Capital Allocation: NPV vs IRR, Hurdle Rates, and Choosing Between Reinvesting, M&A, Debt Paydown, Dividends, and Buybacks

Capital allocation gets discussed as a philosophy, and most of the useful part is arithmetic. A project that returns 42.6% can add less value than one that returns 17.7%. A company can grow earnings 4% a year and be worth $103 million less. And a buyback that looks cheaper than paying off debt at one share price looks worse at another.

This guide gives a decision rule for each of those cases, with checked numbers. It uses a September 2026 backdrop: the 10-year Treasury yield was 5.24% on September 28, up from 4.19% on January 2, so every hurdle rate set at the start of the year is now too low. The discount rate used below, 8.67%, is built step by step in our business valuation guide. Nothing here is investment, tax, or accounting advice.

The five uses of cash

Data on the last two shows what large US companies do. S&P Dow Jones Indices' December 2025 release reported S&P 500 buybacks of $249.0 billion in the third quarter of 2025, up 6.2% after a 20.1% drop in the second quarter, and dividends of $168.1 billion. Over the 12 months to September 2025, dividends set a record $664.9 billion and total shareholder returns set a record $1.685 trillion, of which buybacks were about 60%. Participation fell to 66.6% of companies from 76.8% in the first quarter of 2025, and Apple, NVIDIA, Alphabet, and Meta accounted for over 22% of the quarter's buybacks. That is the most recent S&P release we could verify; later quarterly figures were not available when we checked.

On acquisitions, Bain & Company's 2026 midyear report put 2025 global M&A at $4.9 trillion, up 40%, with deal value up another 41% year over year in the first five months of 2026. How much to pay for one is a valuation question, covered in our acquirer's guide to M&A and the due diligence guide.

Choosing among projects: NPV, IRR, and payback

Three measures dominate, and they answer different questions.

• Net present value (NPV) discounts every cash flow at the cost of capital and sums them. It is the dollar value a project adds.
• Internal rate of return (IRR) is the discount rate at which NPV equals zero. It is a percentage, which makes it easy to compare with a hurdle rate and easy to misuse.
• Payback is the time to recover the investment. It ignores everything after recovery and the time value of money before it, so use it as a liquidity check only.

Illustration: when IRR and NPV disagree

Two mutually exclusive projects, discounted at 8.67%. Project A costs $1,000 and returns $650 a year for 3 years. Project B costs $5,000 and returns $1,300 a year for 7 years. We ran the figures in a script and checked them.

IRR and payback favor A. NPV favors B. If you can only do one of them, B adds $962 more value at 8.67%. NPV ranks A first only at discount rates above 14.5%. That is the IRR of what B adds to A: an extra $4,000 up front, returning $650 a year in years 1 to 3 and $1,300 in years 4 to 7. At a 16% discount rate, A's NPV of $460 beats B's $250.

Rules that follow:

1. For mutually exclusive projects, rank by NPV at your cost of capital. Use IRR to see how much room the project has before the discount rate erases its value.
2. Use the profitability index only when cash is limited and you are choosing among many projects. Under a budget, take projects in descending order of index. In our numbers, project A plus a $2,000 project C (NPV $937) add $1,592 for $3,000, against B's $1,617 for $5,000, so the freed $2,000 has to earn more than $25 elsewhere to make the pair the better choice.
3. Check for sign changes. A project with a later cleanup or decommissioning cost, such as cash flows of -100, +260, -165, can have more than one IRR (here 10% and 50%), and IRR then has no clear reading. NPV does not have this problem.
4. IRR assumes each interim cash flow is reinvested at the IRR itself. The modified IRR, which reinvests at the cost of capital, gives 28.5% for A and 13.1% for B. The gap narrows, but MIRR still ranks A first because it is a percentage earned on a smaller base.

Setting a hurdle rate

The floor for the hurdle rate is the weighted average cost of capital. Damodaran's January 2026 cost of capital data put US non-financial companies at 7.72% overall, 9.34% for software, 7.23% for business and consumer services, and 4.36% for utilities. Those were computed when the 10-year Treasury yielded about 4.2%. With the yield 105 basis points higher in September, rebuilding the illustration's WACC with today's risk-free rate adds about one point, holding the equity premium, beta, and borrowing spread constant.

A few practical rules:

• Use one WACC per business line, not one for the whole company, if the risk differs.
• Add a premium for a specific project's risk, such as a new market or unproven technology, and write down the reason. Do not add one to compensate for optimism in the forecast; fix the forecast.
• Refresh the rate when the risk-free rate moves materially. A hurdle set in January is stale by September in a year like this one.
• Where a project raises capacity or depends on a customer, run the NPV at the hurdle plus and minus two points, the way the valuation guide runs a sensitivity table.

