Implications: New home sales remained in low gear in September, as affordability continued to keep potential buyers on the sidelines. Sales are now at an annual pace of 676,000, right around pre-pandemic levels, which has been a ceiling of sorts for activity the past couple of years.  Unfortunately, ongoing conflicts in the Middle East and between Russia and Ukraine are having upward impacts on energy prices and inflation which have introduced new challenges. First, financing costs have risen, with the average 30-yr fixed mortgage rate up roughly 130 basis points since early this year.  Second, the Federal Reserve has begun hiking interest rates again, which will disproportionately impact the already weak housing sector. But while buyers are unlikely to get much help from interest rates, the good news is that prices have been trending lower for new builds in the past several years. Median sales prices are down 13% from the peak in October 2022.  Meanwhile, the Census Bureau reports that from Q3 2022 to Q2 2026 (the most recent data available) the median square footage for new single-family homes built rose 2.5%. So, buyers are seeing a drop in the price per square foot, not just smaller/lower cost options.  This is partially the result of developers offering incentives to buyers in order to move inventory.  Supply has also put more downward pressure on median prices for new homes than existing homes.  The supply of completed single-family homes has been trending down recently but is still up 260% versus the bottom in 2022. This contrasts with the market for existing homes, which continues to struggle to convince current homeowners to give up the low fixed-rate mortgages they locked-in during the pandemic. While financing costs continue to add uncertainty and keep buyers on the sidelines, less expensive options and an abundance of inventories may give home sales a modest boost through the end of 2026. In other housing news this morning, the FHFA index increased 0.3% in August and is up 2.5% in the past year, while the national Case-Shiller index increased 0.3% in August and is up 2.1% in the past year.

Click here for a PDF version

Implications:  New orders for durable goods surged in September, rising 1.8%, the second fastest monthly gain this year.  While the headline was boosted by a large increase in commercial aircraft orders, underlying activity remains robust due to a resurgence in capital investment.  Aircraft are often ordered in bulk, which is why orders excluding transportation provide a much better check on the broader economy. These orders continue to rise, up 0.5% in September, though growth has slowed slightly from earlier this year when monthly gains were consistently above 1.0%. That said, these orders are still up 11.1% in the past year.  The increase in September was led by industrial machinery (+1.2%), computers & electronic products (+0.8%), and primary metals (+0.5%).  Electrical equipment was the only major category to decline in September but is still up 4.8% in the past year.  All other categories have experienced a surge in growth in the past year – primary metals, industrial machinery, and computers & electronic products have each experienced double-digit growth in the last twelve months. The most important number in today’s release is core shipments – a key input for business investment in the calculation of GDP – which surged 1.2% in September, marking the eighth consecutive month of growth.  Core shipments were up at a 16.5% annualized rate in Q3 versus the Q2 average, an acceleration from the 11.6% pace in Q2.  These shipments are up 11.5% over the past year, the largest annual gain outside the COVID years since 2012, as capital spending from the hyperscalers, the reshoring of production, and a more favorable tax environment for investment have helped drive a strong increase.  In other manufacturing news, the Kansas City Fed Manufacturing Index, a measure of factory sentiment in that region, declined to +11 in October from +14 in September.  The Richmond Fed index, a measure of mid-Atlantic factory activity, remained at -2 in October.  On the employment front, initial claims fell 6,000 two weeks ago to 194,000 while continuing claims rose 25,000 to 1.728 million. These figures suggest continued job growth in October.

Click here for a PDF version

We estimate the US economy grew at a 3.3% annual rate in the third quarter.  Solid.  Robust even, considering that net immigration is near zero and possibly negative, while Baby Boomers continue to gradually drift into retirement.

However, don’t get too excited.  Even if we get 3.3% for Q3, Real GDP will be up a modest 2.0% from a year ago.  And there are still signs that much of the growth is concentrated in AI and data centers.  Excluding that, the economy is tepid and could possibly get weaker if the Federal Reserve keeps raising short-term interest rates.

