023…

This Week’s Rig Count

The US rig count increased by four over the week ending September 18th, as the number of rigs targeting oil was up by two, the count of rigs targeting natural gas was up by two, and miscellaneous rigs were unchanged…for a quick snapshot of this week’s rig count, we are again including below a screenshot of the rig count summary table from Baker Hughes…in the table below, the first column shows the active rig count as of September 18th, the second column shows the change in the number of working rigs between last week’s count (September 11th) and this week’s (September 18th) count, the third column shows last week’s September 11th active rig count, the 4th column shows the change between the number of rigs running on Friday and the number running on the Friday of the same week of a year ago, and the 5th column shows the number of rigs that were drilling at the end of that reporting period a year ago, which in this week’s case was Friday, the 19th of September, 2025…

++

August’s consumer and producer prices and existing home sales; July’s wholesale trade

Major reports released last week included the August Consumer Price Index and the August Producer Price Index, both from the Bureau of Labor Statistics, the Wholesale Trade, Sales and Inventories report for July from the Census Bureau, and the Existing Home Sales Report for August from the National Association of Realtors……this week also saw the Consumer Credit Report for July from the Fed, which indicated that overall consumer credit, a measure of non-real estate debt, expanded by a seasonally adjusted $20.1 billion in July, or at a 4.2% annual rate, as non-revolving credit expanded at a 4.8% rate to $3,829.0 billion, while revolving credit outstanding grew at a 2.5% rate to $1,357.2 billion…

CPI Rose 0.4% in August on Higher Prices for Fuel, Housing, and Phone Service

The consumer price index was 0.4% higher in August, as higher prices for fuel, rent, new and used cars and trucks, car and truck rentals, vehicle maintenance and repairs, motor oil, air fares, lodging away from home, eating out, school tuition and fees, appliances, computers, telephone services, TVs, sports vehicles, and trash collection were partly offset by lower prices for utilities, furniture, smartphones, toys, motor vehicle and health insurance, internet services, tax preparation, moving expenses, and admissions to entertainment venues ….the Consumer Price Index Summary from the Bureau of Labor Statistics indicated that the weighted average of seasonally adjusted prices for consumer goods and services was 0.4% higher in August, after being 0.1% higher in July, 0.4% lower in June, 0.5% higher in May, 0.6% higher in April, 0.9% higher in March, 0.3% higher in February, 0.2% higher in January, 0.3% higher in December, after being 0.2% higher the two months ending in November 0 f 2025, 0.3% higher last September, and 0.3% higher in August of last year….

The unadjusted CPI-U index, which was originally set to have prices of the 1982 to 1984 period equal to 100, rose from 333.918 in July to 334.980 in August, which left it 3.396548% higher than the index reading of 323.976 from August of last year, which is reported as a 3.4% year over year increase, matching the 3.4% year over year increase that was reported for July, with that widely cited year over year change simply reflecting the effect of last August’s rounded +0.3% increase dropping out of the comparison and being replaced by the current month’s rounded +0.4%, and not telling us anything more about inflation beyond that….with modestly higher prices for food and much higher prices for energy this month, seasonally adjusted core prices, which exclude both food and energy, were 0.3% higher over the month, as the unadjusted core price index rose from 337.133 in July to 338.041 in August, which left the core index 2.44598% ahead of its year ago reading of 329.970, which is reported as a 2.4% year over year increase, down from the 2.5% year over year core price increase that was reported for June, and well below the 6.6% annual increase reported for September 2022, which had been the largest annual increase in core prices in forty years…

The volatile seasonally adjusted energy price index was 2.1% higher in August, after being 1.5% lower in July, 5.7% lower in June, 3.9% higher in May, 3.8% higher in April, 10.9% higher in March, 0.6% higher in February, 1.5% lower in January, 0.3% higher in December, 1.1% higher in November, 1.5% higher last September, and 0.7% higher last August, and is now 16.3% higher than in August of a year ago….the price index for energy commodities was 4.2% higher in August, on a 4.1% increase in the price index for gasoline and a 10.1% increase in the price index for fuel oil, while the price index for “other energy commodities”, including propane, kerosene, and firewood, averaged out to be 0.9% higher….meanwhile, the price index for energy services was 0.4% lower, after rising 0.3% in July, as the price index for utility gas service was 1.1% lower in August, but is still 4.4% higher than it was a year ago, while the electricity price index was 0.2% lower, after rising 0.1% in July…. energy commodities are now averaging 28.0% above their year ago levels, with gasoline prices averaging 27.4% higher than they were a year ago, while the energy services price index is up 4.0% from last August, as electricity prices are still averaging 3.8% higher than a year ago…

Meanwhile, the seasonally adjusted food price index was 0.1% higher in August, after being 0.1% higher in July, 0.2% higher in June, 0.2% higher in May, 0.5% higher in April, unchanged in March, 0.4% higher in February, 0.2% higher in January, 0.7% higher in December, 0.1% higher over the two months ending last November, and after being 0.2% higher last September and 0.4% higher last August, and is still 3.0% higher than a year ago….the price index for food purchased for use at home was unchanged in August, while the price index for food bought to eat away from home was 0.3% higher, as average prices at fast food outlets was unchanged and average prices at full service restaurants rose 0.4%, while the price index for food at employee sites and schools was 0.6% higher, and prices for other food away from home averaged 0.3% higher…

In the food at home categories, the price index for cereals and bakery products was unchanged, as average bread prices rose 0.1%, the price index for flour and prepared flour mixes rose 1.4%, and the price index for cookies was 1.9% higher, while the price index for breakfast cereal fell 1.6%, and the price index for fresh cakes and cupcakes was 3.5% lower.…meanwhile, the price index for the meats, poultry, fish, and eggs food group was 0.1% higher, as the price index for poultry rose 0.4%, the price index for pork rose 0.4%, and egg prices were 2.9% higher….in addition, the seasonally adjusted price index for dairy products was 0.3% higher, even as average milk prices fell 1.2%, as the price index for cheese and related products rose 0.7% and the price index for other dairy products was 1.8% higher …on the other hand, the fruits and vegetables price index was 0.4% lower, as the price index for fresh vegetables fell 0.9%, the price index for fresh fruit fell 0.5%, and the price index for canned fruits was 0.3% lower.…meanwhile, the beverages price index was 0.2% higher, as the price index for carbonated drinks rose 1.2% and the price index for beverage materials other than coffee but including tea was 1.1% higher….lastly, the price index for the ‘other foods at home’ category was 0.1% higher, as the price index for salad dressing rose 5.0%, the price index for peanut butter rose 1.1%, the price index for snacks rose 1.1%, and the price index for prepared salads was 2.3% higher…

Among the seasonally adjusted core components of the CPI, which was 0.3% higher in August, after being 0.2% higher in July, unchanged in June, after rising by 0.2% in May, by 0.4% in April, 0.2% in March, by 0.2% in February, by 0.3% in January, by 0.2% in December, by 0.2% over the 2 months ending in November, and by 0.3% last August, the composite price index of all goods less food and energy goods was 0.1% higher in August, while the more heavily weighted composite index for all services less energy services was 0.3% higher..

