The Great Pyramid of Giza took about 26 years to build and was completed in 2560 BC, give or take a few years. It was 481 feet tall, the base was 756 feet, and was constructed with 2.3 million blocks of limestone with some of them weighing 70 tons. The Great Pyramid of Giza is the oldest of the Seven Wonders in the World. Nikola Tesla theorized that the pyramids were not just tombs but were advanced devices for transmitting or generating wireless energy, a concept he explored in his own work on wireless power transmission. He believed their advanced design was a result of fundamental mathematical laws and ratios, and he was also fascinated by their numerology, particularly the numbers 3,6,9. “If you only knew the magnificence of the 3, 6, 9, then you would have a key to the universe.” There are thousands of pyramids on earth and the one quality they all share is that none of them were built with the tip of the pyramid as the base.
Wall Street doesn’t adhere to the ancient wisdom that managed to build pyramids and instead embraces ‘themes’ that occasionally flips the stock market upside down, so that the tip of Wall Street’s pyramid carries more weight than the base. It’s been problematic for the overall stock market whenever the Top 10% of all stocks comprise more than 65% of the market’s total capitalization (78% now). That was the case in 1929, 1937, 1959, 1972, and 2000. The S&P 500 lost -86.8% after 1929, fell -55.2% after 1937, declined -49.9% after Nifty-Fifty top in 1973, and -50.5% after the Dot.Com Bubble. The lone exception was in 1959. Although the S&P 500 shed -14.2%, it avoided a wipeout because the overall market strengthened enough to offset the selling in the stocks in the Top 10%. The spike in 1973 was the result of the Nifty-Fifty craze when Wall Street decided portfolios only needed the 50 stocks it described as the Nifty-Fifty one decision stocks. “You only needed to buy.”
The upside down overweight extends to the S&P 500 now. JP Morgan recently identified 41 AI related stocks that now account for 47% of the S&P 500’s total capitalization, even though they only represent 8% of the 500 stocks. In March 2023 those 41 stocks comprised 26% of the S&P 500, so their weighting has almost doubled. The Magnificent Seven (Apple, Amazon, Alphabet (formerly Google), Meta, Microsoft, Nvidia, and Tesla) now comprise 37% of the S&P 500 capitalization, with Nvidia representing 7.5%. Nvidia is the tip of the current upside down pyramid and on Wednesday November 19 Nvidia reports it earnings, so the stock market has the potential to react strongly, especially if Nvidia’s numbers and outlook disappoint.
In the November Macro Tides I discussed how Nvidia’s GPU’s and chips are depreciated by the companies that use them in data centers and how a shorter depreciation schedule can artificially lift earnings.
“Depreciation is the accounting method of allocating the cost of a tangible asset over its useful life, reflecting the decrease in its value due to wear and tear or obsolescence. In the Gold AI Rush, companies are spending a lot of money on GPU’s (General Processing Units) and will depreciate the cost over time. The amortization period for GPU chips varies significantly, typically ranging from 1 to 6 years. Earnings are reduced by the cost of amortization, so earnings can be boosted if GPU chips are amortized over a longer time frame. Datacenter GPUs may only last from one to three years, depending on their utilization rate, which is normally between 60% to 70%. At that rate a GPU will last about 2 years. Hyper scalers like Google, Microsoft, and Amazon use GPUs for a wider range of workloads and can amortize them over longer periods, typically 3 to 4 years.”
“Nvidia has been doubling the computing power of its GPU chips every 10 months, which makes older less powerful chips worth less. Usage may make a GPU last less than 3 years, but technological obsolescence can cause the value of GPU chips to fall faster. The speed at which GPU chips will depreciate in value compared to an amortization schedule creates an accounting problem that can lower earnings. In order for a data center to remain functional at a high level (and attract paying customers), the owners will need to spend more money on new chips before the old chips have produced an investment return. If the demand for data centers slows at some point in the future, some owners may consider walking away from the data center just as commercial landlords give the keys back to the back once vacancy and cash flow weakens and results in a decline in the valuation of an office building. This risk may not manifest for another 2-4 years but it’s real and underappreciated today. Compare the AI data center amortization treadmill to the lifespan of fiber optic cables that last at least 40 years.”
At the time of the November Macro Tides, the topic of depreciation seemed a bit arcane but I included it since it would become important at some point. In the last two weeks an increasing amount of attention has focused on depreciation. Since 2020 all of the big buyers of Nvidia GPU’s have extended their depreciation times from 3 years to 5 or 6 years. The extension increases the risk that earnings could be negatively impacted as companies are forced to write off the cost of older GPU’s faster in coming years, which will be a drag on future earnings. If earnings growth slows, the P/E ratio afforded to all the companies in Artificial Intelligence will fall and exacerbate a decline in their stock valuations. This process is what has developed after the Top 10% of stocks became the tip of the upside down investment pyramid.