How strong is the tech sector?  Even if data center construction is (oddly) unchanged in September, staying at the August level, nominal spending on data centers will be up at an annual rate of 139% in the third quarter.  That is NOT a typo.  Meanwhile, the US production of semiconductors and related electronic components rose at a 47% rate in Q3, the fastest pace on record for any quarter going back to at least the early 1970s.

The bottom-line is that AI and data centers are booming, the rest of the economy is not.

Consumption: Auto sales declined at a 5.2% annual rate in Q3.  However, “real” (inflation-adjusted) retail sales excluding autos rose at a 3.5% rate and real service spending – which makes up the lion’s share of consumer spending – appears up at close to a 4.0% pace.  Combined, this brings our estimate of real consumer spending to a 3.7% rate, adding 2.5 points to the real GDP growth rate (3.7 times the consumption share of GDP, which is 68%, equals 2.5).

Business Investment: We estimate a 12.3% growth rate for business fixed investment with widespread gains in equipment, commercial construction, and intellectual property.  If we are right, with the exception of the immediate bounce back in 2020 after the COVID shutdown-induced collapse in the first half of that year, that would be the fastest growth for business fixed investment for any quarter since 2011.  A 12.3% growth rate would add 1.8 points to real GDP growth.  (12.3 times the 15% business investment share of GDP equals 1.8).

Home Building: Residential construction looks to have grown at a respectable 5.5% annual rate in the third quarter, but don’t get optimistic, yet.  Housing still faces headwinds in the form of local regulations, a lack of workers, higher mortgage rates, and a strict immigration policy that makes more units available for rent.  In the meantime, look for a 0.2 point add for Q3.  (5.5 times the 4% residential construction share of GDP equals 0.2).

Government: We estimate that government purchases rose at a 1.7% annual rate in Q3 after being roughly unchanged in Q2, which should add 0.3 points to the GDP growth rate (1.7 times the 17% government purchase share of GDP equals 0.3).

Trade: The trade deficit soared in the third quarter due to a surge in goods imports and a modest decline in goods exports, although this forecast may change when a preliminary trade report is released Wednesday morning.  For now, we’re projecting net exports will reduce the Q3 real GDP growth rate by 3.2 percentage points.  That’s huge and is partially due to the AI/data center buildout, which is driving a surge in imports of tech equipment, including semiconductors from Taiwan. 

Inventories: We estimate that businesses added to inventories in the third quarter after drawdowns in each of the previous five quarters. Going back to at least the late 1940s we have never experienced such a long drawdown in inventories without a recession.   However, some of it is due to the unusual nature of the Trump tariffs, which allowed for an inventory surge in early 2025 to front-run those tariffs.  The growth in inventories in Q3 should add 1.7 points to real GDP growth.

Add it all up, and we get a 3.3% annual real GDP growth rate for the third quarter, which masks extremely rapid growth in AI/data centers and slow growth elsewhere.

Brian S. Wesbury – Chief Economist

Robert Stein, CFA – Deputy Chief Economist

Click here for a PDF version

This week’s edition of “Three on Thursday” focuses on the S&P 500 Index’s (the “Index”) performance over the first three quarters of 2026. So far this year through Q3, the Index has hit all-time highs 27 times, with a maximum drawdown (peak to trough) of 9.1%.  For a deeper insight into what has shaped the market over the first three quarters, click the link below.

