Since 2023, we’ve lived in a micro-driven market, which is to say if you decompose a given stock’s performance into company-specific (or industry-specific) factors and macro developments, a majority of the return will be attributable to the former.

That’s a low correlation, high dispersion environment, and indeed the dispersion trade can be self-fulfilling to the extent selling vol on the index to finance long single-stock vol positions perpetuates surface-level calm and drives underlying churn.

I’ve argued previously that the crowded nature of that trade’s almost self-evidently perilous to the extent it creates the conditions for an exaggerated response to a macro shock, as index vol shorts are unwound into an indiscriminate downtrade (i.e., into an across-the-board de-risking episode).

As the figure reminds you, we’ve experienced a couple of mini-“Corr 1” moments over the last eight or so months. One of those episodes (the August growth scare/JPY carry unwind) was pretty ugly.

Anyway, if you ask Goldman’s David Kostin, 2025 will be another year defined by a micro-driven market. He cited three factors, the first of which are expectations for decent growth (macro calm). The second factor’s AI, the “development and adoption” of which “should create differentiation across stocks,” according to Kostin, who went on to cite “elevated policy uncertainty” as another driver of dispersion.

That latter point — about policy uncertainty — calls to mind a familiar paradox in this discussion: You might suggest that intuitively, rampant ambiguity around big-picture issues should drive higher correlation, but remember that policy shifts tend to create winners and losers. As Kostin put it, “the healthy return dispersion environment in 2024 was marked by declining macroeconomic uncertainty and the growing importance of thematic debates around AI and the US election” (emphasis mine).

This topic can be a bid maddening for the uninitiated. Consider the figure below, from Kostin’s note.

So, looking back two decades, the only years which exhibited higher dispersion than 2024 were 2020 and 2009 which, to the point above, will undoubtedly seem counterintuitive to some readers: If the GFC and COVID don’t count as “macro-driven” markets, what do?

But, as Kostin patiently reminded clients, dispersion’s a function of correlation and single-stock volatility. Although correlation naturally spikes during recessionary trades, so does single-stock vol. Depending on the severity of the shock, the sheer magnitude of the vol spike can eclipse the rise in correlation for the purposes of calculating dispersion, hence the very high reading for 2020.

Last year, 6%ile realized correlation (i.e., historically low) was conducive to dispersion and thereby to stock-picking. Implied correlations (illustrated by the first chart above), “suggest little change in this low correlation environment over the next six months,” Kostin wrote.

The figures give you some additional context for the current environment, both in terms of the above-mentioned disparity between single-stock and index vol (on the left) and correlation (on the right).

Assumptions about a low-correlation, high-dispersion environment in 2025 depend, pretty much by definition, on the idea that Donald Trump’s trade war and the associated economic policy uncertainty, manifest as a “winners / losers” market, not as a macroeconomic shock. I wrote about that in the latest Weekly.

The figure below shows Goldman’s attempt to quantify the share of typical stock returns attributable to micro versus macro factors.

The upshot: Nearly three-quarters of the typical S&P 500 stock’s return is currently (or recently) explainable by company- or sector-specific developments, leaving just a quarter to macro factors.

Looking ahead, Goldman said “debates over trade, tax, fiscal and other policies represent potential catalysts for additional return dispersion.”

That’s certainly true. And while I personally doubt much will ultimately come of Trump’s trade crusade, anytime you set about shaking the foundations (and Trump’s doing that on multiple fronts), there’s a risk you collapse the whole edifice. If the building implodes, everything goes in the same direction: Down.

Finally, I’d be remiss not to note that the dawn of what some view as a Russia-style oligarchy in America is highly conducive to dispersion. From Election Day to December 17, shares of Tesla rose 98%. I’ll just leave that there.