Harbor Coffee, our fictional coffee business, reports $180 million of operating cash flow in FY3. Its cash balance does not grow. Where did the money go?
Three $60 million uses explain it: capital spending, dividends and share repurchases. Nothing went missing.
The harder question is whether those were good choices. A company can account for every dollar and still waste your share of it.
Every dollar has another possible use
Capital allocation is how management decides where the company's money goes. The main uses are reinvestment, acquisitions, debt reduction, dividends and buybacks. Keeping cash is a choice too: it preserves room to act later.
Start with the money already spoken for: maintaining the business, paying bills and debts when due, and keeping enough cash for operating commitments. What remains is discretionary capital — money management can put toward other uses.
Spend $100 on a project and that $100 is no longer available for a payout. Funding both requires more cash or new financing.
A project can make money and still be the weaker choice.
What each choice needs to justify
Judge each choice against what else the money could have done:
| Choice | Benefit to test | Missing evidence |
|---|---|---|
| Reinvest | Preserve or grow cash | Return on new spending |
| Acquisition | Extra earning power | Price and integration costs |
| Debt reduction | Less interest and risk | Terms and cash needs |
| Dividend | Cash for owners | Better uses forgone |
| Buyback | More value per share | Price versus value |
Expansion needs to earn enough on the new money to justify its cost and risk. That is the return test in growth quality. An acquisition must justify both the purchase price and the cost of combining the businesses. A bigger business can be a worse investment.
Debt repayment can save interest and leave fewer loans to refinance. Check repayment fees and the cash buffer left afterward. A risky project promising a higher return does not automatically beat that breathing room.
Share buybacks need a price check: paying more than a share is worth harms the owners who stay, even though fewer shares remain outstanding. A lower share count alone cannot prove success.
A high payout, no payout or rising earnings per share says little without the alternatives. The choice between dividends and buybacks has its own trade-offs; neither always comes first.
Follow the cash once
Harbor's FY3 capital spending, or capex, is $60 million: $40 million for maintenance and $20 million for growth. Dividends and buybacks each use another $60 million. No net borrowing, acquisitions or other cash flows change this picture in FY1–FY3.
Starting with operating cash flow, the arithmetic in millions is:
The zero is the change in cash. Harbor still ends FY3 with $105.4 million, the same balance it started with.
Starting from free cash flow, capex is already subtracted: $180 million minus $60 million leaves $120 million. Harbor's figure is after interest and tax. That $120 million pays the two $60 million distributions. Subtract capex again and you count the same spending twice.
Across FY1–FY3, Harbor's reported operating cash adds up to $511.5 million. Capex, dividends and buybacks take almost equal shares; the sliver is the $5.4 million added to cash.
Grade the record after the spending
A capital-allocation record connects past spending to the benefits it delivered. Compare what management expected with what happened, using the alternatives available at the time. Match results to the promised timetable: spending this year may produce cash only later.
Harbor's three-year capex memo splits the $173 million into $114 million for maintenance and $59 million for growth. Operating margins held at 20%, and company-wide returns on invested capital barely changed. Those averages mix old investments with new ones.
The missing evidence is what the $59 million of expansion spending earned. The case supplies neither the projects' original targets nor their results. You need the extra cash they generated, when it arrived and any further spending needed to finish them. Compare those results with what debt reduction or another project could have offered at the time.
Harbor's FY3 buyback spent $60 million on 1 million shares at $60 each. The $66 year-end quote is a later market price, not evidence that $60 was a bargain. The useful comparison is with a defensible estimate of what a share was worth when Harbor bought it.
Harbor passes the funding check: all uses reconcile and operating cash covered them. Expansion returns and buyback value remain unproven. Before giving management a higher grade, you need those results and the commitments competing for future cash. The stock chart cannot grade these decisions for you.
Try the test on a real disclosure
For an optional 60-second check, open Berkshire Hathaway's 2012 shareholder letter, printed pages 19–20. In the Dividends discussion, read the paragraph beginning "The third use of funds" and the warning at the top of page 20.
Which missing fact prevents a lower share count from proving that a repurchase succeeded?
The missing fact is the price paid relative to a conservative estimate of what each share of the business is worth. Funding matters too: the company must still meet its operating needs and commitments. A shrinking share count can conceal an expensive mistake.
Buffett was explaining his choices at Berkshire in that period, not setting a payout ranking for every company. The next lesson asks what management was rewarded for doing when it chose where to spend.
In short
- Cash is discretionary only after obligations, upkeep and operating needs are covered.
- Free cash flow already subtracts capex; do not charge the same spending twice.
- A good business or project can be a poor purchase at the wrong price.
- Knowing where the cash went does not prove it was well spent.