Portfolio Construction

What market neutral does in a drawdown

A market neutral strategy is built to remove the direction of the market from a return. It does not remove everything else, and the episodes in which these strategies have lost money badly were rarely caused by the market going down. This note sets out what neutrality actually removes, what it leaves behind, and what a drawdown is really testing.

Manuel E. De Luque MuntanerCEO & Founder

21 August 202611 min read

Key takeaways

Neutrality is a statement about one exposure — the direction of the market — and about nothing else. It is not a statement about risk, volatility or capital preservation.
The exposures that remain are financing, leverage, crowding, liquidity, counterparty and the residual factor bets inside the portfolio. These are what have historically produced the losses.
The best-documented failure of equity market neutral, in August 2007, happened while the strategies' own market thesis was intact: a large similar portfolio was unwound quickly and everyone holding the same position was marked against it.
In digital assets the same shape appears through the basis and funding: a position that earns a spread is short the event in which that spread stops behaving, and it is usually levered.
A drawdown does not test the thesis. It tests the financing, the exit and the assumption that the other holders of the position will not need to leave at the same time.

What neutrality actually removes

A market neutral strategy holds offsetting long and short exposure so that the direction of the underlying market largely cancels. What is left is intended to be the return of the manager's selection, spread or relative-value judgement, separated from whether the asset class went up or down.

That is a precise claim, and it is worth reading precisely. It says that one exposure has been removed. It says nothing about the size of the remaining exposures, nothing about leverage, nothing about liquidity, and nothing about whether the position can be exited. "Neutral" is a description of a hedge, not a description of risk.

The word does a great deal of work in a conversation, which is why it is worth being pedantic about it. An allocator who hears "market neutral" and infers "low risk" has substituted one exposure for the whole risk profile.

The exposures that remain

Removing direction leaves a portfolio that is still exposed to several things, and in the episodes where market neutral strategies have lost badly, these are what did it.

Financing and leverage — a small residual return is usually levered to become a meaningful one. Leverage is provided by someone, on terms that can change, and it is the terms that fail first.
Crowding — a good relative-value idea is rarely proprietary. Being right about the position and wrong about who else holds it is a distinct risk, and it does not appear in any exposure report.
Liquidity asymmetry — the long and the short leg rarely have the same liquidity. Unwinding a hedged position is two trades, and the harder one sets the cost.
Counterparty and venue — the hedge depends on the other side, the borrow, the clearing arrangement and the venue continuing to function. A hedge that cannot be relied on in stress is not a hedge in stress.
Residual factor exposure — neutral to the market is not neutral to size, momentum, liquidity or quality. A portfolio can be beta-neutral and still be a concentrated bet on something.
Basis and funding — where the return comes from a spread, the position is short the event in which the spread stops behaving. That event is uncommon and is not therefore unlikely.

August 2007, and why it is still the reference case

The clearest documented episode is the week of 6 August 2007, when quantitative equity market neutral portfolios recorded severe losses. It is worth studying not because equities are digital assets, but because of what caused it.

The account set out by Khandani and Lo attributes the episode to the rapid unwind of one or more large quantitative market-neutral portfolios — plausibly a forced liquidation, driven by pressures in an entirely unrelated market. That unwind pushed prices against everyone else holding similar positions, which triggered their own stop-loss and deleveraging rules, which pushed prices further.

Two things about that are worth holding on to. The first is that the strategies' analytical thesis was not what failed; several of the positions performed as intended once the unwind was over. The second is that the trigger came from outside the strategy's own market entirely. A portfolio can be neutral to its market and still be fully exposed to the balance sheets of the people standing next to it.

The Financial Stability Board's supervisory work describes the same mechanism in general terms: leverage, concentration and crowded relative-value positions amplify a shock, because tightening funding terms produce margin calls, forced deleveraging and sales into a market that is already moving.

The digital asset version of the same shape

In digital assets the most common market neutral construction earns a spread rather than a direction — most familiarly, holding an asset while selling a future or a perpetual against it, and collecting the difference between the two.

The mechanism is easy to describe and its failure modes follow directly from it. The spread exists because someone is paying for leveraged long exposure. When that demand fades, the spread compresses and the return goes with it; when it inverts, the position pays instead of earning. Neither outcome requires the price of the underlying to do anything in particular.

Leverage is what makes the strategy interesting and it is also the transmission mechanism. A levered spread position faces margin against a mark, and the mark moves with the spread rather than with the thesis. Add a venue that is simultaneously the exchange, the clearer and the custodian of the collateral, and the counterparty exposure is not a footnote to the strategy — it is part of it.