Click here to view the full report

Implications: New home building continues to disappoint, with starts falling to a 1.221 million annual rate, well below even the most pessimistic forecast from any Economics Group tracked by Bloomberg and the second lowest pace since the initial onset of COVID.  The 5.2% overall drop in September was due to single-family and multi-unit structures, which fell 4.6% and 6.7%, respectively.  Looking at the big picture, home construction has been on a downward trend since peaking a month after the Federal Reserve began the previous tightening cycle back in March 2022.  One key issue for homebuilders is affordability, which has taken a turn for the worse following the conflict with Iran, as surging energy costs have pushed overall inflation temporarily higher, prompting the Federal Reserve to begin increasing its short-term interest rate target for the first time since 2023. This has resulted in a reversal of 30-year mortgage rates, which have moved roughly 110 basis points higher since February and now sit around 7.2%, more than double the levels that prevailed through much of 2021. Another issue is immigration enforcement, which has reduced the number of new homes the US needs to keep pace with population growth, while at the same time making it tougher for builders to find and replace workers.  Meanwhile, some of the tariffs have boosted the cost of construction and restrictive local building regulations make it tougher to build.  To combat these headwinds, earlier this year homebuilders had been focused on completing projects, but it looks like that activity has dried up with home completions coming in at a 1.192 million annual pace in September, close to the lowest level in more than seven years.  Given these developments it is no surprise to see the NAHB index (a measure of homebuilding sentiment) falling to 29 in October from 32 in September, the lowest level since 2012, where a reading below 50 signals that a greater number of builders view conditions as poor versus good (now the 30th consecutive month that has been the case.)  Even though higher interest rates do nothing to change the energy supply constraints that are driving consumer prices higher, homebuilding is bearing much of the brunt of the Fed’s efforts to fight inflation as one of the most interest rate sensitive sectors in the economy.

Click here for a PDF version

Milton Freidman taught us all about monetary policy…but for some reason academics and the Federal Reserve have decided to ignore him.

There appear to be two reasons for this.  First, he was a “classical liberal.”  What today we would call a conservative.  He believed in small government and freedom.  Clearly there are not many academics left who believe in that.

Second, he argued it would be better to replace the humans at the Fed with a computer program to grow the money supply at a steady pace.  Obviously, the humans at the Fed have a serious problem with this.  So, ignore him.

Friedman told us exactly why we had the inflation of the 1970s…too much money printing.  The same thing happened during COVID…the money supply soared and so did inflation.  Now, we live in the aftermath of that failed experiment and the Fed is having a very hard time deciding what to do to bring it back under control.

Every day, analysts by the boatload debate whether we should move interest rates up or down.  And it seems like every Fed member who speaks only gets a headline if they say they will vote to hike or cut, when, and how many times.

This was important historically, because under the monetary policy regime that existed prior to 2008 the money supply and interest rates were directly linked.  The Fed operated in a “scarce reserve” system.  If the Fed said it wanted interest rates lower then what it really meant was that it was going to increase the amount of reserves in the banking system.  If it wanted the federal funds rate to rise, then it squeezed the amount of reserves in the system.

Back then, banks borrowed and lent reserves to each other.  Banks all had Federal Funds Trading Desks; there was an actual market for reserves.  In that regime, there were two ways to decide whether monetary policy was tight or loose.  One, just look at money growth.  If it was growing faster than the economy, then inflation would likely pick up…the classic “too much money chasing too few goods” of the Friedman era.

But the second was to compare interest rates to nominal GDP.  Nominal GDP is the average growth rate of all the sectors in the economy.  It includes “real” growth and “inflation.”  If you are lending money, you should be compensated for the loss in the value of your money due to inflation plus a real return for the risk of lending or the opportunity cost of not making a different investment.  So, nominal GDP is a good target for interest rates.

Economists have come up with alternatives to this rule.  One was invented by John Taylor.  The Taylor Rule wasn’t exactly targeting nominal GDP, but it was close.  All of these methods attempt to use fundamental macroeconomic values to determine a real rate plus inflation that would keep interest rates in a “neutral” position.  Not too tight or too loose.

All of these rules show exactly why the policies of the 1960s and 1970s caused inflation but the policies of Paul Volcker and Alan Greenspan in the 1980s and 1990s did not.

In the 1960s and 1970s interest rates were held well below nominal GDP.  The way the Fed was able to hold rates this low is by adding money to the banking system.  So, low interest rates and excess money growth were the same.

