I had added ₹6 crore and ₹4 crore correctly. The first amount was coupon cash received from a defined investment account. The second was an increase in the valuation of its unsold portfolio, from ₹100 crore to ₹104 crore. In a 2021 management paper I called the ₹10 crore total cash available to support an ₹8 crore discretionary office refit. The error survived the arithmetic because the arithmetic was not where the error lay. I had changed the meaning of one component while making the recommendation easier to read.
Treasury challenged the account before implementation. No disposal had produced the extra ₹4 crore receipt that my description implied. I withdrew the recommendation and supported a smaller ₹5.2 crore commitment considered against the identified ₹6 crore cash source. The ₹0.8 crore left in that bounded account was not the insurer's total liquidity, a measure of claims resources or proof that the smaller expense was permissible. Other approvals, accounting treatment and capital and risk assessment remained separate. The correction needed to preserve those distinctions rather than replace one overstatement with another.
The case is fictional, and it does not prescribe the accounting or regulatory treatment of an Indian insurer's investments. Its management question is narrower: what did the amounts in the recommendation actually represent? That question is simple enough to be overlooked by an experienced finance leader. A favourable performance account may contain several kinds of movement. When the decision shifts from describing performance to authorising a commitment, the relevance of those movements changes. The paper has to make that shift explicit instead of allowing the same attractive total to serve both purposes.
## A total can answer a different question
The coupon receipt and valuation movement could both be relevant to understanding what happened in the investment account. Their relevance did not make them interchangeable. A reader interested in performance might reasonably want to see both, alongside the basis and limits of the valuation. A reader deciding what cash can support a proposal needs a different account, including the source, timing, restrictions and permissions. Finance should connect those views without implying that a comprehensive performance number is also a bank balance or an unrestricted spending authority.
The problem becomes more likely when a recommendation passes through several levels of summary. A detailed schedule distinguishes movements, while a shorter page combines them under a convenient label. The combined number then appears in a slide or a discussion without the original distinctions. Nobody has necessarily made an arithmetic mistake. Yet the decision can now rest on an amount whose meaning has changed. I was responsible for the recommendation, so I could not treat the supporting schedule as an adequate defence. A correct appendix does not excuse a misleading executive account.
It would also be inadequate to solve the problem by banning every use of a combined investment measure. Different measures can be useful for different questions. The discipline is to state what a measure contains and avoid transferring its authority into a decision it does not support. A valuation movement can matter greatly without being cash. A cash receipt can be real without being available for any purpose management prefers. Those distinctions are not obstacles to a decision. They describe the conditions under which a decision can be understood and authorised responsibly.
## Correcting the paper does not generate money
The revised proposal did not recover ₹4 crore or create ₹0.8 crore of liquidity. It changed the recommendation to reflect an already-existing cash state. That is a worthwhile correction, but its value should not be exaggerated into a financial result. We retained a separate ₹21 lakh expense for reviewing the presentation and process. The cost was not netted into the bounded refit account, and the case does not claim that it produced a measured return. The organisation paid for work needed to make future recommendations clearer.
There was an operating consequence as well. Reducing the refit commitment from ₹8 crore to ₹5.2 crore meant that colleagues had to sequence the office work differently. A smaller commitment is not automatically a better one; it has a scope and opportunity cost. The correct funding account made that tradeoff visible. It did not prove that the original project had no value or that every discretionary expense should be delayed. The board and management still had to assess the proposal on its purpose, permitted funding and practical consequences.
I wanted my own role to remain visible because the recommendation had carried the authority of finance. The mistake was not merely a wording issue for someone to tidy after the meeting. The word cash had done work in the decision. It made an amount appear ready to support a commitment. Withdrawing the recommendation acknowledged that consequence. It also gave colleagues a reason to challenge meaning even when the numbers reconcile, rather than assume that a senior sponsor's summary must be the correct interpretation of a supporting account.
## Keep the distinction through the approval
The next step was to make the state of each relevant amount explicit at the point where it supported a recommendation. A receipt, a valuation, a conditional contribution and a future expected inflow can all appear in financial planning. They should not acquire the same certainty or availability merely because the plan needs a funding total. The relevant distinction depends on the decision, and it must survive the final summary. That is more useful than adding a long list of qualifications to every number without explaining which condition matters for the action being proposed.
Permission also remains a separate question. Identifying cash does not determine whether a company may use it for a particular purpose, distribute it or treat it as eligible capital. Those matters require the appropriate professional and governance assessment. I would not want the lesson from this case to become a simplistic instruction to spend whatever has reached a bank account. The financial account should show the source and its state so that the authorised decision-makers can consider the relevant conditions. Clear arithmetic supports that process; it does not replace it.
A board can ask management to explain those conditions without rebuilding every investment schedule. It can examine whether the recommendation uses a measure appropriate to the decision, whether material restrictions are visible and whether a favourable headline depends on treating unlike amounts as equivalent. The question should be specific enough to reveal a mistaken assumption. Asking only whether finance has checked the numbers may invite confirmation of the addition while leaving the meaning untouched, which is exactly how my original paper had become persuasive.
I now regard the translation from financial record to executive recommendation as a substantive part of the CFO's work. It is not simply a matter of making the presentation shorter or more polished. The translation decides which distinctions remain available to the people authorising a commitment. In this case, the distinction between a receipt and unsold appreciation was lost, and the proposed action had to change when it was restored. The lesson I would offer another leader is to ask what the total is being asked to prove before taking comfort from the fact that it adds up.