Placed side by side, three consecutive years of budget deliberations at a multi-business group tend to reveal something more striking than the movement of the total capital envelope: the percentage distribution of that envelope across operating lines barely moves. The aggregate may expand in a growth year and contract in a defensive one, yet the ratio of one line's share to another's holds in place like a template carried in institutional memory. The deliberation itself is organised in a way that confirms the template, since the discussion proceeds not from the question of where an incremental dollar would earn the highest return, but from each line's prior-year share and the permissible band of deviation around it. The reference point a business unit invokes in defending its request is its own allocation history rather than the return available on alternative uses.
A second pattern observable in the same room concerns the asymmetry with which requests are examined. Capital destined for a new line, or for a proposition without an operating history, arrives accompanied by a detailed business case, a sensitivity analysis and a payback schedule, and is tested line by line; maintenance, replacement and capacity spending inside an established line passes with markedly less interrogation, presented as it is as the continuation of a structure that already exists. Both requests, however, consume the same capital and carry the same opportunity cost. Viewed in isolation, the implicit privilege granted to continuity reads as prudence, and within a single cycle it usually is. Compounded across cycles, it operates as a mechanism that locks the group's capital into assets whose returns are declining, without any single decision ever having authorised that outcome.
The name for this pattern is capital-allocation failure — the routing of capital toward uses determined by historical distribution and internal bargaining position rather than toward those offering the highest risk-adjusted marginal return. At the centre of the mechanism sits a silent convention about who owns capital by default. In most groups, capital is not treated as a resource the centre redistributes each year, but as property already held by the operating lines, which the centre may reclaim only upon showing cause. Constructing the burden of proof in that direction inverts the arithmetic of the allocation decision itself: the operative question is no longer where this capital would earn the most, but whether the case for removing it from its present holder is strong enough to survive the internal friction that removal would generate.
A second layer feeding the mechanism is the difference in quality between decision inputs. A mature line, having a long record behind it, supplies a granular, internally consistent and reassuring data set; a new line, by its nature, arrives with projections that are wide, interval-based and difficult to defend under cross-examination. To the extent that the decision mechanism penalises uncertainty directly rather than pricing it, capital is systematically directed toward the side with superior measurability — which is, in most portfolios, the side sitting on the flat part of its growth curve. The third layer is bargaining power. The seniority of the executive representing a line, the length of that executive's history inside the group and proximity to the chief executive influence the allocation outcome independently of the content of the submission, and this influence appears in no budget schedule, no variance report and no post-investment review.
Characterising these tendencies as errors would be misleading, since each of them lowers a real cost under the conditions that produced it. Capital remaining attached to its existing lines spares the organisation from renegotiating its internal settlement from zero every year, allows unit managers to plan on horizons longer than the budget cycle, and preserves the trust on which cross-unit cooperation rests. The weight placed on measurability protects capital from being committed to undocumented assertions. The difficulty lies not in the shortcut itself but in its persistence after the conditions that justified it have changed: once a line's marginal return falls below the group's cost of capital, the preservation of that line's historical share ceases to finance institutional peace and begins to finance the erosion of institutional value.
The institutional cost first surfaces not in the income statement but in the spread between the consolidated return on capital and the return generated by the strongest line in the portfolio. It is entirely possible for no individual business unit in a group to be performing badly while the consolidated figure sits well below what the best line achieves on its own capital; the distance between the two is the price of the allocation decision. That price is never posted as an expense, never appears as a variance, and never enters any executive's performance assessment. It accumulates instead across years, becoming visible only in the gap between the group's compound growth rate and that of the markets it competes in — by which point the decisions that produced it are a decade old.
The second cost sits on the exit side. Holding a low-return asset has the appearance of a decision not taken, whereas the reallocation of working capital, maintenance spending and, most scarcely, management attention to that asset in every budget period constitutes an implicit investment decision made repeatedly. Because the decision is never taken explicitly, it is never tested explicitly either; the group, having never priced the divestment alternative, does not know where the difference between the asset's market value and its carrying value has gone. In a transaction process that information gap converts directly into negotiating loss, since a counterparty evaluating the weakest line of a portfolio invariably prices it with a colder eye than the seller, and prices the seller's unfamiliarity with its own alternatives alongside it.
The third cost emerges on the examination table. When a buyer or a lender asks for five years of capital expenditure mapped by business line against realised return, groups without allocation discipline are largely unable to produce the mapping — the spending record exists, the return record exists, but the decision record that would connect them does not. This absence works in two directions during a valuation discussion. Because past investment cannot be shown to be the product of a repeatable logic, the discount applied to forward projections widens; and to the extent that the buyer concludes allocation judgment resides with the founder or a single senior executive rather than in a process, the earn-out structure and the post-closing commitment conditions attached to that individual expand accordingly. Both effects are priced quietly, and neither is usually attributed in negotiation to the absence of a decision log.
Neutralising the tendency is a matter of institutional architecture rather than individual awareness, and the architecture has four components. The first is the reversal of the burden of proof: capital is defined as belonging by default to the centre, each line rejustifies its share on a zero-based footing rather than against the prior year, and continuation items face the same threshold of examination as new ones. The second is keeping the decision record at the moment of proposal rather than the moment of approval — once the assumptions relied upon, the alternatives considered and the return threshold applied are written down before the outcome is known, comparing result against assumption three years later becomes possible. The third is separating the allocation rhythm from the budget calendar, since capital allocation that remains a subordinate heading within the annual budget cannot escape the prior-year template forming the budget's untouchable spine. The fourth is institutionalising the counter-argument: every material allocation proposal has a participant explicitly charged with defending the alternative use.
BEIREK's intervention in this area is built on converting capital allocation from a budgeting exercise into a traceable decision process. In multi-asset groups and portfolio transformations, the work begins with a capital map: every investment item of the recent period is reconciled across four axes — business line, deciding authority, governing assumption and realised outcome — so that the disconnection between the spending record and the return record becomes visible as a document rather than an impression. An explicit allocation threshold anchored to the group's cost of capital is then defined for each line, together with a review trigger that activates when performance falls beneath it. The trigger imposes no obligation to divest; it imposes an obligation to reopen the decision, which is a materially different and considerably more useful instrument.
On the process side, a rhythm is operated in which allocation proposals are entered into a central decision register under their proposal date rather than their approval date, each accompanied by an explicit comparison against the alternative use and a short pre-mortem note, and that register is placed on the table as the opening document of the following period's deliberation. The near-term effect of this structure is that the allocation discussion becomes noticeably harder in its first year, since the justification for existing shares is being asked openly for the first time and no line has prepared an answer. The medium-term effect appears when the group enters a transaction or financing process and is able to demonstrate, on paper, that its capital decisions rest on a repeatable logic independent of any single individual; the value of that demonstration is measured in the discount applied to projections and in the scope of the earn-out.
A group's allocation discipline is measured less by how much capital it directs toward its strongest line than by how quickly it withdraws capital from its weakest one. That second interval appears as a heading in no management report and is therefore rarely tracked, while the first is rehearsed in every strategy presentation. The variable that determines long-run return on capital is, to a substantial degree, the one nobody is counting.