Haver’s data services are provided through the DLX platform; delivering tools for data management, visualization, modeling and publishing. DLX is designed to increase productivity and integrates with Excel, PowerPoint, and major statistical software packages; MatLab, R, Python, SAS, E-views and Stata. DLX component application DLXVG3 offers precision graphics, publishing, analytic tools, and easy navigation for all Haver databases.

icon-chart.png

Chart and compare multiple data series with accompanying data table of values

Advanced functions (growth rates, moving averages, aggregation, seasonal adjustment)

Analytical tools (cycle charts, scatterplots, trend lines, and recession shading)

Complete graph customization (axis scaling, sizing, formatting, styles)

Excel and PowerPoint direct export and complete integration (dynamic spreadsheets and presentations update on a click)

DLXVG3

icon-db.png

DLX Add-ins for all versions of Excel provide instant data updates of entire workbooks

Use DLXVG3 to locate data series codes from within any file, easily creating new spreadsheets for analysis

Edit date ranges, archive data, apply functions

Excel Spreadsheet Add-ins

icon-time.png

Haver offers timely, accurate, and professionally supported data via an update service, meeting the strictest demands of the institutional research community

AutoUpdate provides fast and reliable data updates within seconds of source releases

Timer service runs continuously on an hourly basis when DLX is running

Manual updates offer clients the independence to update data retrievals on any desired schedule

All database updates retrieve only the data series that have changed and send only the data points that have been updated; including revisions

Timely Updates

DLX software tools to convert data into many different formats needed for diverse analytical frameworks

Haver supports a DLX API permitting clients to leverage and build processes with DLX data

Database Compatibility

products-dlx.svg

Data Link Express

Platform Independent site delivering DLX Databases across any workstation or device.

products-haverview.svg

HaverView

[Matlab](http://matlab.com) by The MathWorks lets you access, visualize, and analyze historic time-series data to identify patterns or uncover complex relationships. From within a single environment you can:

- Access data from multiple sources including files, spreadsheets, databases, data providers, and the Web
- Store data in financial time-series objects to simplify data management, data transformation, missing-data handling, and date-math calculations
- Perform technical analysis with moving averages, oscillators, stochastics, and indices
- Create custom analysis routines and visualizations or animations to capture, present, and share your analysis processes and discoveries
- Customize your analysis environment with advanced functionality from signal processing, statistics, or econometrics

products-matlab.svg

MATLAB

[Stata Data Analysis and Statistical Software](https://www.stata.com) by StataCorp LP is an integrated statistical package designed for researchers of all disciplines. It provides the ability to do data analysis, data management and graphics and has features such as multiple imputation, factor variables, GMM, competing-risks regression, state-space modeling, predictive margins, and more.

products-stata.svg

Stata

[SAS](https://www.sas.com) provides an integrated environment for predictive and descriptive modeling, data mining, text analytics, forecasting, optimization, simulation, experimental design and more. From predictive analytics to model deployment and process optimization, SAS provides a range of techniques and processes for the collection, classification, analysis and interpretation of data to reveal patterns, anomalies, key variables and relationships, leading ultimately to new insights and better answers faster.

[SASEHAVR](http://support.sas.com/documentation/cdl/en/etsug/67525/HTML/default/viewer.htm#etsug_sasehavr_gettingstarted01.htm) is a seamless interface between Haver and SAS data processing that enables SAS users to read economic and financial time series data that reside in a Haver Analytics DLX (Data Link Express) database. [SASEHAVR](http://support.sas.com/documentation/cdl/en/etsug/67525/HTML/default/viewer.htm#etsug_sasehavr_gettingstarted01.htm) requires both SAS/ETS and Base SAS Software.

products-sas.svg

[RATS (Regression Analysis of Time Series) Econometric Software](https://estima.com) by Estima is a leading econometrics and time-series analysis software package. It provides ability to do linear and non-linear least squares, forecasting, SUR and ARIMA models and in addition, supports advanced techniques such as GMM, ARCH and GARCH models, state space models, and more.

products-rats.svg

RATS

Session 1: Bootcamp Part 1

Visualization with GGPlot2 and data analysis

Session 2: Bootcamp Part 2

Session 3: Univariate Models Part 1

Session 4: Univariate Models Part 2

Session 5: Vector Autoregressions

Session 6: Cointegration and error-correction models

A popular open source option for statistical computing and graphics that is available on many UNIX platforms, Windows and MacOS

Estimating, evaluating, and forecasting ARIMA models

Understanding the concept of non-stationarity and testing for unit roots

Estimating autoregressions, granger causality tests, and impulse response functions

Understanding the concept of cointegration and testing using the Enger-Granger test

Estimating and forecasting with vector error correction models

products-r.svg

Teaches econometrics and forecasting at the University of Colorado Denver. He has over 15 years of econometric experience in both the private and public sector.

Daniel L. Jerrett, Ph.D.

Abdel M. Zellou, Ph.D.

Bootcamp and Econometric Tools 1 & 2

Econometrics Training: Applied Time Series in R

Session 5: Multivariate Models Part 1

Session 6: Multivariate Models Part 2

Haver partners with SmartMacro Labs to offer a two week immersive online course introducing participants to the power of the Python programming language. This applied hands-on training develops statistical economics skills with a concentration on economics and finance.

Gain proficiency in building modern time-series models in Python through real-world applications. Designed for professionals in both the public and private sector, the course does not require prior experience, assuming only that participants have a basic understanding of regression and statistical concepts.