Among the goods components of the core price index, which will initially be used by the Bureau of Economic Analysis to adjust August’s retail sales for inflation in national accounts data, the price index for household furnishings and supplies was unchanged, as  the price index for appliances rose 1.3%, the price index for floor coverings rose 1.5%, and the price index for outdoor equipment and supplies rose 1.0%, while the price index for furniture fell 0.9%, the price index for indoor plants and flowers fell 1.5, and the price index for household cleaning products was 0.6% lower….at the same time, the apparel price index was also unchanged, as a 2.1% increase in the price index for men’s suits, sport coats, and outerwear and a 4.9% increase in the price index for women’s dresses were offset by a 1.7% decrease in the price index for boys’ and girls’ footwear and a 1.6% decrease in the price index for jewelry and watches…. however, the price index for transportation commodities other than fuel was was 0.3% higher, as average prices for new vehicles rose 0.3%, average prices for used cars and trucks were 0.4% higher, and the price index for motor oil, coolant, and fluids was 2.6% higher….on the other hand, the price index for medical care commodities was 0.2% lower, as the price index for nonprescription drugs fell 1.0%, the price index for prescription drugs was unchanged, and the price index for medical equipment and supplies was 1.0% higher…meanwhile , the recreational commodities index was unchanged, as the price index for televisions rose 1.4%, the price index for sports vehicles including bicycles rose 1.8%, the price index for pets and pet products rose 0.7%, and the price index for recreational reading materials rose 2.9%, while the price index for video equipment other than televisions fell 6.1%, the price index for audio equipment fell 1.3%,and the price index for toys, games, hobbies and playground equipment was 3.3% lower… at the same time, the education and communication commodities index was 0.1% higher, as the price index for computers, peripherals, and smart home assistants rose 3.8%, but the price index for telephone hardware, calculators, and other consumer information items including smartphones was 2.4% lower.…lastly, a separate price index just for alcoholic beverages was 0.1% higher, while the price index for ‘other goods’ was 0.5% higher on a 0.8% increase in the price index for cigarettes, a 0.2% increase in the price index for cosmetics, perfume, bath, nail preparations and implements, and a 1.3% increase in the price index for miscellaneous personal goods…

Within core services, the price index for shelter was 0.3% higher, as rents rose 0.3%, and homeowner’s equivalent rent rose 0.2%, and prices for lodging away from home at hotels and motels were 2.7% higher, while the price index for tenants and household insurance was unchanged, and the price index for water, sewer and trash collection services was 0.5% higher but the price index for moving, storage, freight expense was 1.9% lower … however, the price index for medical care services was 0.2% lower, as price index for dental’ services fell 0.6%, and  the price index for health insurance was 0.5% lower…on the other hand, the transportation services price index was 0.5% higher, as the price index for airline fares rose 2.7%, the price index for car and truck rental rose 2.2% and the price index for motor vehicle maintenance and repair rose 1.1%, while the price index for motor vehicle insurance was 0.8% lower….at the same time, the recreation services price index was unchanged, as the price index for purchase, subscription, and rental of video rose 1.0%, the price index for veterinarian services rose 1.1% but the price index for admission to movies, theaters, and concerts was 0.6% lower…meanwhile, the price index for education and communication services was 1.8% higher, as the price index for elementary and high school tuition and fees rose 1.4% and the price index for wireless telephone services was 5.9% higher.…lastly, the price index for other personal services was 0.5% lower, as the price index for checking account and other bank services fell 2.0%,  the price index for laundry and dry cleaning rose 0.5%, and the price index for tax return preparation and other accounting fees was 0.4% lower…

Producer Price Index Rose 0.4% in August on Higher Energy & Transportation Prices

The seasonally adjusted Producer Price Index (PPI) for final demand was 0.4% higher in August, as the final demand price index for wholesale goods rose 1.1%, while the more heavily weighted price index for final demand for services was 0.1% higher…that 0.4% PPI  increase in August  followed a revised 0.1% increase in July, when the final demand price index for wholesale goods fell 0.4%, while the price index for final demand for services was 0.2% higher, and followed a revised 0.1% decrease in June, when the final demand price index for wholesale goods fell 1.4%, while the price index for final demand for services was 0.5% higher, and a revised 0.5% PPI increase in May, when the final demand price index for wholesale goods rose 2.2%, while price index for final demand for services was 0.2% lower, and followed a revised 1.1% increase in April, when the final demand price index for wholesale goods rose 1.9% and the price index for final demand for services was 0.8% higher, and an unrevised 0.8% increase in March, when the final demand price index for wholesale goods rose 2.0%, while the price index for final demand for services was 0.3% higher, and followed an unrevised 0.5% increase in February, when the final demand price index for wholesale goods was 1.0% higher, and the price index for final demand for services was 0.3% higher, and an unrevised 0.6% increase in January, when the final demand price index for wholesale goods was 0.1% lower, but the price index for final demand for services was 0.9% higher, and an unrevised 0.4% PPI increase in December, when the final demand price index for wholesale goods was 0.1% lower, while the price index for final demand for services was 0.6% higher, and a  0.4% PPI increase in November, when the final demand price index for wholesale goods was 0.8% higher, and the price index for final demand for services was 0.3% higher, and followed the report of a 0.1% PPI increase in October, when the final demand price index for wholesale goods fell 0.2%, but the more heavily weighted price index for final demand for services was 0.2% higher, and also followed a 0.6% PPI increase last September, when the final demand price index for wholesale goods rose 0.6% and the price index for final demand for services was also 0.6% higher….those post-shutdown reports followed a revised 0.2% decrease last August, when the final demand price index for wholesale goods rose 0.2%, but the more heavily weighted price index for final demand for services was 0.3% lower….on an unadjusted basis, producer prices are 5.4% higher than a year ago, while the core producer price index, which excludes food, energy and trade services, was 0.3% higher for the month. and is still 4.7% higher than it was a year ago…

As noted, the producer price index for final demand for goods was 1.1% higher in August, after being 0.4% lower in July, 1.4% lower in June, 2.2% higher in May, 1.9% higher in April, 2.0% higher in March, 1.0% higher in February, 0.1% lower in January, 0.1% lower in December, 0.8% higher in November, 0.1% lower in October, 0.6% higher last September, and 0.2% lower last August, and is now 7.7% higher than a year ago….the final demand goods price index was 1.1% higher in August because the price index for wholesale energy goods was 4.2% higher, after energy prices had been 1.8% lower in July, 6.8% lower in June, 8.2% higher in May, 7.1% higher in April, 10.5% higher in March, 2.0% higher in February, and 1.7% lower in January,  and as the price index for wholesale foods was 0.1% higher, after wholesale foods had been had been 0.9% lower in July, 0.6% lower in June, 0.6% higher in May, 0.3% higher in April, 0.6% lower in March, 2.3% higher in February, and 1.1% lower in January, and as the price index for final demand for core wholesale goods (excluding food and energy) was 0.4% higher in August, after it had been 0.2% higher in July, 0.2% higher in June, 0.7% higher in May, 0.7% higher in April, 0.4% higher in March, 0.4% higher in February. and 0.7% higher in January….

Wholesale energy prices were 4.2% higher in August on a 24.4% increase in wholesale prices for No. 2 diesel fuel, a 22.8% increase in wholesale prices for home heating oil and distillates, a 4.2% increase in wholesale prices for gasoline, and a 8.8% increase in wholesale prices for natural gas liquids, while the final demand for food price index was 0.1% higher on a 32.3% increase the wholesale price index for eggs for fresh use, a 4.3% increase in the wholesale price index for grains, and a 6.1% increase the wholesale price index for confectionery end products… among core wholesale goods, the wholesale price index for industrial chemicals rose 1.3%, the wholesale price index for oil field and gas field machinery rose 1.6%, the wholesale price index for electronic components and accessories rose 3.4%, and the wholesale price index for cigarettes was 2.5% higher…

Meanwhile, the price index for final demand for services was 0.1% higher in August, after being 0.2% higher in July, 0.5% higher in June, 0.2% lower in May, 0.8% higher in April, 0.3% higher in March, 0.3% higher in February, 0.9% higher in January, 0.6% higher in December, 0.2% higher in November, 0.2% higher in October, 0.6% higher last September, and 0.3% lower last August, and is now 4.5% higher than a year ago….the price index for final demand for trade services slipped 0.2% in July, but the price index for final demand for transportation and warehousing services rose 2.3%, while the core index for final demand for services other than trade, transportation, and warehousing services was unchanged….

