The federal government closed out fiscal year 2025 at the end of September, and in this week’s “Three on Thursday,” we dive into the current state of federal finances, offering a historical perspective as well. With annual deficits now in the trillions and interest payments on government debt at record levels, it’s clear that significant changes are needed. For further insight, click the link below.

Click here to view the report

Implications:  Existing home sales posted a modest gain in September, though activity continues to trudge along at a disappointing pace. The current pace of 4.060 million remains near the lowest since the aftermath of the Great Financial Crisis, and well below the roughly 5.250 million annual pace that existed pre-COVID (let alone the 6.500 million pace during COVID).  That said, affordability has been improving in several notable ways. First, 30-year mortgage rates have been trending lower since May and now sit around 6.3%.  Notably, this is the lowest rate since 2023 and with the Federal Reserve restarting rate cuts recently, buyers have reason for further optimism. Meanwhile, the median price of an existing home is up just 2.1% versus a year ago. It looks like the inventory of existing homes rising 14.0% in the past year has helped put a lid on prices as more options become available for buyers.  That has helped push up the months’ supply of homes (how long it would take to sell existing inventory at the current very slow sales pace) to 4.6 in September, a considerable improvement versus the past few years, and approaching the benchmark of 5.0 that the National Association of Realtors uses to denote a normal market. One last positive to note is that aggregate wage growth (hourly earnings plus hours worked) has begun to consistently outpace median home price gains over the past year for the first time since 2023, which improves affordability. That said, some challenges remain. Many existing homeowners remain reluctant to sell due to a “mortgage lock-in” phenomenon, after buying or refinancing at much lower rates before 2022.  This remains an impediment to activity by limiting future existing sales (and inventories).  Existing home sales also face significant competition from new homes, where in many cases developers are buying down mortgage rates to compete and move inventory (when interest rates are higher, firms, including homebuilders, forego more potential earnings by holding onto inventories). Despite these cross currents, underlying fundamentals have improved recently, which should contribute to a modest rebound in sales. In other recent housing news, the NAHB Index (a measure of homebuilder sentiment) rose unexpectedly to 37 in October from 32 in September.  Keep in mind a reading below 50 signals a greater number of builders view conditions as poor versus good, now the eighteenth consecutive month that has been the case. On the manufacturing front, the Empire State Index – a measure of factory sentiment in the New York region – rebounded sharply to +10.7 in October from -8.7 in September. Meanwhile, its counterpart the Philadelphia Fed Index fell to -22.2 in October from -12.3 in September.

Click here for a PDF verion

As we recently argued, investors don’t need to worry about the federal government shutdown showdown causing a recession.  Before the current shutdown, the federal government had been shut for eighty days in the prior thirty years, with none of those days during a recession.  In other words, the US economy has been less likely to be in recession when the government is shut than when it’s open.

Equity investors appear to understand that, with the S&P 500 down only 0.4% as of the Friday close compared to immediately prior to the shutdown.  Stocks are still overvalued as far as our Capitalized Profits Model is concerned.

In the meantime, we got some relatively good news on the federal budget late last week, with the deficit clocking in at $1.775 trillion for Fiscal Year 2025 (which ended September 30), $41 billion smaller than the deficit the prior year and $34 billion smaller than what the Congressional Budget Office projected the deficit would be only two weeks ago.

And yet the overall budget picture remains unsustainable, which ultimately is what this shutdown is hopefully all about: will we reign in federal spending enough to get our fiscal house in order?

In the meantime, unfortunately, the steady stream of government reports on the economy has dried up because of the shutdown.  As professional economists, we’re not a fan of this particular feature of shutdowns.  It’s not like spending on data crunchers to generate weekly or monthly economic reports is an important political issue.  Nevertheless, we understand that if the rest of the government is shut, the data specialists are going to be furloughed, as well. 