Proficiency working in Jupyter notebooks and the Python environment

products-python.svg

Remote Training: Applied Time Series in Python

US labor's share of total income at a series low

Tian Yong Woon

In this week’s Letter, we look at Southeast Asia’s race to host the data centres behind the AI boom, and at how much of the gain stays at home. Asia already anchors the AI hardware supply chain, from Taiwan’s chip fabrication to South Korea’s memory, while Southeast Asia has centred on packaging, testing and assembly (chart 1). The region is now moving to the final stage: hosting the computing capacity itself. Malaysia offers the clearest example. National growth held above 5% in 2024 and 2025, but Johor grew 8.0% last year as its services sector expanded 9.9% (chart 2). Foreign direct investment underpins the boom, with Malaysia’s FDI position in information and communication tripling between 2022 and 2025 (chart 3). Strains on power and water have followed, prompting new tariffs and tighter approvals. Thailand is on a similar path. Digital industries, including data centres, took almost half of approved FDI in 2025, while approvals elsewhere fell (chart 4). Concentration in the water-stressed Eastern Economic Corridor and gaps in oversight have led the Board of Investment to pause new data centre applications. Singapore offers a cautionary tale. Operating surplus in its data processing and hosting industry rose roughly sevenfold between 2018 and 2024 (chart 5). Yet resource constraints prompted a moratorium on new data centres from 2019 to 2022, which partly redirected investment to Johor. Finally, we ask where the gains go (chart 6). Imported equipment absorbs much of the build phase, while foreign ownership, depreciation and thin staffing shrink what residents keep once facilities are running. The wedge between GDP and national income is largest where facilities are foreign-owned and capital-heavy, as under the region’s FDI-led model.

**The AI supply chain and Asia**
Asia already sits at the heart of the AI hardware supply chain (chart 1). Taiwan leads in logic fabrication, packaging, testing and server assembly, while South Korea leads in high-bandwidth memory. China holds leading positions in packaging, testing and data centres. Southeast Asia’s role has so far sat further along the chain. Malaysia plays a significant part in packaging and testing, Vietnam in testing and Thailand in server and rack assembly. The region’s ambitions are now moving to the final stage. Malaysia and Singapore already play a significant role in hosting data centres, while Thailand, Indonesia, Vietnam and the Philippines are emerging players. For economies that earned their place by assembling and testing chips, hosting the computing capacity those chips power is a natural next step. It is also a resource-hungry one, as the rest of this Letter shows.

261110T1.jpg

**Malaysia**
Malaysia’s economy has held up well. Real GDP grew above 5% in both 2024 and 2025, helped by electrical and electronics (E&E) exports, which rose 18.3% to MYR 711.6bn last year. Within that national picture, Johor stands out (chart 2). The state’s GDP grew 8.0% in 2025, outpacing Malaysia as a whole for a third straight year. The gap has also widened, from 0.6 percentage points in 2023 to 2.8 points in 2025. Services have done most of the work, and now make up 54.8% of Johor’s economy. Johor’s services output expanded 9.9% in 2025, against 4.0% for its manufacturing, and accounted for two-thirds of the state’s growth. The ongoing data centre boom has been a key driver, with the industry classified within information and communication services.

261110T2.jpg

Foreign direct investment underpins the boom. Malaysia’s FDI position in information and communication, the sector that includes data centres, tripled from MYR 39.0bn in 2022 to MYR 117.4bn in 2025 (chart 3). Net inflows over four quarters have climbed from MYR 2.9bn in 2022 to MYR 38.1bn in the year to Q2 2026. Johor is the main destination, hosting over 70% of Malaysia’s operational data centre capacity, and the Johor-Singapore Special Economic Zone offers preferential tax incentives for the sector. The facilities are hungry for power and water, and authorities have responded on both fronts. A dedicated ultra-high-voltage tariff took effect in July 2025, followed by a separate water tariff for data centres in August. Approvals have also tightened. Johor has capped water usage effectiveness, and the federal government said in October that states without sufficient water will not host new projects.

261110T3.jpg

**Thailand**
Thailand is on a similar path. Approved FDI applications in digital industries, which include data centres and cloud services, reached THB 553.0bn in 2025, up from THB 95.2bn in 2024 (chart 4). Their share of total approved FDI rose from 13.1% to 48.5%, and they more than accounted for the overall increase. Approvals outside digital industries fell 7.0%. The projects cluster in the Eastern Economic Corridor (EEC) south-east of Bangkok, notably in Chonburi and Rayong, where parts of the region already face water scarcity. The concerns echo Malaysia’s: strain on power and water, and gaps in oversight. Some operators can set up without seeking Board of Investment (BOI) promotion, and Thailand lacks a unified system for overseeing the industry. Policy has shifted accordingly. The BOI has accepted no new data centre applications since April 2026, and the cabinet approved a national data centre policy committee in August.

261110T4.jpg

**Singapore**
Singapore offers a cautionary tale. Data centres sit within its information and communications (ICT) sector, which grew at double-digit rates around the turn of the decade and peaked at around 18% in 2021 (chart 5). The industry that houses them, data processing and hosting, also includes internet search. Its operating surplus, the profit component of value added, rose roughly sevenfold from SGD 1.1bn in 2018 to SGD 7.6bn in 2024, with the sharpest jump in 2019. That growth came up against resource constraints. Singapore imposed a moratorium on new data centres from 2019 to 2022, yet the industry’s surplus still nearly doubled over the period, to SGD 5.9bn. The cap on new capacity was also a partial driver of data centre investment into neighbouring Johor. For its neighbours, Singapore’s experience shows that the gains from data centres eventually run up against the resources needed to host them.

