Equities

Mean Reversion Is Thematic for 2026

As we head into 2026, one of the stock market’s primary concerns from a few months ago has faded from focus: the “cockroaches” concept. It was not long ago that Jamie Dimon circulated his view that the bad-debt debacles at First Brands (from alleged fraud) and Tricolor (a shoddy subprime auto loan book) could be like household pests: if you see one or two, look behind the drywall and there will be many more.

The vision of cockroaches infesting private credit was, for a spell, a primary sentiment driver and a key reason for the market’s mini comeuppance in November. However, as things stand now, there hasn’t been much in the way of new bad debt revelations. Maybe that will change, but for now, latching onto the cockroaches for a general bear thesis has lost its muscle.

Another concern in some circles is the backup in Japanese government bond yields. Beyond the effect on that nation’s economy, a reasonable theory has been that Japan could drag the U.S. Treasury market along with it. That would, as the theory goes, adversely affect the bull case for global equities.

The 10-Year JGB yield has started to flirt with 2%, up from around 1% in early 2025. Thus far, the U.S. bond market has not reacted, and equities have not shown concern. So long as the repricing of the JGB market remains orderly, we are also okay with it, but this is an area we will be watching in 2026.

In equity asset allocation, we favor small caps, along with the Quality and Value factors. In the case of Quality, we have some performance charts that are touching extremes last seen in 1999 and 2000.

For example, inside small caps specifically, the junky Russell 2000 Index just witnessed its boldest nine-month outperformance bout over the better-quality S&P 600 Index since 1999. On a reversion-to-the-mean basis, it makes complete sense to round up investment committees to scrutinize exposure to unprofitable firms and low-profitability holdings in small-cap mandates.

For an idea of how bold such a mean reversion could be across the market, we calculate that the S&P 500 outperformed the S&P 500 Quality Index by 11 percentage points in the six months off the April 8 broad market lows. The last time this happened was in 1999. Not much was rectified on this front in November or December, either.

In other words, you don’t have to look far to find froth. Thirteen of the 15 largest S&P 500 Tech sector components can’t get past a <5x sales screen, while eight of them are above 10x sales. Taking it further, check for >30x sales, and we still count four of those 15. It’s a market that is priced with what appears to be absolute certainty that Silicon Valley’s AI promises will necessarily come true.

As things stand, the Magnificent 7 is putting its money where its mouth is, aggressively. The group goosed its collective capital expenditures by 61% in the year through the most recent earnings season. At the same time, net cash flow from operating activities only grew 22%, resulting in a collective decline in Mag 7 free cash flow. FCF also declined in 2022, a year that witnessed the Tech sector sell off to the tune of 27.6%, while the S&P 500 fell 18.1%.

We understand quite well why free cash flow growth is turning negative: Everyone in the AI game is attempting to set their own foundation for the future, when the promises of the technology will change the world. But here and now, the Mag 7’s price-to-free cash flow ratio is 52.6. That figure will seemingly rise in 2026 if capital expenditures once again handily outpace operating cash flow growth, which is essentially the universal consensus.

Because of such firms’ overbearing size in an index such as the S&P 1500, the total U.S. stock market now has 60% of its market capitalization in firms that trade for more than 35x forward earnings. Generally, when looking for the other 40% of firms, the investor will move down the sector list to the down-and-outs. The bottom five YTD performers are Energy, Consumer Discretionary, Materials, Real Estate and Consumer Staples.

Consider an outperformance case for two of those five: Consumer Discretionary and Consumer Staples. These two tend to show up in size when we run buyback screens and, in many cases, Value screens. We are okay with the U.S. consumer in 2026 because we believe the general ill health of the middle class is exaggerated.

For example, the National Retail Federation anticipates holiday shopping season sales will grow 3.7%–4.2% this year. Our “real world” inflation measures, on a year-over-year basis, have been clocking in below 2% this winter. If our inflation metrics are accurate, and we believe they are, that means the typical person will buy more presents for loved ones this year than they did last year. That defies the storyline of the busted consumer.

Additionally, the news focuses on specific pain points, such as electricity, coffee and red meat, each of which has come to be outrageously priced. But we are a little stunned that the collapse in unleaded gasoline gathers little attention as a household budget offset.

With the national average gas price under $3 per gallon, down from $5 in mid-2022, we calculate that the combination of price, modern-day fuel economy and average hourly earnings above $31 makes the current situation very close to the lowest household auto fuel burden on record.

If we have it right, the “troubled, but not busted” consumer theme could provide relief for both Staples and Discretionary sectors relative to the S&P 500 in 2026.

In developed markets, we want to check for correlations to the market’s primary risk, the U.S. Tech sector. Since that group skews toward mega-cap growth, it is little surprise that mandates with a Value bias and exposure to smaller firms tend to show up as having the lowest correlations to U.S. Tech. Interestingly, in developed markets, we are finding that screens for both quality and dividends are clocking in with lower correlations to it than our screens for dividends alone.

In emerging markets, we find that deep value concepts have had a historical tendency to snap into outperformance mode after the calendar years in which they woefully underperformed beta.

For example, we saw emerging markets’ value and deep value struggle particularly badly relative to cap-weighted indexes in 2015, 2017 and 2020. The years subsequent to those (2016, 2018 and 2021) each witnessed bold snapback outperformance by the group relative to cap-weighted baskets. Like those three episodes, 2025 was a particularly bad year for the group; contrarians take note.

To engage emerging deep value is to be content with heavy exposure to Financials, Materials and Energy at the expense of Tech, and Chinese tech specifically. Given the question mark we are presenting on the AI capex build-out, such positioning seems reasonable, as hunt-for-yield concepts inside emerging markets have considerably lagged U.S. large-cap growth since the October 2022 lows.