ROIC versus WACC

Return on invested capital (ROIC) is net operating profit after tax divided by invested capital. If it exceeds WACC, the business earns an economic profit on the capital it already has. The more useful question is what the next dollar earns.

Illustration: a business has $44 million of after-tax operating profit on $400 million of capital, so ROIC is 11.0%. At an 8.67% WACC, the spread is 2.33 points and the economic profit is $9.3 million. With no growth, the business is worth $44 million divided by 0.0867, or $507.7 million. Now suppose it grows 4% a year, funded by reinvesting at some return on the new capital. The standard value-driver formula is value = next year's profit x (1 - growth / return on new capital) / (WACC - growth).

At a return equal to WACC, growth adds nothing. Below it, growth destroys value even while revenue and earnings per share go up. This is the test to apply to any plan that says "growth" without a return: what does the incremental dollar earn, and is that above the hurdle?

Debt paydown versus buybacks

Paying down debt earns the after-tax interest rate with no market risk. Suppose the company borrows at the 10-year Treasury yield plus 2.50 points, which is 7.74%. At a 25% tax rate that costs 5.8% after tax. Prepayment terms matter, and our business debt guide covers fixed versus floating rates and covenant headroom.

A buyback earns the earnings yield, which is one divided by the price-earnings ratio:

Against a 5.8% after-tax debt cost, a buyback beats paying off debt on earnings yield only when the P/E is below about 17. That is a simplification, since a buyback can also be sensible if the stock is worth more than its price, and paying off debt reduces flexibility if the company later needs to borrow. But a repurchase at 30 times earnings while carrying 7.74% debt needs a specific reason. Also note that the earnings yield ignores growth, so a fast-growing company's real return on a buyback can be higher, but that is a bet on the growth rather than a certain return.

The 1% excise tax

Under Internal Revenue Code section 4501, a covered corporation, meaning a domestic corporation whose stock trades on an established securities market, pays 1% of the fair market value of its stock repurchased during the tax year. The base is reduced by the fair market value of stock the company issues during the year, including stock issued to employees. Exceptions include total repurchases of $1 million or less, repurchases contributed to employer retirement plans and ESOPs, repurchases by dealers in the ordinary course, repurchases by regulated investment companies and REITs, and repurchases treated as dividends.

Illustration: a company repurchases $500 million of stock and issues $120 million to employees through equity awards. The base is $380 million and the tax is $3.8 million. The S&P release estimates the tax reduced S&P 500 operating earnings by 0.36% in the third quarter of 2025 and by 0.40% over the 12 months to September 2025, so it is a small drag on the aggregate and a real but minor cost to a single decision. Companies that issue a lot of stock compensation can offset much of it. It does not change the ranking of buybacks against paying down debt unless the margin is thin.

Dividends, and how to choose

A dividend is the residual: cash the business cannot deploy at or above its hurdle rate. It tells shareholders the company has no better use, and it commits the board to a payout investors will expect to continue. A buyback is flexible, since the board can pause it, but it taxes shareholders only when they sell, and the company pays the 1% tax. The dividend investing guide covers what shareholders do with the cash.

A workable order of decisions:

1. Fund maintenance spending and any project with an NPV above zero at the hurdle rate, ranked by NPV, or by profitability index if cash is rationed.
2. Look at acquisitions only after checking whether the same money would earn more in the first step. Price them against synergies, as our M&A guide does, and plan for integration, using the post-merger playbook.
3. Keep liquidity and debt at a level the business can survive a bad year with. See the cash flow and working capital guides.
4. Compare paying down debt with buying back stock on after-tax yield, and compare both with intrinsic value from a DCF.
5. Distribute what is left, choosing the form by how stable the surplus is: a dividend for a steady surplus and a buyback for a variable one.

Audit your past decisions

Capital allocation improves through review. Once a year, list the largest projects and acquisitions from three to five years ago, restate the forecast NPV, and set it beside what happened. Look for patterns: forecasts that ran optimistic in one division, synergies that never arrived, projects that missed the hurdle by more than the sensitivity range. Then adjust either the forecasting method or the hurdle premium for that division. Growth and private-equity-owned companies face the same trade-offs with a different capital structure, discussed in the private equity value creation guide.

This guide is for informational purposes only and is not investment, tax, or accounting advice. Consult a qualified finance professional and tax adviser before making capital allocation decisions.