Volcker and Greenspan held interest rates close to nominal GDP and that meant the money supply was growing more in-line with the economy.  This is why inflation was out of control in the 1970s but came back down and stayed down in the 1980s and 1990s.

All of this changed in 2008, when Ben Bernanke changed the Fed’s policy regime to one of “abundant reserves.”  Essentially, through Quantitative Easing, the Fed flooded the banking system with reserves.  Reserves were no longer scarce and banks no longer traded them.  All those Federal Funds Trading desks are gone.  What this means is that there is no longer a market for reserves.  This de-linked interest rates and money supply growth; they are both determined independently.  The Fed just sets short-term interest rates where it wants them.  We have called this price fixing because there is no longer any market input.  If the Fed wants to hold rates at zero, it just does it.

Why did the Fed do this?  Partly because of the theoretical work done by Michael Woodford, the most recent recipient of the Nobel Prize in Economics.  Many academics and members of the press are praising his work.  We do not.

To Woodford, monetary policy isn’t just about monetary policy itself but the public’s expectations of monetary policy in the future.  So, for example, he thought that in the aftermath of the Financial Crisis the Fed should commit to keeping rates low for an extended period, a policy it repeated during COVID with horribly inflationary results.  In turn, Quantitative Easing – the Fed buying hundreds of billions and ultimately trillions in Treasury debt – was a tool to implement forward guidance.

But a pre-commitment to holding rates low and adding more money to help the economy grow through a crisis also signals that the Fed doesn’t care as much about the actual long-term path of the economy.  This means it’s committing to potentially bad policy.  If the Fed were omniscient about the future path of the economy, the economy wouldn’t have run into certain problems in the first place.  Forward guidance discounts the possibility of something unexpected happening. 

Some still believe forward guidance helped the Fed deal with the Financial Crisis, but the stock market kept falling well after short-term interest rates bottomed and the Fed started Quantitative Easing.  Instead, the real catalyst for recovery was ending overly strict mark-to-market accounting policies, which then gave financial firms some breathing room.  Until this is widely accepted, the Fed will continue to make mistakes.

So, to go back to our opening comments, Milton Friedman wanted to take human foibles and political pressure out of monetary policy.  Bernanke, Yellen, Powell, and even Greenspan near the end of his tenure, were swayed by Woodford (and others) that the Fed could manage public expectations about future actions and actually control the economy completely.

But making a mistake in monetary policy is one of the worst things any country can do.  A stable value for the purchasing power of a nation’s currency is as important to an economy as a constitution.  And, clearly, this recent experiment in monetary policy has been a complete failure.

The sooner we come to recognize this, the better.  Unfortunately, it’s hard to take power away from bureaucratic institutions once they have taken it.

To give a Nobel Prize to Woodford, a godfather of abundant reserves, puts an academic star of approval on this policy.  Friedman also won a Nobel Prize for his work, but because it criticized the human management of monetary policy, the Fed wants to ignore it.  Powell actually said the relationship between M2 and inflation “no longer holds” and is something we “have to unlearn.”

Clearly, seeing M2 grow 42% during COVID followed by 9% inflation would suggest he was wrong.  Nonetheless, many have alternative explanations for our current bout of inflation and still cling to the belief that money doesn’t matter.

Clearly, we are critical of the current monetary policy regime.  Our new Fed Chair, Kevin Warsh, is also critical of this regime.  Nonetheless, any wholesale change in the system does not seem to be in the cards, at least anytime soon.  So, for the time being, we are living in a regime where the Fed must set interest rates where it thinks they should be.  How does it do that?

By keeping the funds rate close to nominal GDP.  During the 1980s and 1990s this worked well, but then at the end of Greenspan’s tenure he made a mistake and held rates too low in the aftermath of the dot.com crash.  This helped fuel the housing bubble which led to the Great Financial Crisis, where the Fed kept rates at rock-bottom near-zero levels for several years.  Yes, inflation stayed tame in spite of prolonged low rates, but that was largely because of banks having to rebuild capital after the mark-to-market meltdown and then having to meet higher capital standards.