But this week there will be a major exception, with the Bureau of Labor Statistics having called back enough staff to generate the monthly report on the Consumer Price Index, albeit later than usual.  The CPI report was considered “special” in that the Social Security Administration, which is providing essential services, is required by law to calculate and announce the annual Cost-of-Living Adjustment (COLA) for beneficiaries.  

We think consumer price inflation probably ran a little on the hot side in September, although that should diminish in October due to the recent drop in oil prices.  In addition, we will get a report on Existing Home Sales this week, because that is generated by the National Association of Realtors, which is a private group, not run by the government.  We’re estimating a mild pick up in sales.  

Going back a few weeks to when the shutdown started, we do have some economic reports for September, just not as many as usual, and they showed some cracks in the economy.  The worst news was the ADP Employment report, which showed a decline of 32,000 in private-sector payrolls for the month.  Remember, this comes on the heels of news from a month ago that annual revisions to the official payroll report will show job growth of only about 70,000 per month in year ending March 2025 versus a prior estimate of about 145,000.    

Meanwhile, the two ISM reports were tepid, with the manufacturing index at 49.1, lingering below the key 50 level, signaling contraction, and the ISM Services index falling to exactly that key 50 level, the second lowest level this year. Adding to the tepid figures were regional reports from Federal Reserve Banks, showing growth in manufacturing in New York and shrinking manufacturing in Philadelphia.

The best report for the month was on auto sales, as automakers reported that sales remained steady in September at a 16.6 million annual rate, up 3.4% from a year ago.  However, these sales may have been boosted by the expiration of federal tax credits electric vehicles purchased before September 30.    

We also have all the usual economic reports for July and August, and so we still feel comfortable with our estimate that Real GDP grew at about a 3.5% annual rate in the third quarter.  But we highly doubt that continues into the fourth quarter, where our very early guess is that Real GDP grows at a much slower 1.5% rate, instead.  

As far as the shutdown goes, it looks like it’s not on the verge of ending just yet.  The longest shutdown in Trump’s first term lasted five weeks.  And with the stakes even higher and his Administration’s ambition on spending cuts even more dramatic, this one looks ready to last well into November.

Brian S. Wesbury – Chief Economist

Robert Stein, CFA – Deputy Chief Economist

Click here for a PDF version

In this week’s “Three on Thursday,” we take a closer look at construction spending in the United States. Each month, the U.S. Census Bureau releases its Construction Spending report, which tracks how much is being put in place across residential, nonresidential, and public projects. Construction is more than bricks and mortar—it’s a leading indicator of broader investment trends, tied to everything from consumer housing demand to government infrastructure outlays to AI-driven expansion. To learn more, click the link below.

Click here to view the report

Over the past two and one-half decades the federal government has buried taxpayers under a mountain of debt, now approaching $38 trillion.

During this time the key problem has been spending, not a lack of tax revenue.  Over the past 25 years, taxes have remained relatively stable as a share of GDP, while spending continued to rise.  We estimate that spending was 23.2% of GDP in the year ending September 30 (Fiscal Year 2025) versus 17.7% in 2000.  In other words, the reason we have a debt problem is because we have a spending problem.

That’s the bad news, and it means policymakers still have a very long way to go before we can claim our fiscal house is in order.  But just because the overall budget picture remains bleak doesn’t mean we shouldn’t recognize improvements when they happen, and there is some progress.

The Congressional Budget Office recently estimated that budget deficit for the year that that ended on September 30 (FY 2025) came in at $1.809 trillion, slightly smaller than the $1.817 trillion of FY 2024.  Although that’s only an improvement of $8 billion in dollar terms, there are reasons for at least some modest hope on the deficit.

First, we project that nominal GDP grew 4.8% in FY 2025, which means that even if the deficit remained roughly the same in dollar terms, it declined relative to GDP.  After clocking in at 6.3% of GDP in FY 2024, it looks like it was 6.0% in FY 2025.  That may seem like only a minor improvement, just 0.3 percentage points of GDP, but it’s a shift from the expansion of the deficit in the prior two years.  It’s movement in the right direction.    