Among trade services, seasonally adjusted margins for health, beauty, and optical goods retailers fell 3.6%, margins for fuels and lubricants retailers fell 11.3%, and margins for apparel wholesalers were 2.5% lower, while margins for automobile retailers rose 4.2%  and margins for hardware, building materials, and supplies retailers were 5.3% higher….among transportation and warehousing services, average margins for truck transportation of freight rose 2.0%, margins for air transportation of freight rose 1.5%, and margins for airline passenger services were 4.2% higher….among the components of the core final demand for services index, the price index for cable and satellite subscriber services rose 2.9%, the price index for credit intermediation, including trust services not otherwise listed rose 4.5%, and the price index for arrangement of cruises and tours rose 2.0%, while the price index for application software publishing fell 3.3%, the price index for portfolio management fell 1.6%, and the price index for management, scientific, and technical consulting services was 4.6% lower…

This report also showed the price index for intermediate processed goods was 1.8% higher in August, after being 0.4% lower in July, 1.0% lower in June, 2.7% higher in May, 2.6% higher in April, 3.0% higher in March, 1.5% higher in February, 0.1% higher in January, 0.1% higher in December, 0.6% higher in November, 0.1% lower in October, 0.2% higher last September, and 0.4% higher last August….the price index for intermediate energy goods rose 7.3% in August as refinery prices for No. 2 diesel fuel rose 24.1%, refinery prices for jet fuel rose 21.3%, refinery prices for gasoline rose 4.2%, and producer prices for natural gas liquids were 8.8% higher….on the other hand, the price index for intermediate processed foods and feeds fell 0.1%, as the producer price index for processed poultry fell 1.9%, the producer price index for meats fell 1.0%, and the producer price index for dairy products was 1.0% lower…meanwhile, the core price index for intermediate processed goods less food and energy goods was 0.5% higher, as the producer price index for basic organic chemicals rose 1.7%, the producer price index for paperboard rose 2.6%, the producer price index for copper and brass mill shapes rose 3.5%, the producer price index for nonferrous wire and cable rose 2.4%, the producer price index for steel mill products rose 1.7%, and the producer price index for switchgear, switchboard, and industrial controls equipment was 2.4% higher….average prices for intermediate processed goods are now 11.5% higher than in August 2025, the 22nd year over year increase in 42 months, but are way off the 26.6% year over year increase of November 2021, which had been a 46 year high…

Meanwhile, the price index for intermediate unprocessed goods rose 1.1% in August, after rising 0.7% in July, falling 6.5% % in June, rising 3.0% in May and 1.9% in April, falling 0.2% in March, rising 5.9% in February, rising 4.3% in January, and 2.1% in December and 2.4% in November, after falling 1.3% in October, falling 0.5% last September, and falling 1.8% last August….that was as the August price index for crude energy goods rose rose 1.5%, as crude oil prices rose 5.2%, unprocessed natural gas prices fell 11.6%, and coal prices were 1.0% lower… on the other hand, the price index for unprocessed foodstuffs and feedstuffs was 0.1% lower, as the producer price index for slaughter cattle fell 6.4%, the producer price index for raw milk fell  8.7%, and the producer price index for raw cane sugar and byproducts was 0.9% lower….meanwhile, the index for core raw materials other than food and energy materials was 2.1% higher, on a 5.4% increase in the price index for raw cotton, a 4.2% increase in the price index for copper base scrap, a 3.3% increase in the price index for aluminum base scrap,  and a 4.7% increase in the price index for recyclable paper….this raw materials price index is now 12.8% higher than a year ago, the 18th year over year increase in the past 43 months, which followed a run of twenty-seven consecutive year over year increases, which came after the annual change on this index had been negative from the beginning of 2019 through October of 2020…

Lastly, the price index for services for intermediate demand was 0.3% higher in August, after being 0.1% higher in July, 0.8% higher in June, 0.4% higher in May, 0.9% higher in April, 0.1% higher in March, 0.3% higher in February, 0.6% higher in January, 0.6% higher in December, 0.2% higher in November, 0.4% higher last October, 0.3% higher last September, and unchanged last August.…the price index for intermediate trade services was 1.0% higher, as margins for building materials, paint, and hardware wholesalers rose 1.6%, margins for intermediate hardware, building material, and supplies retailers, rose 5.3%, and margins for intermediate paper and plastics products wholesalers rose 1.3%….at the same time, the price index for transportation and warehousing services for intermediate demand was 0.5% higher, as the intermediate price index for truck transportation of freight rose 2.0%, the intermediate price index for air transportation of freight rose 1.5%, the intermediate index for air mail and package delivery services, excluding by USPS, rose 4.8%, and the intermediate index for transportation of passengers was 4.1% higher….on the other hand, the core price index for intermediate services other than trade, transportation, and warehousing services was 0.1% lower, as the intermediate price index for executive search services fell 5.7%, the intermediate price index for portfolio management fell 1.6%, the intermediate price index for television advertising time sales fell 2.4%, and the intermediate price index for management, scientific, and technical consulting services was 4.8% lower….over the 12 months ended in July, the price index for services for intermediate demand was 5.1% higher than it was a year earlier, the seventieth  consecutive annual increase in this index, after it had briefly turned negative year over year at the onset of the pandemic, from April to August of 2020, even as the current annual increase is still lower than the record 9.5% year over year increase that was indicated for July 2021…

Wholesale Sales Rose 0.8% in July; Wholesale Inventories Were Up 1.3%

The July report on Wholesale Trade, Sales and Inventories (pdf) from the Census Bureau estimated that the seasonally adjusted value of wholesale sales was at $801.3 billion in July, up 0.8 percent (±0.4 percent) from the revised June level, and was up 13.0 percent (±0.5 percent) from wholesale sales of July 2025… the June preliminary sales estimate was revised to $794.9 billion from the $794.1 billion in sales reported last month, and hence the May to June change was revised to a 2.9 percent (±0.5 percent) decrease from the 3.0% percent (±0.5 percent) decrease that was reported last month.…as an intermediate activity, wholesale sales are not included in GDP except insofar as they are a trade service, since the traded goods themselves do not represent an increase in the output of the goods produced or finally sold….

On the other hand, the monthly change in private inventories is a major factor in GDP, as additional goods on the shelf or in intermediate storage represent goods that were produced but not sold, and this July report estimated that wholesale inventories were valued at a seasonally adjusted $958.9 billion at month end, up 1.3 percent (+/-0.2%)* from the revised June level. and 5.7 percent (±1.2 percent) higher than in June a year ago….at the same time, the June preliminary estimate revised from the $944.7 billion reported a month ago to $946.7 billion, which revised the June change from up 0.2% to up 0.4% from May….

For national accounts purposes, wholesale inventories are adjusted for price changes by category with the appropriate component of the producer price index.   The producer price index for July indicated that prices for finished goods were on average 0.4% lower, that prices for intermediate processed goods were also on average 0.4% lower, but that prices for unprocessed goods were 1.1% higher….hence, once adjusting for those changes in price, July’s real wholesale inventories were around 1.5% more than those at the end of the 2nd quarter, and maybe even more than that…since the key source data and assumptions(xls) for the second estimate of 2nd quarter GDP indicated that real wholesale inventories were down sharply in the 2nd quarter, accounting for over 35% the quarter’s inventory decrease, that this report appears to indicate a modest real increase in aggregate July’s wholesale inventories in comparison would therefore mean the change in inventories has increased, and hence the difference between the 2nd quarter decrease and the July increase would be added to the 3rd quarter’s real growth in GDP…

Existing Home Sales 2.0% Lower in August Despite Lower Prices

   The National Association of Realtors (NAR) reported that their seasonally adjusted count of existing home sales fell 2.0% from July to August, the 3rd consecutive decrease, projecting that 3.98 million homes would sell over an entire year if the August home sales pace were extrapolated over that year, a pace that was 1.2% below the 4.03 million annual sales rate they projected in August of a year ago….July’s home sales, at a 4.06 million annual rate, were revised  but unchanged  from the 4.06 million annual rate shown in last month’s report….the NAR also reported that the median sales price for all existing-home types was $429,100 in August, 1.6% higher than in August a year earlier, which they cite as “the 38th consecutive month of year-over-year price increases“…the NAR press release, which is titled “NAR Existing-Home Sales Report Shows 2.0% Decrease in August“, is in easy to read plain English, so if you’re interested in the details on housing inventories, which were the highest since November 2019, cash sales, distressed sales, first time home buyers, etc., you can easily find them in that press release…as sales of existing properties do not add to our national output, neither these home sales nor the prices for which these homes sell are included in GDP, except insofar as real estate, local government and banking services are rendered during the selling process…

Since this report is entirely seasonally adjusted and at a not very informative annual rate, we like to check the raw data overview (pdf), which gives us a close approximation of the actual number of homes that sold each month, and their prices….this unadjusted data indicates that roughly 363,000 homes sold in August, down by 9.5% from the 401,000 homes that sold in July, and down by 3.5% from the 376,000 homes that sold in August of last year…that same pdf indicates that the median home selling price for all housing types fell from a revised $436,400 in July and from a revised $442,800 in June to $429,100 in August, with the regional median home sales prices ranging from a low of $340,400 in the Midwest to a high of $619,100 in the West…

(the above is the synopsis that accompanied my regular Sunday morning news links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are chosen from the aforementioned GGO posts, contact me…)

record high US oil output, oil supplies at 42 year low; SPR at 43½ year low; WTI at 16 week high; record diesel prices

US oil prices settled over $100 for the first time in 16 weeks; diesel prices at an all time high; production of crude from US wells at a record high; total US oil supplies lowest since March 1984 with the Strategic Petroleum Reserve the lowest since November 1982; oil refined was the most one week in seven years, refinery utilization rate was 2nd highest in eight years..