261110T5.jpg

**Gains from data centres**
Then there is the question of where the gains go (chart 6). The build phase is a one-off. Headline investment looks large, but much of it goes on imported servers, chips and equipment, which higher imports offset in GDP. What remains is domestic value added from construction, fit-out and local services, which fades once building is complete. The operating phase recurs each year, but less of it accrues to residents than GDP implies. If a facility is foreign-owned, the after-tax profits accruing to its owners are counted in GDP but not in gross national income (GNI). Depreciation is further excluded from net national income (NNI), and both measures sit closer to resident welfare than GDP. What stays is local wages and taxes, including corporate tax. Data centres need few staff once running, so their labour share is low and the lasting employment gain small. The wedge between GDP and NNI is therefore largest where facilities are foreign-owned and capital-heavy, as under the FDI-led model above. The chart captures direct effects only, leaving aside supply-chain spillovers and crowding out.

261110T6.gif

Asia

Economic Letter from Asia: AI and Data Centres

Winnie Tapasanun

261109A15.jpg

261109B15.jpg

Wholesale inventories rose a slightly more-than-expected 0.7% (7.6% y/y) in September, the eighth straight m/m gain, after rises of 0.6% in August and 1.2% in July, indicating wholesalers continued to accumulate inventories at a solid pace. A 0.6% m/m September increase had been expected in the Informa Global Markets survey. The year-on-year rate rose to 7.6% in September, the 21st consecutive y/y rise and the fastest pace since March 2023, from 7.5% in August (1.8% in September 2025), remaining well above a low of -0.9% y/y in December 2024 but far below a peak of 25.5% y/y in June 2022.

Durable goods inventories grew 0.6% (8.0% y/y) in September, up for the 11th month running, after a 0.9% rise in August. The September m/m gain in durable goods inventories was led by a 2.5% advance (23.8% y/y) in electrical & electronic goods, followed by increases of 1.3% (4.5% y/y) in hardware, plumbing & heating equipment, 1.3% (4.6% y/y) in miscellaneous durable goods, 0.7% (4.1% y/y) in lumber & other construction materials, 0.5% (15.9% y/y) in professional & commercial equipment, and 0.4% (-2.4% y/y) in furniture & home furnishings. In contrast, durable goods wholesale inventories for metals & minerals (-0.9%; +2.7% y/y), motor vehicles & parts (-0.5%; +0.9% y/y), and machinery (-0.2%; +4.6% y/y) declined m/m in September.

Nondurable goods inventories climbed 0.9% (7.0% y/y) in September, the third successive m/m rise and the seventh in eight months, after a 0.1% uptick in August. The September m/m increase in nondurable goods inventories was driven by a 10.6% rebound (51.9% y/y) in petroleum & petroleum products, followed by rises of 2.6% (3.9% y/y) in apparel, 1.4% (2.7% y/y) in miscellaneous nondurable goods, 1.3% (9.1% y/y) in chemicals, 0.7% (0.6% y/y) in drugs, and 0.2% (4.6% y/y) in paper & paper products. To the downside, nondurable goods wholesale inventories for farm products (-7.4%; +18.9% y/y), alcoholic beverages (-1.8%; -3.3% y/y), and groceries (-0.8%; +1.5% y/y) fell m/m in September.

Wholesale sales jumped a larger-than-expected 2.8% (19.8% y/y) in September, the third straight m/m increase and the 12th in 13 months, on top of a 2.0% gain in August, indicating wholesale demand strengthened further. A 1.1% m/m September increase had been expected in the Action Economics Forecast Survey. Durable goods sales rose 1.3% (23.8% y/y) in September, the third consecutive m/m rise and the 10th in 11 months, after a 2.4% increase in August. The September m/m gain in durable goods sales was led by a 6.7% advance (37.5% y/y) in metals & minerals, followed by increases of 2.7% (19.8% y/y) in machinery, 2.6% (42.9% y/y) in electrical & electronic goods, 0.7% (19.5% y/y) in miscellaneous durable goods, and 0.2% (34.0% y/y) in professional & commercial equipment. In contrast, durable goods wholesale sales for motor vehicles & parts (-1.2%; +1.5% y/y), furniture & home furnishings (-1.0%; +12.5% y/y), hardware, plumbing & heating equipment (-0.5%; +6.8% y/y), and lumber & other construction materials (-0.1%; +6.6% y/y) decreased m/m in September.

Nondurable goods sales surged 4.4% (15.9% y/y) in September, the largest m/m gain since March, stronger than a 1.5% rebound in August. The September m/m increase in nondurable goods sales was driven by an 18.8% surge (62.8% y/y) in petroleum & petroleum products, followed by rises of 4.4% (26.5% y/y) in farm products, 2.3% (-2.6% y/y) in miscellaneous nondurable goods, 1.2% (19.0% y/y) in chemicals, 0.5% (-2.5% y/y) in alcoholic beverages, and 0.1% (2.4% y/y) in groceries. To the downside, nondurable goods wholesale sales for drugs (-1.5%; +4.7% y/y), paper & paper products (-0.8%; +4.9% y/y), and apparel (-0.6%; +1.3% y/y) fell m/m in September.

Wholesale sales outpaced inventory growth in September, pushing the inventory-to-sales ratio in the wholesale sector down to 1.21 in September, matching its lowest reading since May, from 1.23 in August (1.34 in September 2025); it was below a high of 1.42 in June 2023. The durable goods I/S ratio dipped to 1.49 in September, the lowest since July 2021, from 1.50 in August; it was below a high of 1.87 in March 2024. The durable goods I/S ratio ranged from 2.65 for machinery to 0.98 for electrical & electronic goods. The nondurable goods I/S ratio eased to 0.92 in September, the lowest since May, from 0.95 in August; it was below a high of 1.06 in March 2023. The nondurable goods I/S ratio ranged from 2.41 for apparel to 0.39 for petroleum & petroleum products.

