'Risk Control Is Essential': Top Goldman Trader Reflects On The Madness Of March

A choppy week ended with the SPX down ~2% and now below the 200-day moving average (down ~6% MTD), and Brent crude oil now at $112 ($103 last wk, $92 week prior and $72.48 on Feb 27th).

But, as Goldman Sachs global head of custom basket trading, Louis Miller, points out in his latest 'Global reflections': Choppiness creates opportunities, but risk control is essential.

Here's four agenda items he believes are worth calling out amidst this march madness:

1) Our European Theme of Themes Strategy is Performing Well During Turbulent Times…

While all the focus is on US and Asia and we sense an understandable apathy in Europe, one trade that is working well with much less crowding is our EU Theme of Themes basket pair (GSXEMKTP Index, +5.67% MTD). This basket goes long the top EMEA thematic baskets and goes short the bottom thematic baskets based on 1. narrative momentum, 2. narrative reversal, 3. direct coverage momentum and 4. price momentum. Over the past 3 years this market neutral basket of baskets trade has outperformed the European market with much less vol, and could be a nice compliment to a portfolio in an uncertain tape. 

Source: Goldman Sachs

The strategy has worked on both the LONG and SHORT side – GSXEMKTL has o/p vs. market and GSXEMKTS has u/p vs. market

Source: Goldman Sachs

2) The need to be selective on Defensives:

In the US, traditional defensives have disappointed and need to be examined closely. US treasury 2yr yields are up >50bps month-to-date in the same month that US payrolls were negative (even ex-weather adjustments). The healthcare sector has actually underperformed this month and the long leg of quality has not held up well, some of AI immune stocks are starting to look rich and beginning to underperform. On this last point, Faris Mourad explains the market has traded asset-heavy industries as defensively and as AI-immune, which is debatable. See our work on companies that HALO'd too far.

We like going short US Companies that HALO’d too far (GSXUHALT). The basket provides exposure to asset-heavy businesses with low returns and no growth that have rallied too far based solely on their assets and more defensive positioning. We filter for R1K companies in industries with the highest asset intensity ratios and exclude all stocks that are tied to any secular trend (satellites, robotics, quantum, AI, …), focusing on those that rallied significantly year-to-date with unchanged/negative earnings revisions.

GS Research pointed out recently: reflecting the increasing investor desire for “AI insulation,” asset-heavy stocks have outperformed sharply in the last few months. Asset intensity is defined as the ratio of a company’s assets, less cash and intangibles, to revenues. After years of asset-light outperformance, our sector-neutral basket of tangible asset-heavy stocks (GSTHHAIR) has outperformed our basket of asset-light stocks (GSTHLAIR) by ~20 pp since the start of November. GSXUHALT filter for the companies that have rallied in line with GSTHHAIR but have no earnings growth estimates to justify this rally.

GSXUHALT can be paired with thematic trends we like on the long side, following the recent price action that has created the biggest ‘buy the dip’ opportunity in global equities since Liberation Day.

Equities that have “HALO’d too far” have moved in-line with GS research’s High Asset Intensity basket (GSTHHAIR):

Source: Goldman Sachs

GSXUHALT was trading in-line with earnings until late last year, and has diverged aggressively of late:

Source: Goldman Sachs

We asked our research team to pull the PEG ratios for GSXUHALT and our diversified Internet basket (GSTMTINT), and we can clearly see GSXUHALT is now trading expensive relative to growth expectations while GSTMTINT is trading cheap…A few clients are exploring this pair (GSPUHIIT).

Source: Goldman Sachs

3) How to protect from a degrossing episode? 

The market has de-risked but hasn’t degrossed. Having a plan for degrossing risk makes sense, and our team has a lot of fantastic momentum hedges available (recall our barra index GSP1MOMO Index or Sector Factor or EU and APAC momentum offering) and our derivs trading team highlights two limited loss implementations: 1) GSX1BFML 2m 75 90% put spreads cost ~2.64%, 2) GSTMTAIP 2m 75% 90% spread costs ~2.52%. (max loss premium paid) Recall, the work we have done to the platform to provide intraday transparency on options on baskets on Bloomberg for the mid, delta implied vol. We can facilitate real size, and are committed to a good experience on the unwind.

Despite the volatility in the market and index wobbles, Momentum has been resilient with our pure index GSP1MOMO only 2% off the highs (and still up ~12% YTD). The continued geopolitical tensions coupled with extreme crowding in momentum (5y high) and elevated gross positioning in Prime (gross exposure near 5y high) could lead to technical selloff in the factor. Our favorite limited loss implementations. We also have seen clients looking at implementations of Momentum excluding AI related names (GSPUMOXX).