Second, the decline in the deficit this past year would have been larger were it not for a calendar-related issue.  In particular, two years ago (in 2023) October 1st fell on a Sunday and so $72 billion in federal payments (spending) that normally would have been made that day were instead made on Friday September 29, 2023.  That shift changed the budget math because September 29 was still in FY 2023.  This artificially held down official spending in FY 2024 making the deficit appear smaller than it really was.  In addition, due to natural disasters in 2023, some taxpayers were allowed to postpone tax payments.  This boosted revenue in FY 2024 by about $70 billion.  In other words, the actual reduction in the deficit for FY 2025 was really closer to $150 billion, not $8 billion.

Third, although total federal spending was up $228 billion last year (after adjusting for the timing issue discussed above), the gain was due to factors largely outside the control of the new President and Congress, at least in the short term.  Spending on Social Security, Medicare, and Medicaid were up a combined $245 billion in FY 2025.  Meanwhile, interest payments on the national debt were up $80 billion.  In addition, the Environmental Protection Agency, under the direction of the former President, shoveled out $20 billion in extra payments in the waning days of his Administration, which was included in FY 2025.  By contrast, other categories of spending – outside those entitlements, interest, and unusual EPA payments – declined $117 billion.  When is the last time you remember that happening?

Fourth, due to weather-related disasters, some revenue that normally would have occurred in FY 2025 is being postponed into FY 2026.  In other words, while the CBO says federal receipts were up 6% versus last year, the gain would have been more like 8% in the absence of disaster-related deadline changes.

Again, none of this means we are anywhere close to having the debt or deficit situation under control.  A deficit of 6.0% of GDP is still unsustainable.  Moreover, the Supreme Court may throw the whole budget situation for a loop if it declares recent tariffs illegal.  Expect a ruling early next year.  

The current shutdown and political brawl over emergency additions to Medicaid spending during COVID is extremely important for maintaining this progress.  Government spending had been on a one-way escalator, with each so-called crisis (2008 and then COVID) permanently lifting spending to a larger and larger share of GDP.

Somehow the angst over budget deficits has disappeared.  We remember when deficits were all anyone talked about.  Remember ABC’s Sam Donaldson screaming at President Reagan about the size of the deficit?

Lately, the angst and screaming has been about cutting spending, not reducing the deficit.  The good news is that progress is being made. And it’s important to recognize that budget progress when it happens even if more needs to be done.  So, here’s a thumbs up for the progress that has been made in FY 2025, with fingers crossed that this progress is just the beginning.

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 performance over the first three quarters of 2025. In the third quarter alone, the S&P 500 Index achieved a total return of 8.1%, bringing the year to date gains up to 14.8%. So far this year through Q3, the S&P 500 Index has hit all-time highs 28 times, with a maximum drawdown of 18.9%. For more insights, click the link below.

Click here to view the report

No one, that we know of, is saying Artificial Intelligence (AI) isn’t an amazing new technology that will have an important impact on life, investing, and the economy.

What we worry about is that investors have gotten ahead of themselves.  It’s happened before.  On January 3, 1925 the New York Times published a piece on radio stocks that said, “It is considered doubtful if there ever has been an industry in the history of this country which has had such phenomenal growth.”  Those stocks eventually crashed.

We would add the South Sea Bubble of 1720 promised to revolutionize global trade, railroads at the turn of the 20th century, the oil boom of 1920/21, the telecom and computer boom of the late 1990s, and 3D printing more recently.  All of these technologies, inventions and industries have transformed the US and the world.  They have all lifted living standards; they all created immense wealth for savvy investors and entrepreneurs.

But they also generated speculative pricing in markets that eventually led to pain for investors who tried trading them for profit.  There are many reasons for this, but the most important is that the real beneficiary of new developments in technology, when the dust settles, is the consumer, the end user.  Yes, wealth is created by well-run companies (that survive the shake-out) in the industry, but the real benefits are spread far and wide.