The wholesale trade figures are available in Haver's **USECON** database. The expectations figure for inventories is contained in the **MMSAMER** database. Expectations for sales are from the **AS1REPNA** database.

261109C15.jpg

261109D15.jpg

- Wholesale inventories +0.7% in Sept., eighth straight m/m gain; +7.6% y/y, highest y/y pace since March ’23.
- Durable goods inventories +0.6% (+8.0% y/y), 11th consecutive m/m increase.
- Nondurable goods inventories +0.9% (+7.0% y/y), seventh m/m rise in eight mths.
- Wholesale sales +2.8% (+19.8% y/y), 12th m/m gain in 13 mths., w/ durable goods sales up 1.3% and nondurable goods sales up 4.4%.
- I/S ratio down to 1.21, matching its lowest level since May.

U.S. Wholesale Inventories Increase in September; Sales Jump

Robert Brusca

261109E.jpg

Japan's leading economic index, released by the Cabinet Office, has moved up from much lower readings in mid-2025. In July and August, it made two sizable moves higher, reaching a peak in August at an index level of 118. However, in September, the index fell sharply to a level of 116.4, a sharper fall than had been expected (but this reading is still preliminary). After the fall, the September reading is only marginally higher than its readings in May and June of this year. 

The leading index has been boosted by rising share prices until recently, and the interest rate spread has also signaled advance as the Bank of Japan has been raising rates.

As a reference, I include the consumer confidence indicator where we see that confidence has been mildly rising since April of this year; yet, the gradient has been extremely mild and it comes after a more lengthy period of weakness for that indicator, so that the year-over-year increase in confidence shows a 1.1% rise. However, the change over 24 months for consumer confidence is negative. But over the same 24 months, the May index is up by 7.4%, and share prices and the interest rate spread also show increases over that two-year span.

Through September, the Economy Watchers index is showing modest contraction in its headline and modest expected contraction in its future index, with a percentile standing for the Economy Watchers current index in its 47th percentile, whereas the outlook index is only in its 41st percentile; neither one of those is registering much strength. Both of those readings are below their historic medians; the median on ranking data occurs at a percentile standing of 50%.

The detail from the Economy Watchers index tells us that households are expressing some degree of normalcy in current conditions that is showing contraction but that nonetheless has a historic percentile standing above its median, ranking in its 53.4 percentile. The eating and drinking places category is a reading that tends to be volatile and depends on the overall state of the economy. It is showing less shrinkage in September than in August, but it has a queue standing that is elevated, in its 60th percentile, relatively firm. Manufacturers also have a queue percentile standing above neutral. But the overall picture remains modest.

261109.jpg

**Summing up**
The leading index reflects these modest conditions for growth. However, even with these cooling LEI readings, the Bank of Japan is still set for a rate hike, perhaps as soon as its next meeting. While the Cabinet Office leading index has declined, the current conditions index continues to improve in September. Thus, growth in Japan is balanced and strong enough that action against too-high inflation remains likely. The economy also will be benefiting from some fiscal stimulus. But financial conditions and particularly foreign exchange rate behavior could shift and set a new background and create a new policy dilemma at the drop of a hat.

Japan

Japan’s LEI Weakens in September

Michael J. Moran

261106A.jpg

261106B.jpg

Nonfarm payrolls rose 165K in October, marking one of the best readings of the past few years and easily beating the expected increase of 65K. Moreover, the results in August and September were revised upward by a combined total of 84K. The new figures left the six-month average at 82K, a far better performance than the average monthly increase of only 10K in all of 2025.

Several sectors stood out on the strong side of job growth. Employment in the business-service category jumped 70K, the strongest showing since mid-2022, when the economy was still experiencing the post-Covid boom. In addition, business services contributed importantly to the upward revisions in August and September (an additional 52K jobs). Manufacturing industries are showing signs of life, as they posted their fifth consecutive monthly gain (up 19K) and added 10K to the revisions total. The health-care sector, a reliable area for job growth, had posted below-average results in the prior five months, but the gain of 55K in October brought it close to its long-run average.

The headline unemployment rate dipped 0.1 percentage point to 4.1%, a reading comfortably within the range of 4.0-4.5% seen in the past two years (although above readings in the neighborhood of 3.5% in the latter part of 2022 and the early months of 2023). It was a “strong” drop in unemployment in the sense that job growth as measured by the household survey exceeded the increase in the size of the labor force (463K versus 421K). October marked the third consecutive month of strong gains for employment and the labor force in the household survey, ending a string of eight months of generally negative results.

Although the headline unemployment rate eased in October, the broad unemployment rate (the so-called U-6 measure) rose 0.3 percentage point to 7.9%. This measure includes individuals working part-time involuntarily and individuals who would like a job and have searched for employment in the past 12 months but have not looked in the past four weeks. Both of these metrics rose in October, contributing to the increase in broad unemployment. The jump in the U-6 measure reversed a good portion of the noticeable improvement since late last year.

Average hourly earnings rose 0.4% in October, matching the firmest increase of the year thus far. The latest results left the year-over-year change at 3.1%, about equal to the readings in the prior two months, suggesting that the downward trend in wage growth may be starting to stabilize.

The surge in energy prices this year has pushed CPI inflation above the annual increase in average hourly earnings, leaving a drop in aggregate real wage income. However, the financial position of the typical wage earner is difficult to assess because the recent period of negative real wage growth followed several years of real gains, which followed sharp real declines during the inflation burst in 2021-22. The ups and downs over the past several years have been largely offsetting, but individuals in total have registered real wage gains. Still, the surge in prices this year has been sharp enough to generate strains for many households. The strains certainly seem to have had an influence on the midterm elections.