In the November 3 Weekly Technical Review, I noted that Meta’s stock sold off hard after reporting better than expected earnings. “Revenue rose 26% to $51.2 billion and earnings were $7.25 per share versus Wall Street’s estimate of $6.69, which is a big beat. Meta intends to spend much more on AI in 2026. Meta will spend $70 - $72 billion in 2025 (above previous estimates) and could spend more than $100 billion in 2026. Meta’s stock didn’t sell off because it was spending less on AI, it sold off because there is concern it might be spending too much. For the first time we’re seeing the question of return on investment enter the conversation. The overspending concern will spread to more of the AI related stocks at some point in 2026.” The reaction to Meta’s earnings announcement represented the first canary in the Artificial Intelligence’s coal mine.
Warren Buffett is retiring at the end of 2025, but the cash position of Berkshire Hathaway’s has never been larger in absolute terms and as a percentage of the portfolio. Warren Buffett is one of the all-time great investors and the accelerated increase in the allocation to cash ($382 billion) suggests Berkshire Hathaway is expecting a big unwind in the stock market when reality hits the valuation fan. Warren Buffet’s outlook sharply contrasts with Households who have allocated more to the stock market than ever before. The current allocation (52%) is higher than in 2007, 2000, and way higher than in 1972.
The extreme level of concentration in the Top 10% of stocks is a long term warning that future returns will be less, once the current upside down pyramid episode runs its course. History informs us that it will take a period of years i.e. 10 to 15 years. But institutional investors won’t sell because the market is too concentrated or that valuations are too high. There will have to be a good reason or reasons to convince investors that buying and holding is not warranted. It will take time and mounting losses to persuade most institutions and the public that things are different this time, and buying the dip only leads to more losses.
I’m not convinced that investors are ready to give up on the Artificial Intelligence theme just yet. As noted in the November 10 WTR, “A modest decline over a handful of days won’t erase the belief that AI is revolutionary and the vision of more gains to come.” As I noted last week, “After the negative breadth signals in November 2021, the S&P 500 managed to eek out a new high on January 4, 2022. Despite the negative breadth signals on October 28 and October 29 for the Nasdaq, the market can hold up as long as the buying in the 41 related AI stocks is maintained.” My guess is that will change sometime in 2026, and I’ll do my best to identify the timing.
Stocks
The S&P 500 closed below its 50 day simple average for the first time in 140 days on November 17. This is the third longest stretch since 1970 which shows just how unusual the last 6 months plus has been. This doesn’t mean the market is about to go to hell in a hand basket, but it does represent a change in character, no matter how unusual the recent trend has been. The rarity of -3% pullbacks and absence of even a -5% correction underscores how complacent and overly confident most investors have become. On November 14 the S&P 500 was down 80 points in the first 10 minutes of trading, and then rallied 128 points in less than 3 hours as the ‘buy the dippers’ scooped in.
In the November 3 WTR I reviewed what it would take to confirm a high in the S&P 500 technically. “A close below 6551 would be significant since the S&P 500 has made higher highs and higher lows since the low in April, and 6551 is the last trading low.” In the November 10 WTR I updated the key level for the S&P 500. “The intra-day low on November 7 was 6631, which was also above the 6612 level mentioned last week. The key level going forward is 6631 as that is now the most recent trading low.”
After dropping to an intra-day low of 6646 on November 14 and 6639 on November 17, the S&P 500 has so far held above the important level of 6631. This is a two hour chart so it is short term focused, but the price lows are more intermediate term. The top panel measures short term selling pressure as reflected by NYSE Ticks. As the S&P 500 was hitting its low on November 17, Ticks became as oversold as they did when the S&P 500 fell to 6551. The bottom panel is the RSI on the S&P 500 and it dipped below 30 on November 17, as it did when the S&P 500 traded down to 6551, 6631, and today’s low of 6639. This suggests the S&P 500 can be expected to bounce to near 6780 or modestly higher. Obviously, you can throw this outlook out the window if Nvidia lays an egg after the close on November 19.
For the first time since the April low the S&P 500 failed to achieve a higher high, which further elevates the importance of 6631. A drop below 6631 would establish a lower high and lower low, and increase the odds the S&P 500 is in a downtrend. The S&P 500’s RSI is near 40 which is where it was as the bottom at 6551 was formed. The initial decline from 6920 was 289 points and an equal drop would target 6589. This raises the possibility of a quick spike to near 6589 followed by a reversal that lifts the S&P 500 back above 6631.
In the last two WTR’s I’ve recounted the serious negative technical signals that were generated as the S&P 500 and Nasdaq 100 reached new highs on October 28 and October 29. The weakness indicated by those signals suggests the market is vulnerable to a sharper decline. I think that’s more likely in 2026, but Nvidia could unleash a wave of selling pressure which would expose the underlying technical weakness. If the S&P 500 drops below 6631 and doesn’t rebound quickly and reclaim 6631, a drop to 6350 – 6400 is possible.
I thought the doubts about AI and the level of spending were likely to persist and cause an increase in volatility as selling pressure picked up and the dip buyers kept corrections shallow. On November 14 and November 17 that was certainly the case. Seasonality is favorable going into year end and most individual investors will be reluctant to sell winners before year end so capital gains taxes can be postponed until April 2027. Institutional investors will also be reluctant sellers since they want to show they are exposed to the AI trade and its great promise. After year-end these positive dynamics will diminish. These are reasons why the market can hold up into early 2026 despite near term volatility.