A perfect example of this is the combination of national fiber optic networks with more powerful, and decentralized computers run on the WinTel platform (the Windows Operating System + Intel Chips).  It may be hard to remember, but the world completely changed in the late 1980s and 1990s as things moved from large back offices with lots of paper, to electronically powered networks.  From mail to email.

In 1999 Ford launched AutoXchange through a joint venture with Oracle (serendipitous, eh?) to digitize its supply chain.  Ford estimated it could cut annual purchasing costs by 10%, saving it up to $8 billion per year.  The stock market, which was already overvalued according to our capitalized profits model, loved this news.  The dot.com boom of the late 1990s was fueled by developments like this for years.

No one should ever argue that digitizing business activity wasn’t more productive and less costly.  The question is, did it translate directly into permanently higher profits and stock prices for corporations?  History and economics say “NO.”

Why? Because Ford was only one auto company.  All its competitors – Daimler, Toyota, Chrysler, GM, Volkswagen – did the same thing.  And as costs came down, competition made sure those cost savings were passed onto consumers.  No, car prices did not fall, but better cars were built, and prices reflected the reduced production costs.

The story of the swashbuckling telecom industry during the late 1990s – WorldCom, MCI, Global Crossing, Lucent, Qwest, and others – is amazing.  Boom to bust.  The Palm Pilot was an amazing device, basically the precursor for the iPhone, but 3Com went bust, too.  Amazing new technology, lots of speculative activity. 

Few would want to live in a world today that wasn’t transformed by the inventions of the past 200 years.  Clearly, living standards are higher.  But as Amara’s Law (from futurist Roy Amara) says, “We tend to overestimate the effect of technology in the short run and underestimate the effect in the long run.”

It's not clear that AI is boosting revenues or profits right now or is just cannibalizing existing business models.  For example, Google search already operated on algorithms based on payments, patterns of search and whatever other filters it chose to use, like political correctness.  So, what does AI do that’s different other than do it with fewer people writing code?  Maybe that boosts profitability for Google by lowering costs, but even that is not clear.

AI will clearly have a positive impact on the speed of research in medicine, and many other areas of life.  And, like all things in Internet time, it will move faster than we think.  However, this was also true of radio and information, the fiber boom and the speed of communications, the car industry and the proliferation of driving.  But each of those went through speculative phases and then a shake-out as markets and competition revealed winners and losers.

We are in the speculative phase now.  Pricing in a future that is very bright, but that is just not here yet.  Every single model of stock market valuation that we know shows the market exceeding its historical averages – price-earnings, price-sales, market cap to GDP (without including highly valuable private companies), and our capitalized profits model.  And if the economy is slowing, as the jobs market suggests, these valuations are even more extreme.

At some point, things will return to normal…when, or how, is anyone’s guess.  In the meantime pay attention to value, and the basics.  Read our Chief Investment Officer Dave McGarel’s “Market Minute” for a more specific take on all these issues.  We build a new future every day and have been doing so for a long time…history doesn’t repeat, it rhymes.   The challenge for investors isn’t believing in the future—it’s remembering that even the brightest future can be overpriced today.