The average workweek is an underappreciated element of the monthly employment report; it receives little attention in the financial press and is not scrutinized by financial market participants. However, it is an important indicator. The current workforce comprises more than 159 million individuals, and thus a change of just a few tenths of an hour can have a noticeable effect on production. The average workweek remained at 34.4 hours for the third consecutive month in October, a better showing than readings of 34.2 to 34.3 in the prior two years or so.

The employment and earnings data are collected from surveys taken each month during the week containing the 12th day of the month. The labor market data are contained in Haver's **USECON** database. Detailed figures are in the **EMPL** and **LABOR** databases. The expectations figures are in the **AS1REPNA** database.

261106C.jpg

261106D.jpg

- Payrolls posted a solid gain; Aug/Sep revised upward.
- Headline unemployment edged lower, although broad unemployment rose.
- Average hourly earnings exceeded expectations.

October Employment Report: Firm in Most Respects

261106E.jpg

German industrial production rose by 0.7% in September after rising 1% in August. Consumer goods and capital goods sectors also had strong gains in each of the last two months. Intermediate goods showed milder and more erratic output results and trends.

Sequentially, IP headline gains are 0.5% over 12 months; that rises to a 1.5% annualized pace over six months but edges slightly lower to a 1.3% pace over three months. Still, there are output gains on all horizons, and the three-month and six-month gains are each stronger than the year-on-year.

Consumer goods output sequentially is less buoyant even over three months, largely because of a huge drop in July that gains in August and September could not erase. So, the sequential output trend for consumer goods is erratic despite two solid months of back-to-back gains. Capital goods output shows acceleration and building momentum, with its annualized pace at 8% over three months. Intermediate goods output shows contractions over each of the sequential horizons and a tendency toward deceleration.

September data conclude the data for Q3. On a full quarter-over-quarter basis, output is falling in two of three sectors. The exception is capital goods, where output posts a quarterly gain of just 0.4% at an annual rate.

To put growth in a greater perspective, annual growth rates of output ranked on data back to 2006 place total German IP growth at a 50.8 percentile standing. Capital goods have a 52.4 percentile standing for growth, while consumer goods growth is a laggard at a 36.7 percentile standing and intermediate goods at a 42.7 percentile standing. Only total IP growth and capital goods have standings above their respective 50% mark, placing them above their median gain on this horizon.

German manufacturing output, real orders, and real sales show erratic-to-weaker trends. Manufacturing output shows gains on all sequential horizons, while real orders and real sales are both showing significant declines as well as a pattern of deceleration. Manufacturing output and real sales are falling in Q3. But in terms of ranked year-over-year growth rates, real orders and real sales have standings above their historic median rates.

The survey indicators for the German economy show a progression toward strength over the last three months, and so do the sequential readings, for the most part. All the indicators are higher month-to-month. But their percentile standings, conducted on levels of these survey data, all register below median standings on data back to 2006. Recent performance is good, but it is weak in a historic context.

261106.jpg

Data for three early reporting EMU members, France, Spain, and Portugal, show output increases in September. However, their respective sequential footprints are not as upbeat. France shows output declines over each sequential horizon; Spain and Portugal are erratic. However, Spain’s IP growth rate over 12 months is above its median result, while France and Portugal have IP growth rates ranking near the lower third of their respective historic queues of growth rates for output.

On balance, the German data reported are fairly solid, with the surveys being quite upbeat. Real orders and real sales trends are cautionary and must be monitored. The three European reporters underline the fact that Europe is faring less well than Germany. Just another reason to be wary of good news. But what we do have here is a nice slug of good news. Maybe it can last?

Germany

German IP: Solid Gains in September; Strong Survey Results

Andrew Cates

**Half a pass-through**
Interest-rate expectations have moved higher this year, but long-term bond-yield forecasts have risen by much less. The Blue Chip Financial Forecasts panel has repeatedly revised its policy-rate outlook upwards, reflecting a more restrictive monetary environment, while making more modest adjustments to longer-term yields (chart 1). This divergence might suggest that forecasters expect tighter monetary policy to prove less persistent than current interest-rate settings imply, or that other forces are impacting longer-term yields. Japan provides an important contrast. There, long-term yield forecasts have risen more sharply than policy-rate expectations, reflecting growing concerns about government borrowing and fiscal sustainability. The broader message is that bond markets are responding not only to the outlook for monetary policy, but also to differences in fiscal conditions and the credibility of public finances.

261105Z1.jpg

**The market has already voted**
South Korea offers a striking example of the gap between market prices and economic forecasts. The KOSPI has fallen sharply since its June peak, yet consensus growth projections have continued to improve (chart 2). Korea's position at the heart of the global semiconductor supply chain makes it especially sensitive to changes in expectations for AI spending. Its equity market may be signalling a turn in the investment cycle that forecasters have yet to acknowledge. Equities are imperfect predictors, of course, and a correction need not foreshadow an economic downturn. But the divergence may now be too large to dismiss.

261105Z2.jpg

**Borrowers have stopped asking**
The latest Federal Reserve loan officer survey published this week offers a less reassuring view of the American economy. Demand for mortgages, commercial property loans and broader business credit has weakened sharply and together (chart 3). That matters because conventional (i.e. non-AI-related) investment has already been lacklustre. Large-scale technology construction may be sustaining aggregate capital spending, but it does not mean the wider private sector is eager to expand. Monetary policy could already be restraining the borrowers it reaches most directly, even as the most conspicuous investment boom continues elsewhere. This is an uncomfortable imbalance for the Fed — and a warning against treating strong headline investment as evidence of broad-based strength.