US oil prices rose for a fifth time in six weeks after the US “destroyed” at least eight Iranian oil tankers in the Persian Gulf and Gulf of Oman, Houthi attacks on Saudi oil facilities forced them to shut down, and Iran fired advanced missiles at U.S. warships in a new escalation of the war…after rising 9.7% to $91.48 a barrel last week after the US launched a new wave of attacks on Iranian infrastructure and shipping and Iran retaliated against US bases and allies in the region and their shipping, the contract price for the benchmark US light sweet crude for October delivery edged higher in early Asian trading on Monday after U.S. forces struck three Iranian crude carriers following Iranian missile attacks aimed at two U.S. warships, and traded at a six-week high, while ​Brent, the international benchmark​, settled 1.1% higher after Iran vowed to strike energy infrastructure across the Middle East in response to further U.S. attacks on its assets, further escalating a conflict that had sharply reduced oil supply from the region…oil prices surged nearly 3% during Asian trading hours on Tuesday after Yemen’s Houthis attacked energy facilities in Saudi Arabia, while Iran threatened the US with what it called “economic warfare,” increasing fears that the war would disrupt more oil supplies from the region, and touched a 13-week high in early trading in New York after fresh Houthi attacks on Saudi energy facilities forced their operations to shutdown​ , and settled $1.55 higher at $93.03 a barrel after the Houthi attacks on Saudi energy facilities set oil installations ablaze and threatened  a ‌major expansion of the Middle East war…oil prices climbed more than two percent across Asia on Wednesday, as escalating US-Iran tensions continued to roil global energy markets, then extended their rally Wednesday morning in New York, as the recent escalations in the U.S.-Iran war stoked further supply fears, and settled $3.02 higher at $96.05 a barrel after U.S. forces struck five Iranian oil tankers and Iran targeted a U.S. base in Jordan and attacked 10 ships…oil prices extended their gains in early global trading on Thursday after Iran said it had attacked 10 ships near the Strait of Hormuz, putting the spotlight back on the Strait and its role in global energy supplies, but traded slightly lower in London on Thursday morning on profit-taking after the strong ​price rise the previous day, then rallied to top $100 for the first time since May in early New York trading after the EIA reported the US Strategic Petroleum Reserve​ was near a record low and Cushing stocks ​were just off ‘tank bottoms’ and contined to ralliy to settle $6.43 higher at $102.48 a barrel for an eighth straight gain, the longest streak higher in more than three years, following the news that the Iran-aligned Houthis ​had seized control of Yemen’s port of Mocha on the Bab el-Mandab Strait, further increasing their threat to oil flows through the Red Sea….oil prices extended their climb during early Asian trading on Friday, reaching their highest since May​, as hostilities in the Gulf continued and the prospect of peace ​s​eemed even more remote, but retreated during morning trading in New York as reports that Iran was in negotiations with Gulf states to establish a jointly managed shipping route through the Strait of Hormuz weighed on prices, and settled $2.43 lower at $100.05 a barrel, reversing early gains after the Financial Times reported that foreign ministers in the Middle East were trying to work out a temporary deal with Iran to manage shipping through the Strait of Hormuz, but still finished 9.4% higher for the week, settling above $100 for the first time in over four months…

meanwhile, natural gas prices finished lower for the first time in five weeks as traders shifted their focus to the Autumn “shoulder season”, when little demand for either cooling or heating is expected….after rising 3% to $2.975 per mmBTU last week on lower production, higher LNG demand, and forecasts for record cooling demand in September, the price of the benchmark natural gas contract for October delivery opened 2 cents lower on Monday and gradually arced lower ​t​hrough the morning, as traders focus​ed on the impending shoulder season and fading cooling demand, and settled 5.9 cents lower at $2.916 per mmBTU​, as electric power demand dropped sharply and fading cooling demand outweighed strong LNG feedgas needs….natural gas prices opened 6.3 cents lower on Wednesday, tumbling overnight as shoulder season conditions took hold, and settled 9.4 cents lower at $2.822 per mmBTU as fading power demand and the approach of milder fall temperatures pressured the market, despite strong LNG feedgas ​demand  and expectations for a relatively lean storage ​r​eport Thursday…natural gas prices started Thursday 4.8 cents lower, moving down overnight as traders looked to easing demand as the season moves deeper into September, then dropped to an intraday low of $2.753 as a bearish storage report hit the wire, before posting a steady recovery in the following hours to settle 1.2 cents higher at $2.834 per mmBTU, supported by lower production and strong LNG feedgas demand, despite fading weather-driven demand and a larger-than-expected storage build…natural gas futures showed little sign of recovery on Friday morning, as traders assessed a bearish storage surprise and seasonal declines in power sector demand, and continued to slump through morning trading, weighed down by seasonally strong production readings and forecasts for waning weather demand, and settled 0.3 cents lower at $2.831 per mmBTU, and thus ended 4.8% lower for the week as rapidly fading power sector demand and the approaching shoulder season outweighed strong LNG feedgas and lower production.

The EIA’s natural gas storage report for the week ending September 4th indicated that the amount of working natural gas held in underground storage rose by 40 billion cubic feet to 3,254 billion cubic feet by the end of the week, which left our natural gas supplies 76 billion cubic feet, or 2.4% below the 3,333 billion cubic feet of gas that were in storage on September 4th of last year, but 148 billion cubic feet, or 5.2% above the five-year average of 3,106 billion cubic feet of natural gas that had typically been in working storage as of the 4th  of September over the most recent five years….the 40 billion cubic foot injection into natural gas storage for the cited week was more than the 35 billion cubic foot injection into storage that the market had been expecting ahead of the report, but it was less than the 69 billion cubic foot of gas that were injected into natural gas storage during the corresponding week of 2025, and also less than the average 52 billion cubic foot injection into natural gas storage that had been typical for the first week in September over the past five years…

The Latest US Oil Supply and Disposition Data from the EIA

US oil data from the US Energy Information Administration for the week ending September 4th showed that despite a big drop in our oil exports, we had we had to pull oil out of our stored crude supplies for the nineteenth time in twenty weeks, and for the 41st time in sixty-seven weeks, in part due to another increase in demand for oil the EIA could not account for…. Our imports of crude oil rose by an average of 53,000 barrels per day to 6,824,000 barrels per day, after rising by an average of 612,000 barrels per day during the prior week, while our exports of crude oil fell by an average of 1,066,000 barrels per day to average 3,417,000 barrels per day, which, when used to offset our imports, meant that the net of our trade of oil worked out to an import average of 3,407,000 barrels of oil per day during the week ending September 4th, an average of 1,119,000 more barrels per day than the net of our imports minus our exports during the prior week… At the same time, transfers to our oil supplies from Alaskan gas liquids, from natural gasoline, from condensate, and from unfinished oils were unchanged from the prior week at 887,000 barrels per day, while during the same week, production of crude from US wells was 85,000 barrels per day higher at a record high of 13,947,000 barrels per day.  Hence, our daily supply of oil from the net of our international trade in oil, from transfers, and from domestic well production appears to have averaged a total of 18,241,000 barrels per day during the September 4th reporting week…

Meanwhile, US oil refineries reported they were processing an average of 17,586,000 barrels of crude per day during the week ending September 4th, an average of 91,000 more barrels per day than the amount of oil that our refineries reported they were processing during the prior week, and the most we’ve refined in one week in seven years, while over the same period, the EIA’s surveys indicated that a total of 233,000 barrels of oil per day were being pulled from the supplies of oil stored in the US… So, based on all that reported & estimated data, the crude oil figures provided by the EIA appear to indicate that our total working supply of oil from storage, from net imports, from transfers, and from oilfield production during the week ending September 4th averaged a rounded 888,000 more barrels per day than what our oil refineries reported they used during the week.  To account for the difference between the apparent supply of oil and the apparent disposition of it, the EIA just plugged a [ -888,000 ] barrel per day figure onto line 16 of the weekly U.S. Petroleum Balance Sheet, in order to make the reported data for the supply of oil and for the consumption of it balance out, a fudge factor that they label in their footnotes as “unaccounted for crude oil”, thus indicating there must have been a error or omission of that amount in the week’s oil supply & demand figures that we have just transcribed…. Since 623,000 barrels per day of demand for could not be accounted for in the prior week’s EIA data, that means there was a 265,000 barrel per day difference between this week’s oil balance sheet error and the EIA’s crude oil balance sheet error from a week ago, and hence the changes to supply and demand from that week to this one that are indicated by this week’s report are somehow off by that much, and therefore not very useful… However, since most oil traders react to the figures in these weekly EIA reports as if they were gospel, and since these weekly figures therefore often drive oil pricing and hence decisions to drill or complete oil wells, we’ll continue to report this data just as it’s published, and just as it’s watched & believed to be reasonably reliable by most everyone in the industry…(for more on how this weekly oil data is gathered, and the possible reasons for that “unaccounted for” oil supply, see this EIA explainer….also see this March 2023 twitter thread from an EIA administrator addressing these ongoing weekly errors, and what they had once hoped to do about it).