Brian S. Wesbury – Chief Economist

Robert Stein, CFA – Deputy Chief Economist

Click here for a PDF version

Implications:  The US service sector ran in place in September as the ISM Services Index lagged expectations and dropped to 50.0, signaling the same level of activity as last month.  Despite continued weakness in its counterpart survey on the manufacturing sector, the service sector remains far more resilient, with activity expanding in ten out of the last twelve months. Looking at the details, growth was divided among service categories: ten out of eighteen major industries reported growth versus seven that reported contraction.  The drop in the headline index came from the new orders index normalizing from 56.0 (the highest level in ten months) to 50.4.  Meanwhile, the business activity moved into contraction territory (albeit, barely) from its five-month high of 55.0 to 49.9, the first time below 50 since May 2020. Survey comments indicate a fractured growth picture, as some industries struggle with stagnant demand under the pressure of high interest rates and tariffs, while others experience very strong demand from the AI and cloud infrastructure buildout.  This bumpy path has kept service companies defensive in their hiring efforts.  Employment continued to contract in the service sector in September (but at slower pace compared to last month), with the category rising from 46.5 to 47.2.  That makes six out of the last seven months where the employment index has been below 50, signaling contraction.  Service companies – once hamstrung with difficulty finding qualified labor – have begun reducing their headcounts, with more industries (eight) reporting lower employment in September than higher (six).  Finally, the highest reading of any category was once again the prices index, which appears to have stabilized just below 70. While that is near the highest level since late 2022, it is still far from the worst we saw during the COVID supply-chain disruptions, when the index reached the low 80s.  Though inflation pressures remain, the M2 measure of the money supply has grown very slowly for three years, which means we are likely to see lower inflation and growth in the year ahead.  As for the economy, it’s important to remember that Purchasing Manager’s surveys like the ISM Services index and its counterpart on the manufacturing sector often capture sentiment mixed in with actual activity.  Uncertainty from trade policy has been weighing on sentiment, but could be alleviated as more trade agreements are finalized.  However, monetary policy has been tight enough to reduce inflation toward the Federal Reserve’s 2.0% target and is probably still modestly tight today.  And a monetary policy tight enough to reduce inflation may also be tight enough to slow the US service sector.

Click here for a PDF version

The government officially shut down yesterday as Congress failed to pass—and the President failed to sign—appropriations bills (or a stopgap “continuing resolution”) to fund government operations for the start of the new fiscal year on October 1st. Headlines often frame shutdowns as economic calamities in the making, but history suggests otherwise.  For more insights, click the link below.

Click here to view the report

Implications:  Manufacturing activity remained soft in September but did not contract as rapidly as in August, with the index narrowly beating expectations and rising to 49.1. This makes seven consecutive months that the ISM Manufacturing index has been below 50, continuing a pattern that stretched all of 2023 and 2024.  While many believed the downturn was over when the index briefly rose above 50 in January and February, the subsequent seven months of weak readings warrant caution for the industry.  Looking at the details, just five of the eighteen major manufacturing categories reported growth in September, while more than double (eleven) reported contraction.   Unfortunately, the new orders index – which broke into expansion territory last month for the first time since January – contracted in September, falling to 48.9.  Order books were weak heading into this year; now, survey comments blame tariff uncertainty for the weakness, as many customer orders have been placed on pause until stability returns.  In addition, survey comments also report that companies are largely shouldering the added costs where tariffs have been put in place, as customers are unwilling to pay the higher prices and simply choosing not to buy.  This has caused companies to look for ways to reduce overhead, most notably through their hiring efforts.  Though the employment index rose to 45.3, it remains firmly in contraction territory, with only one major manufacturing category (Nonmetallic Mineral Products) reporting an increase in employment in September versus fourteen reporting a decrease.  Despite the stagnation in manufacturing, inflation pressures remain.  The prices index declined to 61.9 – high by historical standards – although below the recent peak of 69.7 in June, and well below the levels during the post-COVID inflation surge.  In other news this morning, ADP’s measure of private payrolls declined 32,000 in September versus a consensus expected increase of 51,000, the third decline in the past four months. However, there are often large gaps between the initial ADP report and the official report from the Labor Department, so we are estimating Friday’s official report will show a nonfarm payroll gain of 80,000 with the unemployment rate remaining at 4.3%.  On the housing front, pending home sales, which are contracts on existing homes, increased 4.0% in August, suggesting a gain in existing home sales in September.  Declining prices might be part of the reason for any pickup in sales, with both the FHFA and Case-Shiller indexes down 0.1% in July.  That’s the fourth straight monthly decline for the FHFA and fifth straight for the Case-Shiller.  Compared to a year ago, FHFA prices are up 2.3% while Case-Shiller prices are up 1.7%, both less than overall inflation.

Click here for PDF version