Follow the internal transfer
I become cautious when a management presentation celebrates an improved business-unit result immediately after changing an internal allocation. The improvement may be informative. It may reveal that an activity had previously received too little credit for the funding it supplied or too little charge for the funding it consumed. But it is not necessarily new income for the institution. In banking, that distinction is essential because an internal funding method can change branch rankings before a single customer changes their behaviour or a single external cash flow changes its amount.
Our review at Lurenshan Commercial Bank reassigned CNY88 million of annual management contribution between branches and central treasury. Eight of thirty-six branches changed performance quartile. Those were significant consequences for the people managing the branches, and it would have been unreasonable to call the change merely technical. Equally, it would have been wrong to describe the reassigned amount as an improvement in bank-wide profit. The first task for finance was to make both facts visible in the same account without allowing one to obscure the other.
The original method gave limited recognition to differences in the funding commitments associated with products and customer behaviour. Colleagues could debate commercial performance using a measure that did not adequately express those differences. Treasury had the relevant funding analysis, but the management account did not bring it into the branch conversation clearly enough. I wanted the revised method to support better decisions about the business we pursued and the commitments we accepted. I did not want a more elaborate allocation simply because a more elaborate model was available.
That purpose required us to specify what the method was intended to influence. If it was supposed to improve the discussion of funding duration, the explanation needed to show duration. If it was intended to recognise different patterns of customer balances, those assumptions needed to be visible and open to challenge. A method that produces a precise charge while leaving its commercial meaning obscure is difficult to govern. The apparent precision may even discourage the questions needed to understand whether the charge is useful in a particular decision.
A fair transition is part of the method
We ran the revised account alongside the existing one for two quarters. This created additional work, and some colleagues wanted to move immediately to one set of numbers. I retained the parallel period because managers had made commitments under the previous approach. The comparison allowed them to identify how much of a changed result came from their business and how much came from the measurement. It also exposed explanations that made sense to the modelling team but did not survive a discussion with the people using the reports.
Finance reconciled the branch and central results to the bank-wide account. Treasury explained the funding assumptions, while commercial teams tested the behavioural consequences. Each contribution was necessary. A branch manager could identify an incentive that might encourage an unhelpful customer practice, but that observation did not establish the cost of funding. Treasury could explain a cost, but that explanation did not settle how a performance measure should be introduced. My role was to hold these questions together long enough for an explicit management decision.
Several managers objected because their branches moved down the table even though their customer activity had not deteriorated. I agreed that the distinction should affect the transition for relevant incentives. We did not conceal the revised result, and we did not pretend that the former measure remained the better long-term signal. We gave the change a defined introduction and retained the comparison. The cost was a more complex reporting period and slower adoption. The alternative would have made a methodological change feel like a retrospective judgement on decisions taken under different expectations.
This experience changed how I think about implementation. The transition is part of the economic method because it affects whether people will respond to the intended signal or spend their energy defending themselves against it. A technically sound allocation introduced carelessly can produce mistrust that persists after the calculations are understood. That does not mean every objection should delay a change indefinitely. It means the decision should distinguish a substantive challenge to the method from a fair concern about how past commitments are being evaluated through a new measure.
Look for a changed decision
Once the revised account was in use, I wanted evidence of changed decisions rather than evidence that a new spreadsheet had become familiar. Could managers explain why two funding relationships with similar balances might have different economic value? Could they identify where a proposed offer depended on assumptions about renewal behaviour? Could they distinguish a branch's commercial contribution from the effect of the central allocation? These questions were more revealing than a general statement that the organisation had adopted a more sophisticated approach to performance management.
In a six-branch pilot, the reviewed category of more stable customer balances rose from CNY5.4 billion to CNY6.1 billion over nine months. That was an observed movement during the period, not a controlled demonstration of the method's effect. Customer activity, pricing and the wider environment also changed. I kept those qualifications in the account. A useful internal method deserves to be assessed carefully; it does not deserve a causal claim merely because a favourable result followed its introduction. The additional balances and their cost still required their own commercial review.
I would also examine decisions that the method discouraged. A proposal that looks less attractive after a fuller funding charge may be abandoned for good reasons, but it may also reveal an assumption worth challenging. Finance should be able to explain the basis of the charge without treating the model as an authority beyond discussion. The point is to make the bank's commitments more intelligible. If the method prevents colleagues from asking a sensible question about a customer or a product, it is serving the report more faithfully than the enterprise.
Directors do not need to reproduce every calculation to oversee this work. They do need to understand the purpose of the method, the material assumptions, the effect on performance assessment and the route for revising it. I would show them the reconciliation to the institution's account and the distributional consequences among businesses. A large change in branch contribution deserves an explanation even when the bank's total is unchanged. The unchanged total is precisely why the board should ask what behaviour management expects the new allocation to produce over time.
My conclusion is practical: follow an internal transfer through to the decision it is intended to improve, and keep it separate from external earnings throughout the discussion. This makes a performance table less comfortable in the short term, because a ranking can no longer stand alone as a complete judgement. It makes the table more useful in the longer term, because colleagues can understand what it measures and what it cannot establish. I would rather explain a redistribution honestly than attach a growth story to income the bank has not actually earned.