(Spoiler alert: scary chart below.)

I have been working to add a quantitative measure of risk exposure to my stock market overview material. There are numerous ways to quantify risk, and I am trying to come up with some sort of composite indicator that includes the full spectrum of “make you want to puke” factors: excessive market valuation, financial instability, geopolitical disruption, political instability, Equity Risk Premium (stock rate of return versus risk free interest return), Things That Go Bump in the Night.

There are plenty of risk measures to choose from and most (except financial instability) are at dangerous levels right now. Using most any of these measures, if market prices were to get back to “normal” (whatever that is) then the stock market could quickly plop down 30% to 50%. If a couple of things went bad at the same time, things could quickly get even worse. Stomach shaky yet?

Anyway, as part of this quest I revisited the data set that Robert Shiller used for his 2000 blockbuster book Irrational Exuberance. He had incredible timing; the book was published just as the horrific Dot Com market crash began. Since publishing, he and Yale University have maintained and updated the data set here . It is a wonderful resource as it has numbers going back all the way to 1871.

Stock pricing is supposedly rational, so it would be reasonable to figure that the general ratio of stock price to company earnings (P/E) would be somewhat constant, or at least it would sort of follow long-term interest rates. It makes sense that it should appear even more smooth with averaging. Shiller’s CAPE P/E measure has a 10-year earnings average to filter out the noise. But even with that decade-long ‘smoother,’ the P/E path is anything BUT smooth.

Here is what I see in the chart; P/E, rather than being steady, follows a kind of rough pattern every few decades. After a severe stock market crash, P/E stabilizes for a while, but then begins an exponential, increasingly rapid, climb. Finally, when that rate of climb ‘turns vertical’, a severe market crash happens. Since 1871 I don’t see any cases of a fast rising P/E climb that smoothly leveled off. Maybe the 1960’s qualify as a P/E leveling, but that amounted in a lost decade for stocks. Yay! Pick your poison.

Right now, P/E is incredibly high and spiking vertically. As the Great Crash of 1929 hit, the CAPE ratio was nowhere near as high as today. The only other time CAPE has climbed this high was also in a spike, December, 1999. The great Dot Com stock market crash began four months later (March, 2000) and hit bottom in 2002 with a total fall of 49%.

I have no Idea if we are on the verge of a similar collapse, or when it might happen. But , I do not like the current P/E spike. It might it might be time to proceed with caution and pay close attention.

Both my long-running and more current data-intensive stock market models are positive, but unenthusiastic about the U.S. stock market for the coming 6 months. Last month the models expected the stock market to twitch around, but not rise much. And that is what happened. Sometimes these models actually are right.

Goling forward, the models expect flat behavior this month as well, followed by small gains over the next few months. For reasons I have not figured out, the current models see market jitters next month or in April.

Viewed from 30,000 feet, the market should be zooming up. Government deficit spending remains high, continuing to juice up GDP. The Trump tariffs have had a slight reduction in the federal deficit, but continuation of the Trump tax cuts and the addition of more give-aways have more than made up the difference. The AI infrastructure build out (software and hardware) continues as a historic spending stmulant.

The stock market, however, seems to be discounting the spending bonanza. My graph of the long term GDP-based trend of the S&P 500 shows it to be about 7% above trend which is not a big deal. The equal-weight version of the S&P 500 is running an anemic 1% below trend. If the overall economy was really healthy we would have seen a true boom/bubble. This looks more like a hospital patient being kept alive with heroic medical care, and held in a coma by heavy drugs.

Consumer spending continues to climb — but, most spending growth has come from the very wealthy, not the broader population. Unemployment is not terribly high at 4.4%, but the graph has a concerning upward slope and most recent employment has come in the low-paid health care industry. Corporate profits remain acceptable. Money supply is increasing at a moderate pace. Last year’s surge of the Magnificent Seven appears to be waning. The high relative growth of the S&P 500 versus it’s equal-weight version appears to have slowed-down and may be topping.

Could be worse, I guess. Another queasy market month would be fine, I guess.

The chart shows my forecasts (generated by my current data-intensive models) for the major US stock market indexes for the coming half year. I’d call that a “confusing” picture without a clear trend other than, perhaps “motion sickness”. The only of these indexes that has done well for the past quarter is ^DJT. (No, not the stock that lost 2/3 of its value this year and went from supposedly being a media Giant, to a crypto Mega Player and now to a Fusion Energy Leader. Good luck with that. This ^DJT is the Dow Jones Transportation Average).

This has been a year during which I have truly appreciated the fact that the stock market forecasts I generate are based on actual data rather than on my emotions.

Because my emotions have been frazzled — mainly because of all the twists and turns of the Trump government. Trump and his very determined team have kept up the tremendous viewer ratings he always craves — no one can stop focusing on every move. And suspense has been amplified; mainly because of all the policy shifts and reversals, but also partially because of the data gaps created through the Federal shut down, partially via lies and obfuscation, and partially because of the audacity of so many moves — or feints.