This week’s rounded 233,000 barrel per day average decrease in our overall crude oil inventories came as an average of 56,000 barrels per day were being pulled out of our commercial stocks of crude oil, while 178,000 barrels per day were being pulled out of our Strategic Petroleum Reserve, the twenty-fourth consecutive Iran war related withdrawal from the SPR, including the four largest draws in SPR history, which left the SPR level at 285,360,000 barrels, the lowest since it was initially being filled in November 1982….with both commercial oil and the SPR both down again this week, that left the Total of all US Oil Supplies at 709,429,000 barrels, down 18.5% from 870,774,000 barrels on April 17th, and the lowest since March 30th, 1984….

Further details from the weekly Petroleum Status Report (pdf) indicated that the 4 week average of our oil imports fell to 6,586,000 barrels per day last week, which was still 2.3% more than the 6,436,000 barrel per day average that we were importing over the same four-week period last year, while the four week average of our exports rose to 3,940,000 barrels per day last week, which was 6.4% more than the 3,703,000 barrel per day average that we were exporting last year year at this time… This week’s crude oil production was reported to be 85,000 barrels per day higher at all time high of  13,947,000 barrels per day as the EIA’s estimate of the output from wells in the lower 48 states was 71,000 barrels per day higher at 13,490,000 barrels per day, while Alaska’s oil production was 14,000 barrels per day higher at 457,000 barrels per day…US crude oil production had reached a pre-pandemic high of 13,100,000 barrels per day during the week ending March 13th 2020, so this week’s reported oil production figure was 6.5% higher than that of our pre-pandemic production peak, and was also 43.8% above the pandemic low of 9,700,000 barrels per day that US oil production had fallen to during the third week of February of 2021.

US oil refineries were operating at 97.8% of their capacity while processing those 17,586,000 barrels of crude per day during the week ending September 4th, down from 98.0% the prior week, but still the second highest refinery utilization rate in eight years ….the 17,586,000 barrels of oil per day that were refined that week were the most we’ve refined in one week since August 16th, 2019, 4.6% more than the 16,818,000 barrels of crude that were being processed daily during the week ending September 5th of 2025, and 0.5% more than the 17,495,000 barrels that were being refined during the pre-pandemic week ending September 6th, 2019, when our refinery utilization rate was at 96.4%, which was close to the pre-pandemic normal utilization rate for this time of year…

Even with the increase in the amount of oil that was being refined this week, gasoline output from our refineries was somewhat lower, decreasing by 587,000 barrels per day to 9,308,000 barrels per day during the week ending September 4th, after our refineries’ gasoline output had increased by 73,000 barrels per day during the prior week… This week’s gasoline production was 2.9% lower than the 9,587,000 barrels of gasoline that were being produced daily over the week ending September 5th of last year, and 10.2% less than the gasoline production of 10,360,000 barrels per day seen during the prepandemic week ending September 6th, 2019….on the other hand, our refineries’ production of distillate fuels (diesel fuel and heat oil) increased by 222,000 barrels per day to 5,348,000 barrels per day, after our distillates output had decreased by 9,000 barrels per day during the prior week.  After that big production increase, our distillates output was 2.3% more than the 5,229,000 barrels of distillates that were being produced daily during the week ending September 5th of 2025, and 0.1% more than the 5,341,000 barrels of distillates that were being produced daily during the pre-pandemic week ending September 6th, 2019….

Even with this week’s decrease in our gasoline production, our supplies of gasoline in storage at the end of the week rose for the sixth time in thirty weeks, increasing by 1,269,000 barrels to 206,938,000 barrels during the week ending September 4th, after our gasoline inventories had decreased by 1,173,000 barrels to a 42 week low during the prior week.  Our gasoline supplies rose this week because the amount of gasoline supplied to US users fell by 371,000 barrels per day to 8,551,000 barrels per day, and because our imports of gasoline rose by 94,000 barrels per day to 464,000 barrels per day, and because our exports of gasoline fell by 62,000 barrels per day to 872,000 barrels per day… After fifty-four gasoline inventory withdrawals over the past eighty-one weeks, our gasoline supplies were 5.9% lower than last September 5th’s gasoline inventories of 219,997,000 barrels, and about 5% below the five year average of our gasoline supplies for this time of year…

After this week’s big increase in distillates production, our supplies of distillates rose for the seventeenth time in thirty weeks, increasing by 2,087,000 barrels to 106,274,000 barrels during the week ending September 4th, after our distillates supplies had increased by 796,000 barrels during the prior week… Our distillates supplies rose by more this week even though the amount of distillates supplied to US markets, an indicator of domestic demand, rose by 288,000 barrels per day to 3,678,000 barrels per day, because our exports of distillates fell by 179,000 barrels per day to 1,556,000 barrels per day, and because our imports of distillates rose by 72,000 barrels per day to 185,000 barrels per day… After 28 withdrawals from distillates inventories over the past 60 weeks, our distillates supplies at the end of the week were 11.9% lower than the 20,638,000 barrels of distillates that we had in storage on September 5th of 2025, and were about 13% below the five year average of our distillates inventories for this time of the year…

Finally, even after the big drop in our oil exports, our commercial supplies of crude oil in storage fell for the 14th time in twenty-six weeks, and for the 26th time over the past year, decreasing by 391,000 barrels over the week, from 424,460,000 barrels on August 28th to 424,069,000 barrels on September 4th, after our commercial crude supplies had decreased by 4,450,000 barrels over the prior week….After this week’s decrease, our commercial crude oil inventories were back to the recent five-year average of commercial oil supplies for this time of year, while they were still about 28% above the average of our available crude oil stocks as of the first weekend of September over the 5 years at the beginning of the past decade, with the difference between those comparisons arising because it wasn’t until early 2015 that our oil inventories had first topped 400 million barrels. After our commercial crude oil inventories had jumped to record highs during the Covid lockdowns in the Spring of 2020, then jumped again after February 2021’s winter storm Uri froze off US Gulf Coast refining, but then fell sharply due to increased exports to Europe following the onset of the Ukraine war, only to jump again following the Christmas 2022 refinery freeze-offs, changes in our commercial crude inventories had been less extreme up until the onset of the Iran war, when they were initially built up to a three year high by mid-April, before falling to the lowest in nearly eight years by late July…This week’s decrease was the second in six weeks, and as of September 4th our commercial crude inventories were 0.1% below the 424,646,000 barrels of oil we had in commercial storage on September 5th of 2025, but were 1.2% more than the 419,143,000 barrels of oil that we had in storage on September 6th of 2024, and 0.8% more than the 420,592,000 barrels of oil we had left in commercial storage on September 8th of 2023…

This Week’s Rig Count

The US rig count increased by three over the week ending September 11th, as the number of rigs targeting oil was up by one, the count of rigs targeting natural gas was up by two, and miscellaneous rigs were unchanged…for a quick snapshot of this week’s rig count, we are again including below a screenshot of the rig count summary table from Baker Hughes…in the table below, the first column shows the active rig count as of September 11th, the second column shows the change in the number of working rigs between last week’s count (September 4th) and this week’s (September 11th) count, the third column shows last week’s September 4th active rig count, the 4th column shows the change between the number of rigs running on Friday and the number running on the Friday of the same week of a year ago, and the 5th column shows the number of rigs that were drilling at the end of that reporting period a year ago, which in this week’s case was Friday, the 12th of September, 2025…

