At a typical investment committee session, the agenda item covering how the cash position is held passes in a few minutes, while a credit margin of a few hundred basis points occupies half an hour in the same meeting. The ordering of the agenda itself declares a preference: the price of debt is treated as a negotiable item, whereas the location of cash is treated as treasury routine. Yet in a period during which inflation exceeds the yield on deposits or short-dated government paper, capital parked on the asset side of the balance sheet can generate erosion comparable in magnitude to the carrying cost of the debt sitting opposite it. Because that erosion never appears as a separate line in the income statement, it never enters the agenda either.
The same pattern grows sharper in capital-intensive project structures. When a developer or an industrial group holds a substantial cash balance against construction commitments falling due over the coming eighteen months, the maturity profile of that balance is generally constructed not around the commitment calendar but around the instrument that generates the least approval resistance. The preferred instrument is overnight or one-month, for the straightforward reason that no one is ever criticised on liquidity grounds. The actual drawdown schedule, by contrast, is knowable and typically spread across a far longer horizon, with a meaningful share of the commitments remaining untouched throughout the first twelve months.
The behaviour has a name — negative-yield allocation, the deliberate residence of capital in an asset whose real return is below zero, justified by preservation — and its mechanism draws on two distinct layers. The first is that loss aversion does not discriminate between nominal and real: so long as the nominal balance holds, the decision-maker registers no loss, even as the position contracts continuously in purchasing-power terms. The second is the economics of institutional approval, under which continuing the existing arrangement requires no fresh signature, no new committee resolution and no additional counterparty review, whereas extending maturity or changing instrument requires all three. Where these two layers combine, inaction becomes the cheapest available option.
The tendency is not, in itself, an error; under certain conditions it is entirely functional. In periods when counterparty risk cannot be priced, when market liquidity narrows, or when the institution's own commitment calendar has become indeterminate, surrendering real return in order to purchase certainty is a defensible trade, and the price paid can reasonably be read as the premium on optionality. The difficulty lies not in the shortcut but in its continuation after the conditions have changed. Once the period of uncertainty closes, the commitment calendar firms up and the range of acceptable counterparties reopens, the position requires justification once more — and nothing in the institutional architecture triggers that re-justification.
The first surface on which the institutional cost becomes visible is not the cash position but the project's financing structure. On one side a cash buffer earning a negative real return is maintained; on the other, construction loan drawdowns are initiated earlier than planned, or bridge financing is arranged at a wider margin. Because the two decisions are taken in separate meetings by separate teams, their net cost is never assembled into a single table. Treasury has preserved its liquidity ratio and the financing desk has closed on schedule; each team performs well against its own metric, while the group's blended cost of capital exceeds the sum of what either decision would suggest in isolation.
The second surface emerges at the valuation and due diligence table. In the review of a company or a portfolio, a large cash item that has sat motionless for an extended period triggers two readings on the buy side simultaneously: either weak capital allocation discipline, or provision against an undisclosed obligation. The first reading compresses the multiple directly; the second brings a widened scope of representations and warranties and, in some cases, a higher escrow percentage. In either case idle cash becomes an item that yields nothing to its owner while supplying negotiating leverage to the acquirer. The expectation that surplus cash will be read as comfort works, at the transaction table, reliably in reverse.
The third surface is the institution's own decision velocity. In organisations that carry a negative real-yield buffer over long periods, that buffer gradually hardens into a psychological threshold, and every investment decision requiring a dip below it encounters an additional layer of approval; the size of the buffer thus begins to operate as an unannounced ceiling on the investment budget. The consequence is not merely that capital waits at a cost, but that opportunities with narrow transaction windows — interconnection capacity allocations, land options, equipment manufacturing slots — are forgone. Since that loss is measured nowhere, the true carrying cost of the buffer is systematically understated.
What neutralises the tendency is not greater vigilance on the part of the decision-maker but the structural division of the decision itself. Three components of cash management should be separated. The first is operating liquidity, covering commitments falling due over the next ninety days and assessed solely on the criterion of accessibility. The second is the commitment buffer, covering payments already contracted but not yet drawn, whose maturity profile ought to be mapped directly onto the payment calendar. The third is the strategic reserve, tied to no contract and therefore requiring evaluation as a capital allocation decision — on the investment committee agenda, held to the same standard as a project. Absent the separation, the third layer continues to be carried on the justification belonging to the first.
Two records are required for the separation to function. The first is a rationale record kept at the moment of proposal rather than the moment of approval, in which the condition underlying the allocation is stated explicitly: which uncertainty, which counterparty constraint, which gap in the calendar. The second is a review cadence that tests the continuing validity of that condition at a defined rhythm, where the question posed in a quarterly session is not whether the position should be changed but whether the condition that produced the decision still holds. Framing the question this way shifts the burden of re-justification from change to continuation, and prevents inaction from occupying the default position.
BEIREK's intervention in capital-intensive project structures concentrates precisely on establishing that separation. The project commitment calendar — the EPC milestone schedule, equipment advance payments, interconnection and permitting fees, interest and commitment charges — is consolidated onto a single time axis and the cash requirement mapped month by month; that map is then overlaid on the existing maturity profile of the cash position, and the gap between the two curves indicates how much of the buffer genuinely needs to remain liquid, with the remainder carried to the committee agenda no longer as a treasury preference but as an explicit allocation decision. In the same exercise the construction loan drawdown plan is modelled alongside the cash utilisation plan, so that the interest cost of early drawdown and the real erosion of idle cash appear side by side in one table, and the aggregate effect of two separately optimised decisions becomes visible for the first time.
Alongside this, the condition record and the review cadence governing allocation decisions are established as components of the project governance structure, with their alignment against the covenant headings in the credit agreement — minimum liquidity thresholds, reserve account obligations, distribution tests — re-examined each period. The object here is not to shrink the buffer; in many structures the covenant package already mandates a minimum level, and that level is not open to discussion. The object is to draw the boundary between what is mandated and what is carried out of habit clearly in every period, and to require the second portion to reproduce its own justification.
Carrying a position with a negative real return is, under the right conditions, a defensible choice; what is not defensible is that the choice, once made, renews itself automatically and is never reopened on any agenda. What demonstrates an institution's capital discipline is not the size of its buffer but whether there exists a record of the date on which that size was set, the condition on which it rested, and the person who proposed it. Absent such a record, the buffer is not a decision but a residue.