• GSTMTAIP 2m 80 95% put spreads cost ~2.52%

• GSX1BFML 2m 75 90% put spreads cost ~2.64%. (max loss premium paid)

Why isn’t quality working? Quality (GSP1QUAL) has typically worked well as a hedge when SPX is drawing down (see table below). However, the trade has not worked this year as the short leg has outperformed. Meanwhile, Size quietly just had its best 5 day stretch in over a year as market participants rotated into defensive trades and the AI trade has bounced back, benefitting Mag 7 and other larger companies. Size (GSP1SIZE) also has a similar hit rate to quality, outperforming 70% of the time during SPX drawdowns

Quality is based on the combination of profitability, earnings quality, and earnings stability. These scores are based on a combination of historical metrics including margins. In the past, these more stable, higher margin companies demanded a premium, particularly in times of macro uncertainty. However, AI makes this dynamic look different in the forward space. In a world where there is increased competition, high margins may not be sustainable. While not all quality companies are exposed to AI disruption, the overhang of margin compression could translate to a moderation in the P/E premium of the highest vs lowest quality stocks which is well off the lows, implying that there could be more room for these moves to continue.

Factor Performance

Source: Goldman Sachs

Performance of our Quality Index GSP1QUAL

Source: Goldman Sachs

Pricing power that quality stocks have historically had may be deteriorating in a world where AI driven competition may put pressure on margins.

This could hurt quality stocks where P/E premium of the highest vs lowest quality stocks is still well off the lows.

Equity factor valuations relative to history

Source: Goldman Sachs

The underperformance of Quality has been driven by the strong outperformance of the short leg

Source: Goldman Sachs

4) Higher For Longer: Short Japan Inflation Losers (GSXAJIFL)

To position for stagflation, our colleagues in Japan and Europe now have implementations available (recall we flagged GSPUSTAG Index last week). the desk likes going short Japanese companies that are negatively impacted by rising inflation. To gain exposure to this thematic, we have constructed GS Japan Inflation Risk (GSXAJIFL) basket which consists of industries such as utilities, food products, beverages, restaurants, retail etc. that have high exposure to rising input costs but limited ability to pass these on. 

In recent weeks, investors have grown increasingly concerned about a "stagflationary" macro environment due to the ongoing conflict in the Middle East & a consequent surge in commodity prices. Japan, which is heavily reliant on the Middle East for energy, has come under pressure with rising import costs & a weakening yen. (+) 3 Quick Points:

1/ Updating GS Forecasts: GS Japan Economists' Akira Otani & team recently raised Japan's core CPI (excludes fresh food) inflation forecasts, mainly for 2026 (average 2026 core CPI to 2.0% yoy, a 0.2 pp upward revision from early March forecast, and 0.4 pp from pre-war forecast) along with lowering real GDP growth outlook. 

2/ Potential Tax Cuts: Prior to the current conflict, inflation worries had jumped into focus over PM Takaichi's proposal for a 2-year suspension of the consumption tax on food. While intended to provide cost-of-living relief, markets feared the 5 trillion yen in lost revenue would be debt-funded, fueling fiscal slippage concerns. 

3/ BOJ Hike Path: At recent Monetary Policy Meeting (MPM) on March 19, the BOJ decided to hold policy rates citing the situation in the Middle East and the developments in crude oil prices as new risk factors, while noting that they will be paying due attention to its impact on Japan's economic activity & prices. This geopolitical shift adds another layer of complexity to BOJ's normalization path which is already pressured by a weak yen, tightening labor market and an uneven Shunto wage recovery. 

As uncertainties remain over the scale and duration of the ME conflict, the desk likes going short Japanese companies that are negatively impacted by rising inflation. To gain exposure to this thematic, we have constructed GS Japan Inflation Risk (GSXAJIFL) basket which consists of industries such as utilities, food products, beverages, restaurants, retail etc. that have high exposure to rising input costs but limited ability to pass these on. 

GS Economists' Core CPI (excl. Fresh food) Inflation Forecast with Scenario Analysis

Source: Goldman Sachs

Performance

Source: Bloomberg as of 19th March 2026

Finally, somewhat stating the obvious, commodity inflation remains the primary concern...

...and pipeline for that has started to fill (refined products, fertilizer, food, etc...)... and won't be eased anytime soon.

Professional subscribers can read much more from Goldman's Sales & Trading team here at our new Marketdesk.ai portal