++

August jobs report; July’s trade deficit, construction spending, factory inventories, and JOLTS, et al

Major reports released last week included the Employment Situation Summary for August and the Job Openings and Labor Turnover Survey (JOLTS) report for July, both from the Bureau of Labor Statistics, and three July reports that provide us with preliminary input data to 3rd quarter GDP, and also suggest revisions to 2nd quarter GDP: the July report on our International Trade from the Bureau of Economic Analysis, and the July report on Construction Spending (pdf), and the Full Report on Manufacturers’ Shipments, Inventories and Orders for July, both from the Census Bureau…

The week’s major privately issued reports included the ADP Employment Report for August, wherein the national payroll processor reported a 38,000 increase in private jobs in August, the light vehicle sales report for August from Omida, aka Wards Automotive, which estimated that vehicles sold at a 16.8 million annual rate in August, up from the 16.3 million annual sales rate reported in July, and up from the 16.5 million annual rate reported for August a year ago, and both of the widely followed purchasing manager’s surveys from the Institute for Supply Management (ISM): the August Manufacturing Report On Business indicated that the manufacturing PMI (Purchasing Managers Index) fell to 54.6% in August, down from 55.6% in July, which means that a smaller plurality of manufacturing industry purchasing managers reported improving metrics in various facets of their business in August than in July, and the August Services Report On Business; which saw their Services PMI rise to 55.4% in August, up from 54.1% in July, indicating a larger plurality of service industry purchasing managers reported improvement in various facets of their business in August than in July…

Employers Added 162,000 Jobs in August; Unemployment Rate Unchanged

The Employment Situation Summary for August indicated a payroll jobs increase that was above expectations and the highest initial report since March, an unemployment rate that was unchanged, and an increase in both the employment rate and the labor force participation rate….estimates extrapolated from the establishment survey data indicated that employers added a seasonally adjusted 162,000 jobs in August, after the payroll job change for July was revised from a decrease of 23,000 jobs to an increase of 21,000 jobs, while the payroll jobs change for June was revised from an increase of  20,000 jobs to an increase of 31,000 jobs, revisions which combined means that this report indicates an increase of 217,000 more jobs than were reported last month, in contrast to the 50,000 job addition that was expected …the unadjusted data shows that there were actually 154,000 more payroll jobs in August, as downward adjustments to sectors that normally see an increase in August, such as education, were mostly offset by upward adjustments to sectors that normally see a job decrease, such as retail sales and leisure and hospitality….

Seasonally adjusted job increases were spread through the goods producing, service providing, and government sectors, with a loss of 23,000 jobs in the information sector the only notable decrease…since the BLS summary of the job gains by sector is clear and usually more detailed than our usual synopsis, we’ll just quote from that summary here:

• Total nonfarm payroll employment rose by 162,000 in August, higher than the average monthly gain of 31,000 over the prior 12 months. In August, employment increased in food services and drinking places and in local government education. The information industry lost jobs. (See table B-1.)
• Employment in food services and drinking places increased by 59,000 in August, well above the average monthly gain of 12,000 over the prior 12 months.
• Local government education added 42,000 jobs in August, largely offsetting a decrease in the prior month. Local government education has shown little net change since January 2025.
• In August, employment in manufacturing continued its upward trend (+16,000) and is up by 58,000 since a recent low in December 2025. Employment in machinery manufacturing (+6,000) and in fabricated metal product manufacturing (+6,000) continued trending up in August.
• Employment in health care continued to trend up in August (+13,000) but at a slower pace than the average monthly gain over the prior 12 months (+32,000). Over the month, home health care services (+11,000) and hospitals (+8,000) added jobs.
• Information employment declined by 23,000 in August, following losses that had averaged 8,000 per month over the prior 12 months. In August, job losses occurred in computing infrastructure providers, data processing, web hosting, and related services (-8,000), in publishing industries (-7,000), and in broadcasting and content providers (-5,000).
• Construction employment changed little in August (+22,000). Employment in nonresidential specialty trade contractors continued to trend up (+8,000), similar to the average monthly gain over the prior 12 months (+6,000).
• Employment showed little change over the month in other major industries, including mining, quarrying, and oil and gas extraction; wholesale trade; retail trade; transportation and warehousing; financial activities; professional and business services; social assistance; and other services.

The establishment survey also showed that average hourly pay for all employees rose by 10 cents an hour to $37.75 an hour, after it had increased by a revised 5 cents an hour in July; at the same time, the average hourly earnings of production and non-supervisory employees increased by 11 cents to $32.53 an hour….employers also reported that the average workweek for all private payroll employees rose by a tenth of an hour to 34.4 hours, while average weekly hours for production and non-supervisory personnel were unchanged at 33.8 hours…meanwhile, the manufacturing workweek increased by 0.1 hour to 40.5 hours, while average factory overtime was unchanged at 3.1 hours…

At the same time, the seasonally adjusted extrapolation from the August household survey indicated that the number of those who would self-report being employed rose by an estimated 569,000 to 162,746,000, while the similarly estimated number of those who would be counted as unemployed rose by 115,000 to 7,031,000; which together meant that August saw a net rounded increase of 683,000 to 169,777,000 in the total labor force…since the working age population had grown by 133,000 over the same period, that meant the number of employment aged individuals who were not in the labor force fell by a rounded 551,000 to 105,638,000….the increase in the total labor force was enough to raise the labor force participation rate, from 61.4% in July to 61.6% in August, even as it was still down from 62.3% in August of last year….at the same time, the increase in number employed vis-a-vis the population was great enough to increase the employment to population ratio, which we could think of as an employment rate, from 58.9% in July to
59.1% in August, which was still down from 59.6% a year ago…however, the increase in the number counted as unemployed was not enough to raise the unemployment rate as a percentage of the labor force, as it remained at 4.1%….meanwhile, the number who reported they were involuntarily working part time fell by 414,000 to 4,390,000 in August, which meant that the alternative measure of unemployment, U-6, which includes those “employed part time for economic reasons”, fell from 7.9% in July to 7.7% in August, the lowest U6 unemployment rate since last June….

Like most reports from the Bureau of Labor Statistics, the employment situation press release itself is easy to read and understand, so you can get more details on these two reports from there…note that almost every paragraph in that release points to one or more of the tables that are linked to on the bottom of the release, and those tables are also on a separate html page here that you can open it alongside the press release to avoid the need to scroll up and down the page to access the tables..

Job Openings Rose in July, Hiring, Layoffs, and Job Quitting All Fell

The Job Openings and Labor Turnover Survey (JOLTS) report for July from the Bureau of Labor Statistics estimated that seasonally adjusted job openings rose by 89,000, from 7,182,000 in June to 7,271,000 in July, after June’s job openings were revised 177,000 lower, from the 7,359,000 reported a month ago to 7,182,000 with this report…July’s jobs openings were still 2.6% higher than the 7,089,000 job openings reported for July a year ago, as the job opening ratio expressed as a percentage of the employed rose from 4.3% in June to 4.4% in July, and was also up from 4.3% in July a year ago….a 76,000 job opening increase to 429,000 job openings in durable goods manufacturing appears to be the largest percentage increase in openings this month, while a decrease from 383,000 to 316,000 job openings in transportation, warehousing, and utilities looks to be the largest percentage decrease… (see table 1 for more details)…like most BLS releases, the press release for report is easy to understand and also refers us to the associated table for the data cited, which are linked at the end of the release…

The JOLTS release also reports on labor turnover, which consists of hires and job separations, which in turn is further divided into layoffs and discharges, those who quit, and ‘other separations’, which includes retirements and deaths….in July, seasonally adjusted new hires totaled 5,054,000, down by 278,000 from the revised 5,332,000 who were hired or rehired in June, as the hiring rate as a percentage of all employed fell from 3.4% in June to 3.2% in July, while it was also down from the 3.3% hiring rate in July of a year earlier (details of hiring by sector since March are in table 2)….meanwhile, total separations fell by 265,000, from 5,337,000 in June to 5,072,000 in July, while the separations rate as a percentage of the employed also fell from 3.4% in June to 3.2% in July, while it was also down from the 3.3% separations rate in July a year ago (see table 3)…subtracting the 5,072,000 total separations from the total hires of 5,054,000 would imply a decrease of 18,000 jobs in July, as opposed to the revised payroll job increase of 21,000 for July reported by the August establishment survey later in the week, but still with the expected +/-110,000 margin of error in these incomplete employment extrapolations…