Gold
After Gold topped at 4381 it quickly plunged $494 before bottoming at 3887. Gold was expected to retrace a portion of the $494 decline, but not expected to rally to a new high. The 50% retracement was 4134 and the 61.8% retracement was 4192. On November 13 Gold rallied to 4244 and quickly shed more than $200 in the next two trading days. I thought the rally from 3887 was another opportunity to sell into strength as Gold reached the 50% and 61.8% retracement levels.
Gold is in the process of correcting the Wave 3 rally from the September 2022 low of 1616 to the October 2025 high of 4381. This correction is Wave 4 from the 2015 low of 1046 and will take months to complete. As discussed previously Gold is expected to ultimately decline to near 3325.
“Gold traded up to 4381 on October 20, so Wave 3 was $2765 (4381 – 1616 September 2022 Wave 2 low). A 23.6% retracement is $653 and the 38.2% is $1056 from 4381 the Wave 3 high. This generates targets of 3728 if Gold retraces 23.6% and 3325 if Gold retraces 38.2% of Wave 3. I think Gold will trade below 3750 in coming months. Chart wise the range between 3300 and 3500 is a target since the recent breakout occurred at 3403 and the April high was 3496.”
In the shorter term, the initial decline of 494 suggests Gold could trade down to 3750, which is just above the 23.6% target of 3728 (4244 – 494 = 3750). If Gold trades down to 3750 it will create a short term trading opportunity for another retracement rally.
Gold Stocks
GDX was expected to rally after it plunged -16.95 (-19.9% in 7 trading days) from the high of 85.08. The 50% retracement was 76.60 and the 61.8% retracement was 78.60. “If Gold pushes a bit higher (above its 61.8% retracement target of 4192), GDX might be able to reach 78.60.” I noted that the rebound was another opportunity to lighten up as a rally above 85.08 was not expected. Gold did exceed the 61.8% retracement (4244 vs. 4192 target) and GDX popped to 79.97 on October 12 before falling by more than -6.2% in the next 2 trading days.
The Wave 4 correction in Gold is expected to last for months and GDX can be expected to follow Gold. GDX rallied from 40.26 to 85.08 after bottoming in April. If GDX retraces 38.2% of the 44.82 rally, GDX would fall to 67.96, which it effectively did after dropping to 68.13 on October 27. The swiftness and depth of the drop from the high suggests GDX has the potential to retrace 50% of the rally. That provides a target of 62.67. As Gold works through its Wave 4 correction, there will be sharp rallies that will provide trading opportunities in GDX. If Gold drops to 3750 GDX is expected to decline below 68.13 before the next trading opportunity develops.
Dollar
As noted previously, “The Dollar needs to close above 100.25 to provide more confirmation that the low at 96.21 is in place. If the Dollar does close above 100.25, it will need to hold above 97.46 on any pullback.” The Dollar traded up to an intra-day high of 100.36 but has yet to close above 100.25. As noted last week, a modest pullback was likely. If / when the Dollar closes above 100.25, it should hold above 98.03 on any pullback to maintain a positive intermediate outlook. The Dollar is expected to rally to 104 – 106 in coming months and potentially higher i.e. 110. However, to get to those targets it must first close above 100.25.
I think the FOMC won’t vote to lower the Funds rate at the mid December meeting, although it will be a close call. More data will be coming out now that the government shutdown has ended, but it may not show a decisive weakening in the labor market to convince the Hawks to ease.
Treasury yields
The 5 Wave increase from 3.947% for the 10-year Treasury yield is evident on the daily chart, as discussed in the November 10 WTR. I thought the 10-year yield could dip below 4.066%, before the assault on 4.20% began. On November 12 the 10-year yield fell to 4.056% before reversing higher. The key level for the intermediate term is 4.20%. A close above 4.20% will provide more confirmation that the trend in Treasury yields is up, and will likely be followed by a quick move up to 4.351%. The move higher in Treasury yields will be delayed if the 10-year drops below 3.947%.
TLT declined in 5 waves as it dropped from 92.18 to 88.88. TLT was expected to rally above 89.87 since the 50% retracement of the decline from 92.18 to 88.88 was 90.38 and the 61.8% was 90.85. On November 12 TLT traded up to 90.32. TLT’s comparable level to 4.20% on the 10-year is a close below 88.46.
Major Trend Indicator
The Major Trend Indicator is comfortably above the blue horizontal line. As you can see the Bear market declines in 2022 and 2025 didn’t occur until after the MTI fell below the blue horizontal blue line, as well as the Intermediate decline in 2023 during September and October. This suggests the market could hold up for a while despite the terrible breadth as the S&P 500 and Nasdaq recorded new highs on October 28 and October 29. The MTI is saying a decline of more than -7% isn’t likely.
Seasonality is favorable going into year end and most individual investors will be reluctant to sell winners before year end so capital gains taxes can be postponed until April 2027. Institutional investors will also be reluctant sellers since they want to show they are exposed to the AI trade and its great promise. After year-end these positive dynamics will diminish.