261105Z3.jpg

**Where the credit is going instead**
US credit has not disappeared. It has increasingly changed its route. US banks have expanded lending to non-bank financial institutions at a remarkable pace in recent months, and such loans now account for a much larger share of bank balance sheets than they did before the pandemic (chart 4). Some of that money ultimately supports borrowers through private funds and other intermediaries, making the final destination of credit harder to trace. The IMF has also drawn attention to complex financing arrangements within the technology sector. None of this proves that AI investment is chiefly funded through these channels. It does hint that conventional bank-loan data provide an incomplete picture of leverage and risk. A less visible credit cycle is not necessarily a safer one.

261105Z4.jpg

**Lead times lengthen again**
This week’s manufacturing PMI surveys suggest that supply constraints are returning just as demand is becoming more uneven. Manufacturers across major economies report longer delivery times and renewed pressure on input costs (chart 5). Competition for electrical equipment and specialist components, alongside tariffs and disrupted trade routes, are probably adding to the strain. This is arguably not a repeat of the supply shock of 2021–22. But it is a reminder that inflation can persist even when ordinary borrowers and consumers are retrenching. Central banks face the familiar difficulty of responding to higher prices without unnecessarily amplifying weakness elsewhere.

261105Z5.jpg

**No room at the margin**
Fiscal policy offers little comfort. France's debt burden has risen markedly since 2019, while its primary balance and credit standing have deteriorated. Japan faces its own scrutiny, and the UK is hardly immune. By contrast, several countries once regarded as Europe's fiscal weak links have reduced their debt ratios and improved their standing (chart 6). The old division between prudent core and vulnerable periphery no longer fits the evidence. That helps explain why sovereign bond investors are paying closer attention to fiscal credibility. Yet the deeper concern is what happens if private investment falters. Governments with limited fiscal room may find themselves asked to support an economy at precisely the moment markets are least willing to finance the response.

261105Z6.jpg

The US midterm elections have significantly altered the political landscape in Washington. Democrats have secured control of both houses of Congress, ending unified Republican control and introducing a period of divided government for the remainder of President Trump's term. Concerns about living costs and the war with Iran appear to have influenced voting patterns, contributing to Democratic gains that exceeded expectations. The economic implications are arguably mixed. Divided government could limit the scope for further tax cuts and additional public spending, but may also complicate negotiations over budgets and federal borrowing. The election arrives at an awkward moment for markets. Confidence in the AI investment boom has already been shaken by doubts over financing, valuations and eventual returns. Greater scrutiny of data centres and energy demand adds another uncertainty, but politics did not create the underlying problem. Nor is technology the only source of unease. This week's charts show forecasters struggling to keep pace with changing interest-rate expectations (chart 1), a Korean equity market at odds with increasingly optimistic growth projections (chart 2), and a marked retreat in US loan demand (chart 3). Yet lending to non-bank financial institutions continues to expand (chart 4), supply pressures are reappearing (chart 5), and stretched public finances are unsettling sovereign bond markets (chart 6). The common theme is a world economy whose apparent resilience conceals growing imbalances — and markets that may be recognising them before forecasters do.

<iframe data-testid="embed-iframe" style="border-radius:12px" src="https://open.spotify.com/embed/episode/46b1I8K2lgwYt6eIjOZ21P?utm_source=generator&si=8f58ad559546493b" width="100%" height="152" frameBorder="0" allowfullscreen="" allow="autoplay; clipboard-write; encrypted-media; fullscreen; picture-in-picture" loading="lazy"></iframe>

Charts of the Week: When Expectations Meet Reality

Sandy Batten

261105N4.jpg

261105O4.jpg

Nonfarm business productivity (output per hour) jumped 3.6% q/q saar in Q3 following an upwardly revised 2.1% quarterly gain in Q2 (previously 1.4%). The Action Economics Forecast Survey expected a 2.7% quarterly increase in productivity. Compared with a year, productivity was up 2.1%, slightly slower than 2.3% y/y in Q2. Since productivity growth began to accelerate at the beginning of 2023, it has grown at a well above trend 2.7% annual rate, likely a reflection of AI and other technology advances. Strong productivity is a key factor for boosting real incomes and restraining inflation pressures.

Nonfarm business output (a little different from real GDP) increased 3.7% q/q saar in Q3 versus an upwardly revised 2.4% in Q2 (previously 1.7%). By contrast, real GDP grew 3.4% q/q in Q3. Hours worked edged up 0.1% q/q in Q3 following an unrevised 0.3% quarterly increase in Q2. Compensation growth slowed to 2.9% in Q3 from an upwardly revised 4.0% increase in Q2 (previously 2.6%). Compared to a year ago, compensation growth slowed to 4.5% y/y from 4.9%. With quarterly productivity growth exceeding compensation growth in Q3, unit labor costs fell 0.7% q/q in Q3 after an upwardly revised 1.9% quarterly gain in Q2 (previously +1.2%). The Action Economics Forecast Survey looked for a 1.1% q/q increase in unit labor costs in Q3. The y/y rate of advance for unit labor costs slowed, to 2.4% y/y in Q3 from 2.6% in Q2.

Manufacturing productivity growth slowed markedly in Q3, to 0.3% q/q saar from an upwardly revised 2.9% quarterly gain in Q2 (previously 2.4%). Factory output rose only 1.3% q/q in Q3 following an upwardly revised 5.8% q/q jump in Q2 (previously +5.4%). Hours worked increased 1.0% q/q, down from 2.9% (unrevised) in Q2. Compensation growth picked up to 3.0% from an upwardly revised 2.4% in Q2 (previously 2.1%). With the slowdown in productivity growth and the pickup in compensation growth, unit labor costs jumped 2.7% q/q in Q3 after a 0.4% quarterly decline in Q2 (previously -0.3%). Compared with a year ago, manufacturing unit labor costs were up 4.0% y/y, the highest yearly rate of advance since Q4 2024.