Then, in 2021-22, nominal GDP growth far outpaced short-term rates again, and the results were far different than the tame inflation after the Financial Crisis.  This time inflation soared, with the old relationship from the 1960s-70s coming back with a vengeance.

Clearly, if the Fed actually believes interest rates are more important than the money supply, it has some explaining to do.  Why did seven years of near zero percent interest rates not lead to inflation, but two years of near zero rates did?

All of this brings us back to some simple truths.  If the Fed is going to be setting interest rates where it wants, it should set them according to macroeconomic forces.  And if interest rates are not linked to money growth, then it should not allow money growth to surge or collapse.

For interest rates, they should target nominal GDP.  In the past two years (through the second quarter), nominal GDP is up at a 5.5% annual rate, which signals that a federal funds target rate of 3.875% is probably too low.  It is unlikely the Fed will lift them next week because of the looming election, but we do expect a rate hike later this year.  December seems most likely.  For the money supply a 5.5% to 6.0% growth rate seems appropriate.

The big question for rates is how fast will nominal GDP grow.  The AI investment boom, and the wars in Iran and Ukraine, are impacting all the data.  Nominal GDP might be re-accelerating.  Retail sales (+0.8%), industrial production (+0.7%), and producer prices (1.3%) all grew rapidly and beat consensus expectations.   Consumer prices also went up fast (+0.6%), although as expected.

We get a GDP report next week and, for now, we expect real GDP growth at about a 3.3% annual rate and nominal GDP growth at about a 5.3% pace, which is in line with the past two years.  The robust economic reports for September suggest momentum into the fourth quarter is picking up. 

In the meantime, new Fed Chairman Kevin Warsh is trying to get the Fed out of the game it’s been in for the past roughly twenty years of providing too much “forward guidance” about the future path of monetary policy, so much that he thinks the Fed has sometimes found itself in the position of implementing a monetary policy because the financial markets came to expect it rather than that policy being the most appropriate for the Fed to meet its goals.

Unfortunately, the policy of “forward guidance” still has a great deal of support in some quarters.  Especially, given the recent Nobel Prize recipient.  All of this makes the future more uncertain than it should be.

The bottom-line is that with the new policy regime the Fed has lost touch with the economy.  It has eradicated a lot of market feedback by eliminating banks from trading federal funds.  It has separated interest rates from money supply growth.  It has attempted to manage public perceptions but has now lost touch with them.  Friedman was right, human management of monetary policy increases the risks of making long-term mistakes.

Brian S. Wesbury – Chief Economist

Robert Stein, CFA – Deputy Chief Economist

Click here for a PDF version

Implications: Industrial production came in much higher than expected in September as broad-based gains were coupled with a surge in utility output to push up activity for a sixth consecutive month. Looking at the details, the volatile and tariff-exposed auto sector fell 3.0% in September.  However, manufacturing excluding autos (which we think of as a “core” version of industrial production) rose 0.6%.  As a result, overall manufacturing was still up a solid 0.3%.  Notably, manufacturing ex-autos has been on a consistent upward trend since the start of 2025, a sharp reversal from the downward trend from 2018-2025. Production in high-tech equipment, which has been a consistent source of strength recently due to data center construction and the reshoring of semiconductor production, jumped 1.5% in September. Activity in this sector is up 17.6% in the past year, the fastest of any major category. The manufacturing of business equipment remained unchanged in September but is up 6.8% in the past year, easily outpacing the 1.1% gain in overall manufacturing and signaling a broader reindustrialization.  Looking outside the manufacturing sector, mining activity eked out a gain of 0.2% in September.  The increase was driven by gains in oil and gas extraction which more than offset declines in drilling activity and the extraction of other minerals.  Finally, utilities output posted a gain of 3.5% in September. This series has been on an upward trend since 2023, following nearly twenty years of stagnation, as power hungry data centers have boosted demand for US power generation. On the inflation front, import prices rose 1.1% in September and export prices rose 2.2%.  In the past year, import prices are up 8.1%, while export prices have risen 11.6%.

