A.6.4 Dilution & Buybacks

Basic Prerequisites: 02 · Modigliani–Miller and 03 · Debt, Equity & Seniority.

---

1. Intuition & Practical Objective

The number of shares and how it changes is where capital structure meets shareholder value. Dilution and buybacks are the two levers that change the share count, and both are routinely misunderstood through the lens of the single most-gamed accounting number in finance: EPS = net income ÷ shares outstanding.

The objective of this page is the discipline that separates the arithmetic from the economics:

• Dilution is a wealth transfer, not just a math problem. When a firm issues new shares, it sells part of the residual claim. If it issues at a price below what the shares are intrinsically worth, value flows from old holders to new holders - this is the exact mechanism of the Myers–Majluf “underinvestment” trap in 06.
• A buyback’s EPS accretion is mechanical, not value-creating. Retiring shares raises EPS by construction ($NI$ ÷ fewer shares). But the cash spent was an earning asset; if it was earning at the cost of capital, the buyback destroys exactly the earnings it appears to create. The honest test is ROIC: does the firm have projects that beat the cost of capital? If yes, reinvest; if no, pay out (Jensen 1986’s free-cash-flow point).
• Dividends and buybacks are the same value decision. Both return cash to shareholders. Which is better is a tax and signaling question, not a value question - under MM’s frictionless world, payout policy doesn’t change firm value at all (irrelevance), and the real question is only whether the cash should leave at all (agency - see 06).

---

2. Mathematical Ground Truth & Derivations

EPS mechanics. With $NI$ net income and $n$ shares outstanding,

$$\text{EPS}=\frac{NI}{n}.$$

A buyback of $b$ shares (cash spent $=b\cdot P$) leaves $n-b$ shares, so ceteris paribus $\text{EPS}=\frac{NI}{n-b}>\frac{NI}{n}$ - EPS always rises from a buyback. The trap is the ceteris paribus: if the $bP$ of cash was generating earnings at the firm’s return on assets $a$, then true net income falls to $NI-bP\cdot a$, and the honest post-buyback EPS is $\frac{NI-bPa}{n-b}$. Whether EPS truly rose depends on $a$ vs. the return the market required on the equity retired - i.e., vs. the cost of equity, which is exactly where ROIC enters.

The payout-vs-reinvest decision (Jensen 1986). Define free cash flow as cash flow in excess of what all positive-NPV projects require. For a firm with free cash flow $FCF$ and required return $\rho$ (its cost of capital), investing the cash in a project returning $g<\rho$ destroys value:

$$\text{NPV of reinvesting}=\underset{\text{PV of the weak project}}{\underbrace{\frac{FCF\cdot g}{\rho }}}-FCF<0.$$

The value-maximizing choice is to pay the cash out (dividend or buyback) when $g<\rho$, and to reinvest only when $g\ge \rho$. This is the dividend-policy-vs-ROIC rule: the payout decision is governed by whether internal reinvestment earns above the cost of capital, not by any virtue of paying dividends per se.

Dilution as wealth transfer. Let firm value be $V$ with $n$ shares (intrinsic value $V/n$). Issuing $m$ new shares at price $p<V/n$ (as under information asymmetry, 06) raises $mp$ of cash. New value and per-share value:

$$V^{\prime }=V+mp,\text{per-share}=\frac{V+mp}{n+m}.$$

Old holders’ stake becomes $\frac{n}{n+m}{V}^{\prime }$; the shortfall $\frac{V}{n}$ vs. per-share is a transfer from old to new holders. When that transfer exceeds the project’s NPV, old holders are worse off by investing - the Myers–Majluf underinvestment result.

---

3. Computational Implementation - dilution transfer and buyback EPS, stdlib only

Runs on the standard library. First quantifies the wealth transfer when shares are issued below intrinsic value (the dilution risk); then shows a buyback’s EPS accretion and the ROIC test that exposes it as arithmetic, not value.

The dilution panel is the red flag: issuing at $8$ when shares are worth $10$ hands $40$ of old-holder value to new holders - exactly why managers resist equity issues (bad-news signal) and why dilution risk matters. The buyback panel is the EPS illusion: the naive EPS jumps to $1.556$, but once you subtract the earnings the spent cash would have produced, the “accretion” mostly evaporates ($1.478$) - no value was created by the repurchase itself.

---

4. Failure Modes & First-Principles Breakdowns

1. Celebrating EPS accretion as value. A buyback that raises EPS while the repurchased cash earned at the cost of capital creates zero shareholder value - it only changed the denominator. The failure is reading the arithmetic without the ROIC test (did the firm have a positive-NPV use for the cash?).
2. Missing dilution as a wealth transfer. Issuing below intrinsic value silently moves value from old to new holders. The “dilution” is not the share-count math (that’s inevitable when raising equity) - it is the gap between issue price and intrinsic value. Under information asymmetry that gap is the whole problem (Myers–Majluf).
3. “Buybacks are always good” / “dividends are always good.” Both are ways to return cash; the correct question (Jensen 1986) is whether the cash should leave at all. Paying out when ROIC > cost of capital (plenty of good projects) destroys growth; hoarding when ROIC < cost of capital destroys value through weak reinvestment.
4. Forgetting that payout policy is value-irrelevant under MM. Dividends vs. buybacks differ in taxes and signal content, not in the underlying value transfer. A firm can change its payout form endlessly without creating value - only the decision to return vs. reinvest cash against ROIC creates or destroys it.
5. Ignoring the signaling read of both actions. Because issuing is a bad-news signal and buybacks/dividends a good-news signal (asymmetric information), the announcement moves price even when the cash-flow economics are neutral - a real, if information-driven, channel the arithmetic alone misses.

---

5. References

• Myers & Majluf, “Corporate Financing and Investment Decisions…” (JFE, 1984)
• Jensen, “Agency Costs of Free Cash Flow…” (AER, 1986)
• Brealey, Myers & Allen, Principles of Corporate Finance
• Graham & Dodd, Security Analysis (6th ed.)
• Damodaran, Applied Corporate Finance

---

6. Connected Graph Bridges

• Back: 03 · Debt, Equity & Seniority · Index Hub
• Forward: 05 · Failure Modes · 06 · Advanced Extensions
• Valuation link: Equity Valuation - DCF & payout · Core Financial Ratios - ROIC