Today’s release also contained benchmark revisions from the beginning of 2021. They were modest but generally positive. Over the 22 quarters of revision, productivity grew at a revised 1.62% annual rate, compared to 1.50% prior to the revisions.

The productivity and labor cost data are available in Haver’s **USECON** database. The Action Economics expectations figures are in the **AS1REPNA** database.

261105P4.jpg

261105Q4.jpg

- Output per hour jumped a larger-than-expected and well above trend 3.6% q/q saar in Q3 with an upward revision to Q2.
- Compensation growth slowed to 2.9% q/q resulting in a 0.7% quarterly decline in unit labor costs. 
- Benchmark revisions back to the beginning of 2021 were modest but generally positive, pushing up productivity growth from Q4 2020 to 1.62% from 1.50%.

U.S. Productivity Jumped in Q3 2026

Kathleen Stephansen, CBE

261105A5.jpg

261105B5.jpg

Initial claims for unemployment insurance dropped by 6,000 to 189,000 in the week ending October 31 from 195,000 in the week ending October 24, revised from 194,000. The four-week average was 194,750 in the October 31 week, down from 197,000 in the October 24 week, revised from 196,750. Claims matched the July 18 week’s level and remain range-bound at historically low levels. The Action Economics/Haver Analytics Survey looked for 196,000 claims to have been filed.

The total number of unemployment insurance beneficiaries—also known as “continuing claims”—rose 23,000 to 1.718 million in the week ending October 24, up from 1.695 million in the week ending October 17, revised from 1.701 million. The four-week moving average was 1.710 million in the October 24 week, little changed from 1.709 million in the October 17 week. The insured unemployment rate was unchanged at 1.1% in the October 24 week.

The insured unemployment rate varies greatly across individual states and territories. In the week ending October 17, no state had a rate above 2.00%; the highest rates were in Massachusetts (1.71%), Washington (1.82%), and New Jersey (1.98%). The lowest rates were in South Dakota (0.22%), Louisiana (0.28%), Florida (0.31%), Alabama (0.33%), North Carolina and North Dakota (both 0.34%), Arkansas (0.37%), Iowa and Nebraska (both 0.39%), Kansas and Tennessee (both 0.43%), New Hampshire (0.46%), Georgia (0.47%), Kentucky, Mississippi and Virginia (all 0.48%), and Idaho (0.49%). Rates in other notable states include Texas (0.93%), Pennsylvania (1.21%), Illinois (1.28%), New York (1.40%) and California (1.70%). These state data are not seasonally adjusted.

Data on weekly unemployment claims are from the Department of Labor itself, not the Bureau of Labor Statistics. They begin in 1967 and are contained in Haver’s **WEEKLY** database and summarized monthly in **USECON**. Data for individual states are in **REGIONW** back to December 1986.

261105C5.jpg

261105D5.jpg

- New claims declined by 6,000 to 189,000 in the week of October 31.
- Continuing claims rose by 23,000 to 1.718 million in the week ending October 24.
- The insured unemployment rate held at 1.1% in the week of October 24.

U.S. Initial Unemployment Claims Dropped in the Week of October 31

large_261112F56.jpg

medium_261112F56.jpg

small_261112F56.jpg

thumbnail_261112F56.jpg

261112F56.jpg

large_261112G56.jpg

medium_261112G56.jpg

small_261112G56.jpg

thumbnail_261112G56.jpg

261112G56.jpg

Existing home sales fell 5.5% m/m (-8.8% y/y) to a lower-than-expected 3.75 million units (SAAR) in October, the fourth monthly fall in five months and the lowest level since August 2010, after a 1.0% increase to 3.97 million in September (previously +1.5%; 3.99 mil.), data from the National Association of Realtors (NAR) showed, indicating that affordability constraints, particularly high mortgage rates, continued to suppress housing demand. The Action Economics Forecast Survey had expected October sales of 3.86 million units. Sales were 41.7% below a high of 6.43 million in January 2022 and 43.2% below a peak of 6.60 million in January 2021. The October decrease coincided with a rise in the average 30-year fixed mortgage rate to 7.33%, the highest since November 2023, after increasing to 6.86% in September, according to Freddie Mac. The sales figures are based on closings of sales contracts signed over the past couple of months.

Existing single-family home sales dropped 5.8% (-8.6% y/y) to 3.41 million units in October, the fourth m/m decline in five months and the lowest level since October 2010, reversing a 1.4% gain to 3.62 million in September; sales were 40.2% below a high of 5.70 million in January 2022. Sales of condos and co-ops slid 2.9% (-10.5% y/y) to 340,000 in October, the third consecutive m/m slide and a record low, after a 2.8% decrease to 350,000 in September, remaining 53.4% below a high of 730,000 in January 2022.

Regionally, October existing home sales declined m/m in all regions except the Northeast, where they were flat. Sales in the West dropped 9.7% (-14.5% y/y) to 650,000 in October, the lowest level on record, following a 1.4% rebound in September. Sales in the South fell 6.5% (-8.0% y/y) to 1.73 million, the lowest reading since January 2012, after a 2.2% September rise. Sales in the Midwest slid 3.2% (-8.2% y/y) to 900,000, the lowest level since July 2011, after registering at 930,000 in September and August. Sales in the West, South, and Midwest posted the fourth m/m decrease in five months. Meanwhile, sales in the Northeast held steady m/m (-4.1% y/y) at 470,000 in October, matching the lowest reading since May, after a 2.1% slide in September. Notably, the South remained the dominant region, accounting for 46.1% of total U.S. existing home sales.

