When it comes to the violent rotation out of technology that started in June, early July did not bring any relief. In fact, if we must rate the violence of this rotation on the last trading day before the long holiday weekend, July 2 has to be right at the top of the list. On that day, the ended flat (up just 0.01) while flipping violently in a 113-point range, but the was down sharply (-479.92 or -1.62%), while the and made all-time highs, up 0.76% and 1.14% respectively.
There have been violent rotations before, but typically they are one-day or one-week affairs in an uptrend. This rotation has been going on for a month, and the only other time a rotation was this bad was when the “dot com” bubble burst in March 2000. Then, too, the technology sector declined a lot, over a period of several weeks, while the Dow was up and the S&P 500 flat. While the similarities in this extreme friction are obvious, there is one problem: I do not believe we have reached a climactic top in technology stocks.
Spending on the AI boom continues, and what appears to be happening is we have gotten overheated in some technology names. Because of the viciousness of the rally in April-May, we could experience a sharper correction – much sharper than what we saw in early June – but that would be normal.
The present situation is very different from 2000, when technology companies were borrowing a lot of money to finance what in many cases were money-losing businesses. The miles of fiber optic cables laid then were dark for years, but as internet traffic grew, they were “lit” sometimes 10 to 15years after they were put in the ground. The internet boom that culminated in 2000 laid the groundwork for the boom in the U.S. economy and surging profitability in the past 15years, resulting in a stock market at recent all-time highs.
Strategically, we are still in a bull market. Tactically, we are ripe for a bigger correction.
This dramatic rotation can be seen in the S&P 500 Equal Weight YTD performance of 12.2% vs. an ETF of the “Mag 7” companies down 1.2%, with a big part of the divergence coming in June. If we are to see a deeper correction in the stock market, led by technology, it probably needs to happen between now and when second-quarter earnings begin to be reported near the end of July, basically in the next three weeks. That would cause the overheated part of the technology sector to no longer be so overheated, and for stocks to react better to what are expected to be very strong earnings reports – estimated to be up around 25%.
S&P 500 earnings are dramatically lopsided towards the technology sector. It makes more sense to see a sell-off in the before the chip makers report earnings, which should be spectacular.
The philosophical question is this: Why would companies delivering explosive sales and earnings growth have a correction? Because a tremendous amount of short-term sector gains were built up in the last threemonths and, the desire for investors to book gains or simply protect profits can produce a cascading effect.
We can never know what the trigger may be ahead of time. In August 2024, it was the yen carry trade; last year it was the trade war; this year it could be the Iran war restarting, or something not yet evident to us.
After the friction in the stock market in June and early July, the simple fact remains that it would not take much to have stocks see significant selling into strength tip over more dramatically. Because money has been flowing out of the Mag7 and into the broad market and semiconductor sector, the NDX Index is performing dramatically worse than the SOX or SPX indexes, as the Mag 7 is a huge component there.
If we manage to trade below the NDX June lows in the next three weeks (the red line in the chart above), I think we will experience an intermediate-term correction before earnings kick back into high gear in August.
Any normal correction does not change my optimism based on very high EPS growth and generally better performance of the U.S. economy, mostly resulting from the multiple policy changes of the Trump administration in 2025.
