Global macro and alternative risk premia: how the big institutions diversify

After trend following (Managed futures and trend following: the diversification that works in a crisis) and buying volatility (Long volatility and tail hedging: paying for crashes), one continent of the alternative world is still unexplored: global macro. These funds bet on the big economic forces, interest rates, currencies, inflation, commodities, equity indexes, instead of on individual securities. This is the family of legends: Soros against the pound in 1992, Druckenmiller. It is also a modern industry, largely systematic, that came through 2000-2002, 2008 and 2022 in positive territory while ordinary portfolios were being punished.

This article covers what these strategies really do, where their returns come from (the idea of alternative risk premia, with commodity carry as the case study), what the hedge fund indexes honestly show, and why the practical conclusion for a retiree stays more modest than the pitch. By the end you will be able to screen an "alternative" fund brochure in ten minutes, and to steer clear of the products that wrecked the category in the 2010s.

What a macro fund really does§

The toolkit is wide, but the basic moves fit on one hand.

Directional rates anticipate the central bank cycle: long bonds ahead of the cuts, short ahead of the hikes. Macro funds played 2022 that way, short duration, while the 60/40 lived through its worst year.

Currencies play the gaps between monetary policies and the exchange-rate regimes. This is where the famous trades were made.

Relative value bets on the spread between two related instruments, two maturities on the same curve or two neighboring countries, rather than on a direction.

Commodities follow supply cycles, geopolitics and seasonality.

Then there is the style. The discretionary version leaves those judgments to a team. The systematic version writes them down as rules, and trend following is only one subset of that, industrialized and transparent.

Seen through the category's indexes (HFRI Macro, the SG Macro Trading Index), the profile has three features a retiree can use. Its correlation to stocks stays low, 0 to 0.3 depending on the period. It made money in the three recent hostile regimes, 2000-2002, 2008 and 2022. And its long-run expectation, after fees, is real but modest. That expectation eroded badly after 2010: a decade of zero rates and managed markets starved the strategy of its fuel, which is diverging cycles. The 2021-2023 stretch brought it back.

🔬 What the indexes show, and what they hide

Three cautions, in order of importance.

Dispersion. The gap between the top and the bottom quartile of macro funds runs past 10 points a year, far wider than in trend. You cannot "buy the category" through an index, because hedge fund indexes are not investable. And buying a single fund mostly exposes you to the talent, good or bad, of that one shop.

The biases. Hedge fund databases overstate past performance by 2 to 4 points a year, through survivorship bias and voluntary reporting (Simulator traps: independence, US bias, survivorship takes the mechanism apart).

Convergence. The investable, replicable part of macro, systematic premia and trend first among them, is already available elsewhere at low fees. What belongs to discretionary macro alone, the alpha of judgment, is real at the very best shops and impossible to spot in advance. The research on manager persistence is blunt: past performance predicts close to nothing (Factors (Fama-French, value, momentum) in the withdrawal phase reaches the same verdict on equity funds).

Alternative risk premia: the catalog behind the curtain§

In the 2010s the industry offered to take hedge funds apart and sell the pieces: alternative risk premia (ARP). These are systematic long-short strategies, each one meant to capture a documented source of return (Where returns come from: risk premia covers what makes a premium a premium).

The standard catalog is short. Momentum and trend, which follow moves. Carry, which holds what pays to hold and funds it with what does not. Value, which buys the cheap and sells the expensive. Plus a few more technical premia, such as selling volatility, which is Long volatility and tail hedging: paying for crashes seen from the other side of the counter.

Case study: commodity carry. You do not hold a commodity the way you hold a stock. You hold futures contracts, and you have to replace them as they expire, which is called rolling the position. When the curve is in backwardation, the far contracts trade below spot, and rolling earns a positive return, the roll yield. In contango it works the other way around, and rolling costs money.

So systematic carry goes long the commodities in backwardation and short the ones in contango, and it pockets a spread documented over a century. The core literature, from Gorton and Rouwenhorst to AQR's work, measures gross premia of 3% to 6% a year for the long-short versions.

This is also the hidden lesson of "plain" commodity ETFs. A long-only tracker on a broad index spends most of its life paying the contango, which explains decades of disappointing returns even as spot prices rose. If you hold commodities other than gold, an optimized-roll vehicle is not a refinement: it is what makes the position worth holding at all (Gold in a retirement portfolio handles gold, which is its own case).

The roll yield in one picture. In backwardation, a contract bought below spot converges up to it and pays. In contango it is the reverse: a naive tracker pays that slope at every roll.