Click here for a PDF version

Investor borrowing to fund stock purchases, known as margin debt, has surged over the past three years and currently sits at historic levels. In this week’s “Three on Thursday,” we take a closer look at the $1.5 trillion in margin debt outstanding as of August 2026. To find out more, click the link below.

Click here to view the full report

Implications: Flashy headline, modest details.  Retail sales rose 0.8% in September, easily beating consensus expectations, while sales in previous months were revised higher.  However, most of the surprise can be attributed to a 6.4% jump in gas stations sales, reflecting higher prices at the pump, not a boost to economic activity.  Still, growth was broad-based with eight of thirteen major sales categories rising in September.  We like to follow “core” sales, which strip out the volatile categories for autos, building materials, and gas stations and is important for estimating GDP.  This measure rose 0.3% in September and was up at a solid 5.7% annual rate in the third quarter versus the Q2 average.  This is consistent with our view that real GDP grew at about a 3.0 - 3.5% annual rate in Q3. Within the core grouping, sales at nonstore retailers have been a major bright spot in recent months but slowed in September, rising 0.3% after a 2.0% jump in August. Consumers pulled back on spending in September at stores for sporting goods & hobbies (-0.5%), miscellaneous stores (-3.0%), and general merchandise stores (-0.1%). Meanwhile, sales at restaurant and bars – the only glimpse we get at services in this report – increased 1.1% in September following an upwardly revised +1.4% gain in August.  Sales in this category have shown solid growth since February and stand 7.0% higher than a year ago, outpacing the 6.3% increase for overall sales. It’s important to remember that none of these figures are adjusted for inflation.  Nominal retail sales have risen 6.3% in the past year, but factoring in inflation, “real” inflation-adjusted sales are up 2.5% in the past twelve months. While higher prices have inflated the headline figures, US consumers continue to spend despite a challenging economic backdrop.  In other recent news this morning, initial jobless claims last week ticked up by 1,000 to 199,000; continuing claims declined 4,000 to 1.710 million.  These figures suggest continued job growth.

Click here for a PDF version

Implications: Producer prices surged in September, rising across the board. The Producer Price Index rose 1.3%, the largest monthly gain since 2022, and above even the most aggressive forecast by any Economics Group on Bloomberg. While more than 40% of the headline increase was due to energy (which rose 10.8% for the month), final demand prices excluding energy also rose 0.7% in September, as the second-order effects of higher energy prices work their way through the economy.  Looking at the details, goods prices rose 2.6% in September, the largest monthly increase since these data were first collected in 2009. Half of this increase can be traced to gasoline prices, which jumped 20.3% in September, and are up 64.3% in the past year.  Food prices (+1.2%), and other energy-related categories like diesel, jet fuel, and crude petroleum also rose. On the services side, prices rose 0.8%, led by a 3.9% increase in margins for machinery and vehicle wholesaling.  Further back in the supply chain, price increases for intermediate goods were broad-based, with elevated energy prices continuing to burden businesses at every stage of production. Prices for processed intermediate goods rose 2.5% in September, driven by a 21.0% increase in diesel fuel. Meanwhile, prices for unprocessed intermediate goods also rose 8.4%, led by a 28.3% jump in crude petroleum. Both unprocessed and processed intermediate goods are up by the most in any twelve-month period since 2022. Excluding food and energy, “core” producer prices rose 0.7% in September, while the twelve-month change rose to 5.0%, well above the 3.0% increase for the twelve months ending September 2025. Overall producer prices are up 6.3% in the past year, over double the change from the twelve months ending September 2025. In other recent news, the Empire State Index, which measures manufacturing sentiment in the New York region, rose to 25.5 in October from 7.6 in September. Also, the Philadelphia Fed Manufacturing Index, a measure of factory sentiment in that region, fell to 20.9 in October from 37.8 in September.

Click here for a PDF version