The point is not that the construction is unsound. It is that the sources of loss are structural and identifiable in advance, which means an allocator can ask about them in advance.

What a drawdown is actually testing

A period of stress is not principally a test of whether the manager was right. It is a test of the arrangements around the position, and those can be examined before any stress arrives.

Who provides the financing, on what terms, and what notice is required to change them?
What happens to the position at the point the financing is withdrawn — not what the manager would prefer to do, but what the documents permit?
How similar is the position to what other managers of the same style are holding, and how would anyone know?
Which leg is harder to unwind, and what does unwinding it cost in a market that is already moving?
Does the hedge depend on a single venue or counterparty continuing to operate normally?
What is the residual factor exposure once market direction is removed, and is it deliberate?

What this note does not show

This note presents no performance data. It contains no return, no drawdown depth, no correlation and no recovery period, for any strategy, manager or vehicle — including any of Block Asset Management's. That is a deliberate limit, not an omission: figures of that kind are meaningful only alongside their basis, their fees, their period and their share class, and a public educational note is not where that belongs.

It also is not a prediction. The mechanisms described here explain how these strategies have been observed to fail; they do not establish how any particular strategy will behave in a future episode, and no strategy is being characterised here as robust or fragile.

What the note is for is narrower and more useful: the sources of loss in a market neutral strategy are structural, they are known, and they can be asked about before capital is committed rather than diagnosed afterwards.

References

The sources behind the factual claims in this note. Where a figure could not be traced to a source of this standard, it is not stated.

National Bureau of Economic Research — Amir E. Khandani and Andrew W. Lo — What Happened to the Quants in August 2007?: Evidence from Factors and Transactions Data (NBER Working Paper 14465) (2008)
The documented case in which quantitative equity market-neutral portfolios lost heavily in a single week, attributed to the rapid unwind of a large similar portfolio rather than to the direction of the market. Later published in the Journal of Financial Markets.
Financial Stability Board — Leverage in Non-bank Financial Intermediation — final report (2025-07-09)
The supervisory account of how leverage, concentration and crowded relative-value positions transmit stress: tightening funding terms produce margin calls, forced deleveraging and fire sales that move the very prices the position depends on.

Block Asset Management

How Block Asset Management helps

Nothing below is a claim about outcomes. It describes what is examined before an allocation is made, which is the same set of questions this note argues an allocator should be asking.

Financing terms, read rather than summarised

How leverage is provided, on what terms, and what a provider is entitled to change and at what notice — established from the arrangements themselves.

Liquidity of both legs

A hedged position is unwound as two trades. We assess the harder one, in the size actually held and under stressed rather than normal conditions.

Counterparty and venue concentration

Where an exchange also clears the trade and holds the collateral, that concentration is treated as part of the strategy rather than as an operational detail.

Residual exposure after the hedge

What the portfolio is still exposed to once market direction is removed, and whether that residual is an intended part of the approach or a by-product of it.

Supervision of systematic implementation

Where a strategy is implemented systematically, the models are the provider's. Our role is selection, due diligence and continuous supervision of that provider, and material changes it makes return to the investment committee as a new decision.

Findings recorded, including what could not be established

Where a question cannot be answered from evidence, it is recorded as unanswered, which is what allows it to be weighed openly in a decision.

Market neutral describes a hedge, not a risk profile. The strategies are worth holding for the reason they were designed — a return that does not depend on the direction of a volatile asset class — but the losses they have produced came from financing, crowding, liquidity and counterparties, not from the market going down.

That is the useful conclusion, because every one of those is a question that can be asked and evidenced before an allocation, rather than a property that has to be taken on trust and discovered in the middle of an episode.

If your organisation is evaluating market neutral strategies in digital assets, our investor relations team would be glad to discuss how we assess them.

Important information

This material is provided for information purposes only and is intended for professional and qualified investors. It does not constitute investment advice, an offer or a solicitation to buy or sell any financial instrument, nor a recommendation of any strategy. It presents no performance data of any kind, and nothing in it describes how any Block Asset Management strategy or vehicle has behaved or would behave. A market neutral construction removes exposure to the direction of a market; it does not make a strategy low risk, capital protected or free from loss, and substantial loss remains possible. The historical episodes referenced are illustrative of mechanisms and are not indicative of future outcomes. Past performance is not a reliable indicator of future results.

Related research

Continue reading BAM research

This note is part of Block Asset Management's research on institutional digital asset investing. Explore the wider library, or read how we assess managers and structures before any allocation is made.

Browse all research Our investment approach