Capital Allocation: How to Judge Management

In February 2025, on Dilip Buildcon's earnings call, the CFO gave a number you could write down and check. Standalone net debt would close the year at "around INR1,500 crores," fall to "around INR1,000 crores or even lesser" by March 2026, and the company would be net cash by FY27 (Dilip Buildcon Q3 FY25 concall, Feb 2025). Specific, dated, falsifiable — exactly the kind of statement that tells you, in time, whether you're dealing with a disciplined allocator or a hopeful one.

By November 2025 the same management was explaining why none of it happened. Net debt hadn't fallen to ₹1,000 crore; it had risen to ₹2,102 crore, and the debt-free date had slid from FY27 to FY28 (Dilip Buildcon Q2 FY26 concall, Nov 2025; Inve data, Q2 FY26). I'm not here to pick on one road-builder — this is an illustration, not a view on the stock. But it's the cleanest doorway I know into the one job a CEO cannot delegate, the job that quietly compounds into your return. Where every rupee the business keeps actually goes — the core, an acquisition, debt, your pocket, or just idle on the balance sheet — is decided quarter after quarter, mostly without a vote. A merely good business run by a poor allocator becomes a mediocre stock. The reverse is rarer and worth far more.

This is a teaching piece in the Buffett spirit: a principle, a real Indian example, and a test you can run yourself. The goal is that you finish able to judge an allocator from the public record — and, crucially, to check what they say against what they actually do with the cash.

What is capital allocation, and why is it the whole game?

Every year a profitable company throws off cash it doesn't need to keep the lights on. Management has five things it can do with that cash, and only five:

1. Reinvest in the existing business — capex, capacity, R&D, working capital.
2. Acquire another business — bolt-on or transformational.
3. Pay down debt.
4. Return it to shareholders — dividends or buybacks.
5. Let it sit — cash and investments piling up.

That's the entire menu. The art is matching the choice to the opportunity. Reinvesting at a high return on capital is the best option when such returns are available. When they aren't, forcing growth — chasing a diworsification, overpaying to look bigger — destroys value while looking like ambition. The disciplined allocator does the unglamorous thing: returns cash when reinvestment returns are poor, and pours it in when they're rich. Plenty of managements do the opposite, because empire feels better than discipline.

Buffett's one-dollar test is the cleanest framing ever written for this. For every rupee a company retains rather than pays out, has it created at least a rupee of value over time? Take the earnings a company kept over the last seven to ten years and ask whether the intrinsic value of the business grew by at least that much. If a company has kept years of profit and the per-share value those profits should have built simply isn't there, the money was spent badly — and it should have come back to you.

How do you test an allocator from the public numbers?

You don't need management's intentions. You need their record, and it's all disclosed. Four checks, in order.

One — the retained-earnings test. Sum the profits the company kept over several years and compare it to the growth in earning power and book value over the same window. If a company retained thousands of crores and its operating earnings barely moved, the reinvestment earned a poor return — the cash would have served you better as a dividend. (Profit retained but never showing up as cash is its own warning; see cumulative cash flow vs profit.) This is the single most revealing number, and almost nobody computes it.

Two — incremental return on capital. Look at the change in operating profit divided by the change in capital employed over five years. This tells you what the freshly-invested rupees earned — different from the company's average return. A great legacy ROCE can hide new money going into projects that earn far less; the average hides it, the incremental return exposes it. (For why average return figures mislead, see ROCE vs ROE: what the gap reveals.)

Three — the acquisition trail. Did the company make acquisitions, at what multiples, and what happened to them? Goodwill that later gets written down is capital allocation failing in public. Serial acquirers who never integrate, and who keep buying to mask slowing organic growth, are a recognisable type — and acquisitions funded by mounting debt deserve special scrutiny (see how to spot a debt-trap stock).

Four — the cash-return discipline. When reinvestment opportunities genuinely dried up, did management return cash, or let it pile up earning treasury yields while the stock languished? A growing idle-cash balance with no plan is allocation by avoidance.

The point isn't any single number — it's the consistency between them. A management that reinvests at high incremental returns, acquires rarely and cheaply, and returns the rest is allocating like an owner. One that reinvests at falling returns, buys to grow the empire, and hoards the remainder is allocating like a manager protecting his salary.

What does the concall reveal that the financials don't?

The financials tell you what management did with the cash. The concall tells you how they think about it — and whether the stated logic survives across quarters. This is where the teaching meets the tracking.

Listen for the allocation philosophy stated explicitly: return thresholds, capital discipline, returning cash when they can't deploy it. Then check whether the spending matches the speech. And listen hardest for the falsifiable commitments — "net cash by FY27," "capex normalises to ₹500 crore," "buyback once the cycle turns." These reveal an allocator's character over time, if you remember to check them.