The category's crash test. Retail ARP funds had their hour of glory around 2015-2017. Then came the wreck, commercial and financial, in 2018-2020. The value premium went through its worst episode on record. Selling volatility blew up in February 2018, then again in March 2020. And the "all premia" baskets lost 15% to 25% where the marketing had promised calm decorrelation. Many of them closed.

The lessons hold up. Alternative premia are real but cyclical, and each one has its winters of five to ten years, exactly like trend. Their correlations with each other rise in a liquidity crisis. And fees of 1% to 2% a year eat most of an already modest gross expectation. A 4% gross premium, minus 1.5% of fees and then tax on what is left, does not pay for much of the patience it demands.

The practical verdict for a retiree§

It comes down to a simple hierarchy, three tiers deep.

Tier 1, and it is plenty. The investable, proven part of macro is called trend following. It has its own article, vehicles you can check against a public index (Building it with US-listed ETFs: the retiree's shopping list), and a 5% to 15% slot in the table of defenses (Managed futures and trend following: the diversification that works in a crisis, Defensive assets: the map and the roles).

Tier 2, optional. A diversified commodity fund with an optimized roll, or a multi-premia fund that survived the 2020 purge. Three conditions apply: a real ten-year track record, a flat fee under 1.5% with no performance fee, and a correlation to stocks verified below 0.3 in 2008 and in 2022 alike. Those two crises are opposite in nature, and the test is demanding on purpose.

Tier 3, closed. The big discretionary macro funds that make the history books do not take money from individual investors. Their retail clones have neither the teams nor the results.

Screening an "alternative" brochure comes down to five questions.

What does the fund do when stocks rise three years running? If it makes money too, that is disguised beta.
What is its worst drawdown, and in which year? If it lands on top of the equity drawdown, the diversification is decorative.
What are the total fees, performance fee included?
Which public index does it measure itself against? No benchmark, no purchase.
And the killer: why does this strategy earn anything, who is paying, and why will they keep paying (Where returns come from: risk premia)?

A salesman who answers the fifth question with "our proprietary model" has already given you your answer.

🧮 A full alternative sleeve, sized

Plan: $1.2M, $45,000 a year, 40-year horizon. Total alternative budget capped at 20%. Past that, the psychological gap with an ordinary portfolio becomes unmanageable (The psychology of spending: why spending it is the hard part).

The allocation: 10% trend, checked against the SG Trend, 5% gold, and 5% left open. For that last 5%, two options. A second trend program from another shop, to diversify the manager, which is the plainer choice. Or a surviving multi-premia fund that clears all five questions.

Before buying that third sleeve, a simulator has to show one thing: at least 0.1 of a point of safe withdrawal rate gained on the 1966 and 1973 stress, against the version without it. Otherwise the added complexity has not earned its place. The frequent verdict: once trend and gold are in place, the third sleeve adds almost nothing, and the simple version wins by forfeit.

The essentials§

Global macro bets on macro imbalances: rates, currencies, commodities. It has historically made money in the regimes that punish ordinary portfolios, in 2000-2002, 2008 and 2022, with a correlation to stocks that stays low.
But the category cannot be indexed: huge dispersion between managers, survivorship bias in the databases (2 to 4 points a year), no detectable persistence, and the big funds shut to individuals.
Alternative risk premia (trend, carry, long-short value) are documented, but cyclical and expensive. The 2018-2020 purge of the retail ARP funds (−15% to −25%) showed what a basket sold on decorrelation is worth.
Commodity carry explains one thing that matters in practice: a long-only commodity ETF pays the contango most of the time, and only an optimized-roll vehicle, or a long-short one, holds up.
The retiree's order of business: trend first (the investable part of macro), then at most one extra sleeve, put through the five questions and the simulator test, and the rest of the budget elsewhere. The diversification that counts comes from a total alternative sleeve of 10% to 20%, not from a collection of funds.

Going further§

Gorton & Rouwenhorst, "Facts and Fantasies about Commodity Futures" (2006) and its 2015 update: the foundation of the commodity file.
AQR, "Carry" (Koijen, Moskowitz, Pedersen, Vrugt, 2018): the carry premium measured across every asset class.
Antti Ilmanen, Expected Returns (2011): the reference manual on premia, alternatives included (Forward-looking expected returns (Morningstar, Vanguard, the investment banks)).
HFR and Société Générale (the SG Macro Trading Index): the public indexes for placing any macro fund.
In this book: Managed futures and trend following: the diversification that works in a crisis (the investable part), Long volatility and tail hedging: paying for crashes (the variance risk premium, from the buyer's side), Where returns come from: risk premia (why a premium exists and persists), Defensive assets: the map and the roles (the table of roles).