Breaking down the seasonally adjusted job separations, the BLS finds that 3,056,000 of us voluntarily quit our jobs in July, down by 157,000 from the revised 3,213,000 who quit their jobs in June, while the quits rate, widely watched as an indicator of worker confidence, fell from 2.0% in June to 1.9% in July, and it was also down from the 2.0% quits rate in July a year earlier (see details in table 4)….in addition to those who quit, another 1,666,000 were either laid off, fired or otherwise discharged in July, down by 119,000 from the revised 1,785,000 who were discharged in June, as the discharges rate fell from 1.1% in June to 1.0% in July of all those who were employed during the month, which was also down from the 1.1% discharges rate of a year earlier….meanwhile, other separations, which includes retirements and deaths, were at 350,000 in July, up from the revised 340,000 other separations in June, for an ‘other separations rate’ of 0.2%, the same as in June and as in July of last year….both seasonally adjusted and unadjusted details by industry and by region on hires and job separations, and on job quits and discharges can be accessed using the links to tables at the bottom of the press release…

US Trade Deficit Rose 24.4% in July on Higher Imports of Capital Goods, Lower Exports of Crude Oil

Our trade deficit increased by 24.4% in July, after decreasing by a revised 6.0% in June, as the value of our exports decreased while the value of our imports increased…the Commerce Department’s report on our international trade in goods and services for July indicated that our seasonally adjusted goods and services trade deficit increased by $17.4 billion to a rounded $88.6 billion in July, up from a revised June deficit of $71.2 billion, which had previously been reported at $73.3 billion…trade figures going back to January were also revised with this report, which on net left the 2nd quarter trade deficit about $5.7 billion lower than was previously reported, suggesting a large upward revision to 2nd quarter GDP, the magnitude of which depends on the 1st quarter revisions, which will be recomputed and included in the annual revision to GDP at the end of September…

After rounding, the value of our exports fell by $6.6 billion, or 2.1%, to $310.7 billion in July, on a $6.2 billion decrease to $201.0 billion in our exports of goods, and a $0.4 billion decrease to $109.7 billion in our exports of services, while our imports rose by $10.8 billion, or 2.8 percent, to $399.3 billion, on an $11.4 billion increase to $320.6 billion in our imports of goods, partly offset by a $0.6 billion decrease to $78.7 billion in our imports of services…export prices were on average 1.3% lower in July, so a good part of this month’s decrease in exports was due to lower prices, and real exports were likely only 0.8% lower, while import prices were 0.4% lower, meaning that the increase in the value of our imports was despite lower prices, and that our real imports likely rose around 3.2%..

The news release for this month’s report gives us a brief synopsis of Exhibits 7 and 8 in the Full Release and Tables pdf for May, which details the major reasons for the decrease in our exports and the increase in our imports:

Exports of goods on a Census basis decreased $5.9 billion.

•     Industrial supplies and materials decreased $8.7 billion.
  • Crude oil decreased $4.5 billion.
  • Nonmonetary gold decreased $3.9 billion.
• Capital goods increased $1.9 billion.
• Consumer goods increased $1.7 billion.
  • Pharmaceutical preparations increased $1.0 billion.

  Net balance of payments adjustments decreased $0.3 billion.

Imports of goods on a Census basis increased $12.0 billion.

• Capital goods increased $14.4 billion.
  • Computers increased $6.9 billion.
  • Computer accessories increased $6.6 billion.
  • Semiconductors increased $1.2 billion.
• Industrial supplies and materials decreased $1.8 billion.
  • Crude oil decreased $1.8 billion.

  Net balance of payments adjustments decreased $0.6 billion.

That news release for this month’s report also summarizes Exhibit 19 in the pdf, which gives us surplus and deficit details on our goods trade with selected countries:

The July figures show surpluses, in billions of dollars, with Netherlands ($7.8), South and Central America ($6.6), Hong Kong ($3.1), United Kingdom ($2.5), Brazil ($2.4), Singapore ($1.9), Saudi Arabia ($1.3), Australia ($1.2), and Belgium ($0.9). Deficits were recorded, in billions of dollars, with Mexico ($27.5), Vietnam ($23.3), Taiwan ($18.1), China ($15.2), South Korea ($10.4), European Union ($8.9), Germany ($5.6), India ($5.0), Malaysia ($4.8), Japan ($4.2), Ireland ($3.9), Canada ($3.2), Italy ($2.5), France ($1.3), Switzerland ($0.6), and Israel ($0.5).

• The deficit with Mexico increased $7.2 billion to $27.5 billion in July. Exports decreased $0.2 billion to $32.6 billion and imports increased $7.0 billion to $60.1 billion.
• The balance with Switzerland shifted from a surplus of $2.9 billion in June to a deficit of $0.6 billion in July. Exports decreased $2.0 billion to $4.4 billion and imports increased $1.5 billion to $5.0 billion.
• The deficit with Canada decreased $3.7 billion to $3.2 billion in July. Exports increased $0.5 billion to $29.3 billion and imports decreased $3.3 billion to $32.5 billion.

To gauge the impact of July’s international trade in goods on 3rd quarter GDP growth figures, we use exhibit 10 in the pdf for this report, which gives us monthly goods trade figures by end use category and in total, already adjusted in chained 2017 dollars, the same inflation adjustment used by the BEA to compute trade figures for GDP, except they are not annualized here….from that table, we can compute that the 2nd quarter’s real exports of goods averaged 157,830.7 million monthly in 2017 dollars, while July’s inflation adjusted exports came in at 150,810 million in that same 2017 dollar quantity index representation….figuring the annualized change between those two figures, we find that July’s real exports of goods are running at a 16.6% annual rate below those of the 2nd quarter, or at a pace that would subtract about 1.18 percentage points from 3rd quarter’s GDP if it were continued through August and September…..in a similar manner, we find that our 2nd quarter real imports of goods averaged 250,698.3 million monthly in chained 2017 dollars, while the similarly inflation adjusted July goods imports were at 257,197 million…that would indicate that so far in the 3rd quarter, our real imports of goods have grown at a 10.78% annual rate from those of the 2nd quarter…since imports subtract from GDP because they represent the portion of consumption or investment that occurred during the quarter that was not produced domestically, their increase at a 10.78% rate would subtract about 1.11 percentage points from 3rd quarter GDP….hence, if the July trade deficit is maintained at the same level throughout the 3rd quarter, our deteriorating balance of trade in goods over that of the 2nd quarter would subtract a net of about 2.29 percentage points from the growth rate of 3rd quarter GDP….

You might note that we have not even computed the impact of the usually less volatile change in services here, because the BEA does not provide inflation adjusted data on those, and we don’t have a straightforward way to adjust the various services for all their price changes, but that our exports in services fell $0.4 billion in July, while our imports in services fell $0.6 billion, which would suggest a small boost to GDP from the services side of the international trade ledger…

Construction Spending Fell 0.5% in July after June Spending was Revised Slightly Higher

The Census Bureau report on construction spending for July (pdf) estimated that the month’s seasonally adjusted construction spending would work out to $2,157.6 billion annually if extrapolated over an entire year, which was 0.5 percent (±0.8 percent)* below the revised annualized estimate of $2,167.7 billion of construction spending in June, and 2.8 percent (±1.5 percent) below the estimated annualized level of construction spending in July of last year….the June construction spending estimate was revised less than 0.1% higher, from $2,166.5 billion to $2,167.7 billion, while the annual rate of construction spending for May was revised a bit lower, from $2,168.5 billion to $2,168.4 billion….on net, those revisions mean that construction during the 2nd quarter was about $1.1 billion greater, at an annual rate, than the figures used in last week’s GDP estimate, which would suggest there’d need to be an upward revision of about 0.02 percentage points to 2nd quarter GDP when the third estimate is released at the end of September, assuming the net impacts from the inflation adjustments on the revisions are similar to those we saw in the 2nd GDP estimate…

A further breakdown of the different subsets of construction spending is provided in a Census summary, which precedes the detailed spreadsheets:

• Private Construction: Spending on private construction was at a seasonally adjusted annual rate of $1,614.2 billion, 0.5 percent (±0.3 percent) below the revised June estimate of $1,622.9 billion. Residential construction was at a seasonally adjusted annual rate of $859.0 billion in July, 1.3 percent (±1.3 percent)* below the revised June estimate of $870.6 billion. Nonresidential construction was at a seasonally adjusted annual rate of $755.2 billion in July, 0.4 percent (±0.3 percent) above the revised June estimate of $752.4 billion.
• Public Construction: In July, the estimated seasonally adjusted annual rate of public construction spending was $543.4 billion, 0.2 percent (±1.6 percent)* below the revised June estimate of $544.7 billion. Educational construction was at a seasonally adjusted annual rate of $112.3 billion, 0.2 percent (±1.6 percent)* below the revised June estimate of $112.5 billion. Highway construction was at a seasonally adjusted annual rate of $150.3 billion, 0.2 percent (±4.6 percent)* below the revised June estimate of $150.6 billion.

