市场版：十字路口的科技交易

Recorded: July 6, 2026Published: July 7, 2026Host: Scott T. Chronert, Head of U.S. Equity Strategy, CitiTranscript: Opening Teaser: (00:00) Research @ Citi Markets Edition. Scott Chronert (00:03) Hi, I'm Scott Chronert, Head of U.S. Equity Strategy at Citi Research. Welcome toResearch @ Citi Markets Edition, covering various topics at work within the U.S. equitymarkets. Today, I'm going to discuss our positioning views for Q3 with a focus on tech andthe broadening trade.In our last podcast, we talked about a constructive outlook for theS&P 500 going into the back half of the year. This was premised on whatwe characterized as earnings-growth exceptionalism, which is essentially referring tothe AI influence dynamic that has caused S&P level earnings to grow at a significantlyfaster rate thus far this year relative to most other periods in time, other than comingout of recessions. That earnings tailwind continues in place as we go into the second half and into the Q2reporting period, which will consume a lot of the Q3 timeframe. However, this is where itgets interesting. In our views going into the quarter, we've actually gotten a bit morecautionary on the tech trade and have focused more on the broadening opportunity aswe go into Q3. Let me explain. If you look at the S&P 500 and some of its derivative price-action components more recently, it's quite fascinating. As an example, the S&P 500 isup roughly 9% year to date, but probably 15% for Q2. Surprisingly, though, off 1% to2% from its early June highs. Compare that to the Nasdaq, which is up 16% year to date, but a whopping 27% duringQ2. And compare that to the semiconductor sector within the S&P 500, which is up 33%year to date and about that much during the Q2 timeframe. Interestingly, that semicomponent has also sold off 10% to 11% since early June. So my point here being thatwe've had a very strong influence on this AI trade and related earnings tailwind comingspecifically from the semiconductor component. However, as you begin to look at the backdrop for the semiconductor driver within the AIcomplex, what you run into is a little bit of what we'll call a game of chicken. Thestronger the capex spend by the hyperscalers, the stronger the setup is for thesemiconductor component of that capex influence. You do, however, run into a point where we have to be wary that you can see highermemory costs and higher costs for other inputs begin to become part of that capexplaybook. And on the other side of the semiconductor strength, it begins to set up forconcerns about the return on investment for the hyperscalers and other AIcapex spenders. This is the game of chicken that we're talking about. Essentially, the better the earnings are for the semiconductor component, at some level,the read-through by the marketplace has put more pressure on the hyperscalers, again,to justify ROI. Fascinatingly, for the strength that I mentioned in the S&P 500, thesemiconductor component … the NASDAQ 100, if you think about that old Mag 7 cohort(and we recently have been talking more about the Mag 8), they're roughly flat year todate. So what's happened for this part of the index, which comprises literally 40% or so of theindex, we've seen idiosyncratic behavior among these players. Some up, some down,some sideways. When I say some down, as an example, the software sector within theS&P is actually down 17% year to date. Again, this bifurcation in performance is something that we've been talking about forsome time, but it's also something that we think is going to continue to be with us fromthis dynamic where, OK, are we entering a phase where these mega-cap growers can'tall perform equally, against a backdrop that still favors the AI spending growth that isexpected to persist after the balance of this decade? So what we're doing specifically as we go into the Q3 timeframe is, as Imentioned, we've stepped back in terms of our positioning on the tech componentspecifically, which is an area that we've been quite constructive on, particularly viasemiconductors for the better part of two to three years. And we've allowed this to digestas we go into the Q2 reporting period. We think we'll get much more spin from the companies, whether they're the semimanufacturers, the hyperscalers, some of the other related players in this cohort ofcompanies, that's going to speak to, OK, I get it, the spending strength is there. It's goodfor fundamentals now. But how long can it really persist? And that questioning is whatwe think will keep some edge on the tech trade as we go into the summer months. Now, let's pivot and talk about what else is going on in the market. In the past, we'vesaid that, hey, the AI component of the S&P 500 is probably approximately 50% of theaggregate index. This means that everything else is the other 50%. That everything else is going to align more closely with more traditional … let's callit fundamental, but more importantly,macrometrics. What we've got g