Most people don't. Across the management commitments tracked on Inve, barely half — about 55% — are delivered as stated, and 47% of companies have at least one piece of guidance that quietly went silent, never mentioned again on a later call (Inve data, as of 2026-06-12). Capital-allocation commitments seem especially prone to that quiet disappearance, because deleveraging timelines and capex plans are the easiest to let slide when the cash gets tight. Keeping every dated commitment across a 10–15 stock portfolio in your head, quarter after quarter, is the part no human does well — and that forgetting is precisely what lets a poor allocator escape accountability. That's the job Inve's Promise Tracker does.

A deleveraging timeline that moved twice — what it teaches

Watch one commitment travel across four calls. This is an illustration of how to read an allocator, not a view on the stock.

(Source: Dilip Buildcon Q3 FY25 and Q2 FY26 concalls; Inve Promise Tracker status history, Q3 FY25–Q3 FY26.)

Three forces converged here, and they're worth naming because they recur. First, an anchoring number — "₹1,500 crore by March" — that sounded precise and gave the call a clean headline. Second, a dated extension that quietly reset the goalpost: when the first target slipped, the fix wasn't a smaller commitment, it was a later one ("FY27"), and when that slipped, a later one still ("FY28"). Third, the cash itself moving the wrong way the whole time: standalone net debt went from ₹1,515 crore (March 2024) to ₹2,102 crore (September 2025) (Dilip Buildcon Q3 FY25 and Q2 FY26 concalls), and consolidated borrowings climbed from ₹7,240 crore to ₹10,375 crore over the same stretch (Inve data, Mar 2024 vs Sep 2025). Each force alone is forgivable. Stacked, they describe an allocator whose stated plan and actual cash flow had parted ways.

To management's credit, they said so plainly. "We had originally promised that this year, we will reduce INR500 crores of debt. However, in the current scenario, it has increased a little bit" (Dilip Buildcon Q2 FY26 concall, Nov 2025). The honesty is real, and the cause they cite — a two-year slump in order inflows starving free cash flow — is the kind of external pressure no plan survives. That's the steelman, and it matters. But it's also the entire point of the test: a deleveraging guidance is only as good as the cash flow underneath it, and the way you find out is by holding the February number next to the November one. The number that did the work here isn't ₹2,102 crore on its own. It's ₹2,102 against a guided ₹1,000 — the gap, not the figure.

Now the inversion. How does a beginner get hurt by a story like this? Not by the debt — by believing the date and stopping there. You hear "net cash by FY27," you file the company under "deleveraging story, fixing itself," and you never re-open the file. The damage isn't the missed target; it's the year of unquestioning patience it bought. A dropped commitment costs you the one thing you can't get back — the attention you'd have spent re-checking.

What disciplined capital allocation looks like next to it

For contrast, take a management that said the same kind of thing and did it. Anant Raj guided in early 2024 that it would be net-debt-free by December 2024 — and by FY26 its net debt had sat below ₹50 crore for five consecutive quarters, effectively zero, ahead of the original timeline (Inve Promise Tracker status history, Q4 FY24–Q2 FY26). The balance sheet agrees: borrowings fell from ₹1,591 crore (March 2019) to ₹482 crore (March 2025) while reserves rose from ₹2,442 crore to ₹4,092 crore over the same window (Inve data, Mar 2019 vs Mar 2025). Said it, dated it, beat it. Again, an illustration of behaviour, not a recommendation on either name. And one clean record set beside one missed one shows you the two outcomes, not how often each occurs: most histories sit messily in between, and the real base rate is the ~55% kept-rate cited above — against which a deliver-early case like this is the exception, not the norm. (Anant Raj's full guidance-vs-cash record is in its own deep-dive.)

That's the whole field guide in two records. The disciplined allocator's history is boring on the surface and excellent underneath: debt that falls roughly when guided, the occasional small acquisition that actually integrates, cash returned when reinvestment runs out of road. The undisciplined one's is exciting on the surface — bold targets, ambitious capex — and the real story lives in the gap between the dated commitment and the reported number. You can't judge this from one year. You can judge it from a few years of commitments kept or quietly extended, and a decade of retained-earnings outcomes. Both are public.

How do you put it together without overcomplicating it?

The whole framework collapses into one owner's question: if this management had no access to outside capital ever again and could only allocate the cash the business itself generates, would I trust them to compound it? Everything above is evidence for that one answer.

Run the retained-earnings test for the headline verdict. Use incremental return on capital to see what new money earns. Read the acquisition trail for discipline. And track the dated capital-allocation commitments across concalls to see whether the stated philosophy is real or rhetorical. A management that passes all four is rare — and rarity is the point, because the market frequently prices the empire-builder's ambition higher than the discipline that actually compounds.

Where could this be wrong? A young company reinvesting heavily will show poor near-term incremental returns precisely because the investments haven't matured — judge it on a five-year window mid-build and you understate it. A genuinely transformational acquisition can look reckless for years before it works. And a deleveraging miss driven by a real demand collapse, candidly explained, is a different animal from one driven by an empire-building spree — the cause matters as much as the slip. The framework flags where to look hard; it doesn't deliver a verdict by arithmetic alone. A flawless allocation record can't save a stock from a sector collapse either. Discipline raises your odds; it guarantees nothing.

Frequently asked questions