Digital assets in a multi-asset portfolio: the questions before the allocation

Discussions about digital assets in an institutional portfolio tend to begin, and often end, with how much. It is the most concrete question and the least useful one to start with, because the answer depends entirely on decisions that have not yet been made.

An allocation intended as a source of uncorrelated return is a different instrument from one intended as exposure to a technology theme, even at identical size. The two would be funded differently, measured differently, reported differently, and abandoned under different circumstances. Deciding the number before the role produces an allocation that nobody can subsequently explain.

The funding source is the most revealing decision in the process, and frequently the least deliberate. An allocation carved from an alternatives or absolute-return sleeve implies one thing about its expected behaviour. One taken from growth equity implies another. One funded from cash implies a third, and usually a different tolerance for the volatility that follows.

The reason this matters beyond bookkeeping is that the funding source determines what the allocation is implicitly being compared against. A holding funded from equities will be judged against equities, whatever the paperwork says, and that comparison will be made most forcefully at the least convenient moment.

An institutional portfolio has an existing apparatus for measuring and challenging risk. Digital assets frequently do not fit it cleanly, and the mismatch is worth confronting before an allocation rather than after.

Liquidity mismatches are among the more common sources of difficulty, and they operate at two levels that are easy to conflate.

The first is the liquidity of the underlying exposure: how quickly positions could genuinely be reduced, in size, under stressed conditions rather than normal ones. The second is the dealing terms of the vehicle through which the exposure is held — notice periods, dealing frequency, and any provisions that can restrict redemption.

The two need to be consistent with each other and with the liabilities of the investor. A vehicle offering terms more generous than its underlying can support is not offering liquidity; it is deferring a problem to whichever investor asks last.

A volatile allocation will drift from its intended weight, and it will do so most sharply exactly when opinion about it is strongest. Whether that drift is managed by policy or by discussion is most of what separates a governed allocation from an ungoverned one.

A policy set in advance — the bands, the frequency, who executes and what constitutes an exception — converts a series of contentious individual decisions into an administrative process. Set afterwards, or not at all, each rebalancing becomes a fresh debate conducted under the influence of whatever has just happened, which is the condition under which committees make their least consistent decisions.

The question asked least often at the point of allocation is what would cause it to be reversed. An allocation with no articulated exit condition is difficult to review, because there is no standard against which to review it.

The useful version of the question is not a price level. It is a set of circumstances: a change in the regulatory position that alters what can be held, a failure of the thesis on which the allocation was justified, a deterioration in the operational conditions that made it accessible, or simply the passage of enough time without the intended role being fulfilled.

Writing those down at the outset costs little and changes the character of every subsequent review. It also makes the allocation easier to defend, because a position that was made with stated conditions is a decision rather than a drift.