Construction spending data inputs into 3 subcomponents of GDP; investment in private non-residential structures, investment in residential structures, and into government investment outlays, for both state and local and Federal governments…however, getting an accurate read on the impact of July spending reported in this release on 3rd quarter GDP is difficult because all figures given here are in nominal dollars and as you know, data used to compute the change in GDP must be adjusted for changes in price…the National Income and Product Accounts Handbook, Chapter 6 (pdf), lists a multitude of PPI indices and a number of privately published deflators that are used by the BEA for each of the various components of non-residential investment, so in lieu of trying to adjust for all of those different price indices, we’ve opted to just use the producer price index for final demand construction as an inexact shortcut to make the price adjustment needed to make an approximate estimate…

That price index showed that aggregate construction costs were up 2.2% in July, after rising 0.1% in June, and after rising 0.1% from April to May…on that basis, we can estimate that July construction costs were roughly 2.4% more than those of April, roughly 2.3% more than those of May, and obviously 2.2% more than those of June…we then use those percentages to inflate the lower priced spending figures for each of those months, which is arithmetically the same as deflating higher priced July construction spending, for comparison purposes…annualized construction spending in millions of nominal dollars for the months of the second quarter is given as 2,167,698 for June, 2,168,396 for May, and 2,168,231 for April, while it was at 2,157,581 million in July …thus to compare July’s inflation adjusted construction spending to inflation adjusted spending of the second quarter, our arithmetic formula becomes: 2,157,581 / (((2,167,698 * 1.022) + (2,168,396 * 1.023) + (2,168,231 * 1.024)) / 3) = 0.97277, meaning real construction spending in July was down 2.72% vis a vis the 2nd quarter, or down at a 10.5% annual rate…to figure the effect of that change on GDP, we take the difference between the ‘inflation adjusted’ second quarter spending average and that of July, and then take that result as a fraction of 2nd quarter GDP, and estimate that aggregate July construction spending is falling at a rate that would subtract approximately 0.35 percentage points from 3rd quarter GDP, should we see no improvement from July’s adjusted figures in August or September…

Factory Shipments Rose 0.9% in July, Factory Inventories Rose 0.3%

The July Full Report on Manufacturers’ Shipments, Inventories, & Orders (pdf) from the Census Bureau reported that the seasonally adjusted value of new orders for manufactured goods rose by $5.8 billion or 0.9 percent to $663.6 billion, following a decrease of 0.2% to $657.8 billion in June, which was revised from the 0.3% decrease to $656.5 billion reported last month….however, since the Census Bureau does not even collect data on new orders for non durable goods for this widely watched “factory orders report”, both the “new orders” and “unfilled orders” sections of this report are really only accurate as a revised update to the July advance report on durable goods we reported on last week…on those revisions, the Census Bureau’s own summary, which precedes their detailed spreadsheet of the metrics included in this report, is quite complete, so we’ll just quote directly from that here:

• Summary: New orders for manufactured goods in July, up following two consecutive monthly decreases, increased $5.8 billion or 0.9 percent to $663.6 billion, the U.S. Census Bureau reported today. This followed a 0.2 percent June decrease. Shipments, up nine of the last ten months, increased $5.3 billion or 0.8 percent to $658.8 billion. This followed a virtually unchanged June increase. Unfilled orders, up twenty-four of the last twenty-five months, increased $9.9 billion or 0.6 percent to $1,600.3 billion. This followed a 0.6 percent June increase. The unfilled orders-to-shipments ratio was 6.81, down from 6.84 in June. Inventories, up ten consecutive months, increased $3.5 billion or 0.4 percent to $966.9 billion. This followed a 0.1 percent June increase. The inventories-to-shipments ratio was 1.47, unchanged from June.
• New Orders for manufactured durable goods in July, up four of the last five months, increased $3.6 billion or 1.1 percent to $339.4 billion, unchanged from the previously published increase. This followed a 0.6 percent June increase. Transportation equipment, up following two consecutive monthly decreases, led the increase, $2.6 billion or 2.3 percent to $116.2 billion. New orders for manufactured nondurable goods increased $2.2 billion or 0.7 percent to $324.2 billion. 
• Shipments of manufactured durable goods in July, up ten of the last eleven months, increased $3.1 billion or 0.9 percent to $334.6 billion, down from the previously published 1.0 percent increase. This followed a 1.0 percent June increase. Transportation equipment, up nine of the last ten months, led the increase, $1.4 billion or 1.3 percent to $111.6 billion. Shipments of manufactured nondurable goods, up seven of the last eight months, increased $2.2 billion or 0.7 percent to $324.2 billion. This followed a 0.9 percent June decrease. Petroleum and coal products, up six of the last seven months, led the increase, $1.6 billion or 2.4 percent to $69.8 billion.
• Unfilled Orders for manufactured durable goods in July, up twenty-four of the last twenty-five months, increased $9.9 billion or 0.6 percent to $1,600.3 billion, unchanged from the previously published increase. This followed a 0.6 percent June increase. Transportation equipment, up eleven of the last twelve months, led the increase, $4.6 billion or 0.5 percent to $1,006.3 billion.
• Inventories of manufactured durable goods in July, up ten consecutive months, increased $2.4 billion or 0.4 percent to $604.7 billion, unchanged from the previously published increase. This followed a 0.4 percent June increase. Primary metals, up seventeen consecutive months, led the increase, $0.8 billion or 1.5 percent to $52.0 billion. Inventories of manufactured nondurable goods, up five of the last six months, increased $1.1 billion or 0.3 percent to $362.2 billion. This followed a 0.3 percent June decrease. Petroleum and coal products, up following two consecutive monthly decreases, led the increase, $0.8 billion or 1.6 percent to $47.6 billion. By stage of fabrication, July materials and supplies increased 0.2 percent in durable goods and decreased 0.1 percent in nondurable goods. Work in process increased 0.6 percent in durable goods and 1.1 percent in nondurable goods. Finished goods increased 0.4 percent in durable goods and 0.3 percent in nondurable goods.

To estimate the effect of those July factory inventories on 3rd quarter GDP, they must first be adjusted for changes in price with appropriate components of the producer price index…by stage of fabrication, the value of finished goods inventories were valued 0.4% higher at $329,960 million; the value of work in process inventories rose 0.7% to $273,540 million, and materials and supplies inventories were valued 0.1% higher at $363,362 million.…at the same time, the July producer price index reported that prices for finished goods were on average 0.7% lower, that prices for intermediate processed goods were on average 0.6% lower, while prices for unprocessed goods were 1.8% lower….assuming similar valuations for like types of inventories, those price changes would suggest that July’s real finished goods inventories were about 1.1% higher, that real inventories of intermediate processed goods were about 1.3% higher, and real raw material inventories were about 1.9% higher…since real NIPA factory inventories were down sharply in the 2nd quarter, accounting for about two thirds of the quarter’s inventory decrease, that this report appears to indicate a modest real increase in aggregate July factory inventories would therefore mean that the difference between the 2nd quarter decrease and the July increase would be added to the 3rd quarter’s real growth in GDP…

(the above is the synopsis that accompanied my regular Sunday morning news links emailing, which in turn was mostly selected from my weekly blog post on the global glass onion…if you’d be interested in receiving my weekly emailing of selected links, most of which are chosen from the aforementioned GGO posts, contact me…)