The median price of all existing homes (NSA) slipped 0.4% (+0.4% y/y) to $416,700 in October, the fourth successive m/m decline and the lowest since March, on top of a 2.2% decrease to $418,200 in September ($417,500 previously). Prices remained 15.5% above a low of $360,800 in January 2023. The median price of an existing single-family home slid 0.4% (+0.4% y/y) to $421,900 in October, down for the fourth straight month and the lowest since April, after a 2.3% fall to $423,600 in September. The median price of condos and co-ops, however, rose 0.7% (0.3% y/y) to $365,100, the first m/m rise since June, following a 2.1% September slide to $362,400. Regionally, overall prices fell m/m in October in the Midwest (-2.4%; +1.5% y/y), Northeast (-0.2%; +3.2% y/y), and South (-0.1%; -0.4% y/y), but rebounded m/m in the West (+2.0%; 0.0% y/y).

Inventories of homes for sale climbed to the highest level since November 2019. The number of existing homes for sale (NSA) advanced 1.3% (4.6% y/y) to 1.59 million units in October after holding steady at 1.57 million during the May to September period. The supply of homes on the market at the current selling rate (NSA) rose to 5.1 months in October, matching the highest level since November 2015, from 4.7 months in September. A record low in supply of 1.6 months was in January 2022.

The data on existing home sales, prices and affordability are compiled by the National Association of Realtors. The data on single-family home sales extend back to February 1968. Total sales and price data and regional sales can be found in Haver's **USECON** database. Regional price and affordability data and national inventory data are available in the **REALTOR** database. Mortgage interest rates can be found in the **WEEKLY** database. The expectations figure is from the Action Economics Forecast Survey, reported in the **AS1REPNA** database.

large_261112H56.jpg

medium_261112H56.jpg

small_261112H56.jpg

thumbnail_261112H56.jpg

261112H56.jpg

large_261112I56.jpg

medium_261112I56.jpg

small_261112I56.jpg

thumbnail_261112I56.jpg

261112I56.jpg

- October sales -5.5% m/m to 3.75 mil., fourth m/m decline in five mths.; -8.8% y/y, third straight y/y drop.
- Sales m/m down in the West (-9.7%), South (-6.5%), and Midwest (-3.2%); flat in the Northeast.
- Sales y/y down in all four regions, w/ the steepest decline in the West (-14.5%).
- Median sales price -0.4% (+0.4% y/y) to $416,700, lowest since March.
- Unsold inventory +1.3% (+4.6% y/y) to 1.59 mil. units, highest level since Nov. ’19; 5.1 months' supply, highest since Nov. ’15.

U.S. Existing Home Sales Drop in October to Lowest Level Since August 2010

Our global data offering is unparalleled in terms of coverage, quality, and organization. We offer 290+ databases from 2500 government and private sources, a wealth of key survey and forecast data covering the world’s economies, and a robust offering of archive databases that drive machine learning for big data applications.

Optimized for precision analytics, our data is seamlessly integrated with Microsoft Office and all leading statistical packages through our DLX® software and platform independent HaverView™ website. We further offer direct database feeds, client server or cloud-based delivery, and RESTful API.


Privacy Policy

Disclaimer

Contact us

Service Health

A premier time series data platform for the global strategy, research, and quant communities. 

New York

Tokyo

London

Singapore

Economy in Brief

Viewpoints

Contact Us

Our Data

Our Products

What We Do

Our Authors

About Us

Data News

Featured Data Additions

Resources

Release Calendars

Training Center

Webinars

Newsletters

Download Software

For Clients

Live Updates

#### World Economy Tracked for You 
Haver's worldwide economic data coverage provides a productive resource for analysts, strategists, and risk managers alike. With 45+ years as the industry leader, our breadth of data is unmatched.


#### Diverse Data Sources
From international banks to the farthest reaches of emerging markets, our coverage meets the needs of the global research community.

#### Driven By Quality
Committed to quality and transparency, all data sourced and delivered is continuously monitored and validated to ensure the highest level of accuracy. 

#### Global Data 
From archives to forecasts, primary-sourced from around the globe, we offer 290+ time-series databases.

#### Speed Matters
Delivering precise data as quickly as possible is our priority, which we accomplish in about one minute of release.

Data

#### Comprehensive Solutions
Effortlessly navigate, analyze, and visualize data with our broad array of tools and seamless integration with industry leading statistical and econometric packages. Gain insight and share your perspectives with ease.

#### Secure Data Access 
Effortlessly manage and work with data — delivered to your network or via our cloud services. The choices are easy, and the results are the same, fully secure and consistent access.

#### HaverView™
With a commitment to making data accessible for any work environment, our HaverView web application offers a flexible and productive resource for the dynamic research markets we serve. 

#### Data Exports and Workflow Management
Our DLX software integrates with MS Office allowing seamless publication in PowerPoint and Word as well as intuitive data management in Excel. Flexible data feeds are also supported by a variety of methods. 

Discover Haver!

Data

Software

Connectivity

Research

University

Archive

World Economy Tracked for You

Diverse Data Sources

Driven By Quality

Global Data

Speed Matters

Economic Commentary

U.S. Existing Home Sales Drop in October to Lowest Level Since August 2010

Recent Commentaries

Charts of the Week: Cheaper Energy, Dearer Money

The Economic Implications of the AI Data Center Buildout

German CPI Deflates...on Some Metrics—Shows Promise...and in Promise There Is Peril

U.S. Consumer Prices Increased More Than Expected in October

U.S. NFIB Small Business Optimism Little Changed in October

Japan’s Economy Watcher Index Slips in October; Outlook Fades

Latest Releases

Graph of the Week

US labor's share of total income at a series low

Economy in Brief

Viewpoints