The overall economic picture, however, remains driven by a few main forces. The US and most of the rest of the world is still facing massive government annual deficits that keep overstimulating the economy in the near term (causing inflation), but simultaneously deficits are steadily worsening the overall strength and security of the financial system. (Fortunately, we have not yet seen a big spike up in long term interest rates — the sign of financial panic.)

We are also benefitting from massive investment in building out AI and it’s data infrastructure, but despite all the spending we are seeing unemployment creep up (4.6%) in a pattern that is similar to the lead-up to a typical recession. This market boom is showing its age.

There has been real economic damage caused by the Trump administration, but it has been largely hidden from view. The stock market did indeed keep climbing. But the real value of everything you own has tanked — the value of the US Dollar fell off a -9% cliff.

The Trump tariffs are having the effect of somewhat reducing the deficit, but already the Trump administration is talking about distributing more “helicopter” money in $1,000 and $2,000 checks. The net effect of all of this would appear to be that the economy will continue to be in an aging boom, that could be instantly slammed by a financial panic. I personally am feeling the same sort of “motion sickness” that my forecast charts are showing for the next few months.

My long term GDP-based trend lines are more comforting. They see the S&P 500 index as only mildly (6%) overvalued while the equal-weight version of the same stocks is just 3% above trend. Not too scary. Most recession leading indicators are still quiet.

My personal bet is that January could get ugly — I expect that now a new tax year has begun, a reasonable number of investors holding Magnificent 7 stocks will want to reduce their vulnerability. My older models see subdued gains for the first half year of roughly 4% ,and the probability of at least breaking even as around 80%, which is slightly below normal. The new models are a bit less confident but none-the-less positive.

So, I wish both of my loyal readers a joyful and prosperous New Year. And some luck.

My 2-decade-old market models and my data-intensive AI models basically agree that the U.S. stock market will probably perform fairly well over the next half year with a gain of roughly 6%. The newer models have a few extra worries for the next month or so than the old timers; even so, they are positive/flat. This could change suddenly, of course, if the Government shutdown continues very long — your guess on that is probably as good as mine. Also, I have no idea what the immediate market response would be if the Supreme Court shuts down the legally questionable parts of the Trump tariffs. Again, your guess is as good as mine and my models are silent on the subject.

Overall, the U.S. stock market continues to be split: the AI-supercharged Magnificent 7 stocks are still surging and the established market keeps performing a bit below normal expectations. Right now, the S&P 500 is about 8% above my long term trend line. Last round, it turned south when it reached 10% above trend. (The trend line has been rather reliable for the past few years.)

There is a real split between short-term stock market expectation indicators and long term expectations. Short-term indicators like current stock market index trends and several leading indicators (like the Sahm Rule) have been rather steadily positive. It is as if stock market ’emotion’ has been inoculated in some way so that it no longer gets jarred by erratic moves by the White House. Professional economists, however, are less positive looking forward. In the most simple sense, on one hand they see bad news coming, and on the other hand they see bad news coming. The actual impacts of the Trump tariffs are critical. If tariffs do, indeed, produce a ton of revenue, that means a net reduction in government overspending/stimulus and so a drain on the economy. The other option is that the tariffs don’t actually reduce the deficit significantly — heading the U.S. toward a debt crisis.

By Halloween we should see if stock market investors will continue to keep “whistling past the graveyard”. Expect fewer chocolates this year if the tariffs remain intact.

My stock market forecasting tools are still blessed by not being able to read the daily newspaper. My hometown Washington, DC is essentially under martial law with gun-toting National Guard troops patrolling the streets reacting to a trumped-up ’emergency’ that does not exist, but my stock market performance models (except for September) remain reasonably positive. The forecasting models, with an excellent long term track record, indicate that a wide swath of U.S. economic data is basically benign and so the U.S. stock market will probably perform fairly well over the next 6 months. The September exception is that the S&P 500 (^GSPC on the chart) looks like it has gotten ahead of itself. Both my 18 year-old models with about a dozen economic variables and my current data-intensive models reach similar conclusions.

The macroeconomic data remains positive. Unemployment is still just 4.2%; money supply is growing at a 4% annual rate; inflation, only a bit higher than the Federal Reserve wants, is 2.7%; and corporate profits as of last quarter are growing mildly. Leading economic indicators generally are modestly positive with very low probabilities of a near-term recession.

The stock market division between highflyers like Nvidia versus the rest of the world decreased last month. As the graphs below show, the S&P 500 (NVDA accounts for 8% of the index) is now just 5% above its long term GDP-based trend line, and the equal weight version of the same stocks (SPXEW) is now exactly at the long term trend line.

I have a hunch that September will show stock market indecision, but volatility could easily shoot up toward the end of the month and into October. First, the question of the basic legality of the Trump tariffs will wind its way through the courts, eventually through the Supreme Court. Second, there is a significant chance of a U.S. government shutdown at the end of September since the FY2026 budget will probably need some measure of Democrat approval. And third, there is increasing concern over Trump’s health. (The helicopter flight path from the White House to Walter Reed Medical Center goes directly over my house.) I find a bit of comfort with the thought that the stock market, having weathered so much back and forth uncertainty in the Trump second term’s first months, may have been inoculated from instant major responses to potential new calamities. Wishful thinking no doubt.