In Part 1, we learned how to calculate FCFE for a single year. Now we project it forward and turn it into a number you can act on.
In Part 1, we introduced First River Bank and worked through the FCFE formula. We calculated that the bank earned $500 million in net income, needed to retain $100 million to keep its capital ratio intact as it grew, and had $400 million left for shareholders. That $400 million is FCFE — the cash ceiling.
One year of FCFE tells us what the bank generated. But what we want to know is what the bank is worth. For that, we need to answer two more questions: how much FCFE will this bank generate in the years ahead, and what is each future dollar worth in today’s money?
That is the entire job of a DCF — a discounted cash flow model. In this post, we will build one for First River Bank, step by step. By the end, we’ll have an intrinsic value per share. Then we’ll compare it to a hypothetical market price and see what margin of safety, if any, the stock offers.
Where the Growth Rate Comes From
In Part 5 of the Intrinsic Value series, we derived a growth rate for a manufacturer using two inputs: the reinvestment rate and the return on invested capital. The logic was: growth comes from putting money back into the business, multiplied by how productively the business uses that money.
The same logic applies to a bank — just with bank-specific terms.
For a bank, the equivalent of “invested capital” is book equity — the shareholders’ equity on the balance sheet. The equivalent of ROIC is Return on Equity, or ROE — how much net income the bank earns for every dollar of equity it has.
ROE = Net Income ÷ Book Equity
The equivalent of the reinvestment rate is the retention ratio — the fraction of net income the bank keeps rather than paying out. If a bank earns $500M and pays $200M in dividends, it retains $300M. The retention ratio is $300M ÷ $500M = 60%. The payout ratio — what gets distributed — is the remaining 40%.
Put them together:
Growth Rate = ROE × Retention Ratio
This is the fundamental growth equation for a bank. It says: how fast a bank grows depends on how much it earns on its equity and how much of those earnings it plows back into the business.
First River Bank’s numbers:
| Item | Amount |
|---|---|
| Net Income | $500M |
| Book Equity | $5,000M |
| ROE | 10% |
| Dividends Paid | $200M |
| Payout Ratio | 40% |
| Retention Ratio | 60% |
Growth Rate = ROE × Retention Ratio
= 10% × 60%
= 6.0%
First River Bank is expected to grow at 6% per year. That is the number we’ll use for our five-year projection.
The Five-Year FCFE Projection
Now we project forward. In each of the next five years, net income grows at 6%. The bank retains 60% for regulatory capital and growth. The remaining 40% — the payout ratio — represents the FCFE available to shareholders.
This is the bank version of the table we built in IV Part 7 for Maple Ridge Manufacturing.
| Year | Net Income | FCFE (40% payout) |
|---|---|---|
| 1 | $530.0M | $212.0M |
| 2 | $561.8M | $224.7M |
| 3 | $595.5M | $238.2M |
| 4 | $631.2M | $252.5M |
| 5 | $669.1M | $267.6M |
Each year’s net income is the prior year multiplied by 1.06. FCFE is 40% of net income — the slice that, in a well-run bank at its minimum reinvestment level, belongs to shareholders.
The Discount Rate: Why We Use Cost of Equity, Not WACC
In IV Part 6, we built a discount rate called WACC — the Weighted Average Cost of Capital. WACC blends the cost of equity and the cost of debt together, weighted by how much of each the company uses. The idea is that a firm is funded by two groups — equity holders and lenders — and the discount rate should reflect what both groups require.
For a bank, WACC doesn’t work. Here’s why in one sentence: a bank’s deposits are not financial leverage — they are the raw material of the business. Treating deposits as “debt” and blending their cost into a discount rate would be like treating a manufacturer’s inventory as a liability. It misrepresents what the business is.
When we value a bank, we are valuing what flows to equity holders — and only equity holders. The deposit obligations are already reflected in the income statement (the bank pays interest on deposits before we reach net income). The FCFE we calculated is already net of all obligations to depositors. So the only discount rate that makes sense is the one that reflects what equity holders require.
That rate is the Cost of Equity — the same rate we calculated as the first building block of WACC in IV Part 6.
Cost of Equity = Risk-Free Rate + (Beta × Equity Risk Premium)
For First River Bank:
| Input | Value |
|---|---|
| Risk-Free Rate (10-yr Treasury) | 4.25% |
| Industry Beta (regional banks) | 0.70 |
| Equity Risk Premium | 4.72% |
Cost of Equity = 4.25% + (0.70 × 4.72%)
= 4.25% + 3.30%
= 7.55% (call it 7.6%)
Bank stocks typically carry a beta below 1.0 — they are sensitive to interest rate cycles and credit conditions, but their business model is relatively stable compared to, say, a tech startup. A beta of 0.70 reflects that.
First River Bank’s equity holders require a 7.6% annual return. That is the rate we use to discount every future FCFE back to today.
Discounting the Five Years
With the growth projections and the discount rate in hand, we can find the present value of each year’s FCFE. We divide each year’s cash flow by 1.076 raised to the power of that year — the same mechanic we used in IV Part 7.
| Year | FCFE | Discount Factor | Present Value |
|---|---|---|---|
| 1 | $212.0M | ÷ 1.076¹ | $197.0M |
| 2 | $224.7M | ÷ 1.076² | $194.2M |
| 3 | $238.2M | ÷ 1.076³ | $191.5M |
| 4 | $252.5M | ÷ 1.076⁴ | $188.8M |
| 5 | $267.6M | ÷ 1.076⁵ | $186.2M |
Total PV of 5-Year FCFEs: $957.7M
Terminal Value — Everything After Year Five
First River Bank won’t stop operating in year five. It will keep earning, keep paying out cash to shareholders, keep growing — ideally for decades. We can’t project every year individually, so we do what we did in IV Part 7: we capture the value of everything beyond year five in a single number called the terminal value.
The logic is the same. A business that generates cash flows growing at a steady, modest rate forever has a present value equal to next year’s cash flow divided by the discount rate minus that growth rate. We use the risk-free rate — approximately 4.25% — as our long-run terminal growth rate. This anchors the bank’s far-future growth to something it can plausibly sustain: roughly the pace of the overall economy.
Terminal FCFE = Year 5 FCFE × (1 + Terminal Growth Rate)
= $267.6M × 1.0425
= $278.9M
In the stable terminal phase, a bank’s competitive advantages tend to erode — competitors enter, spreads compress, ROE converges toward the cost of equity. To reflect that, we adjust our stable-phase growth and reinvestment assumptions. The terminal payout ratio rises slightly as reinvestment needs fall.
Stable Reinvestment Rate = Terminal Growth Rate ÷ Stable ROE
= 4.25% ÷ 9.0% (ROE converges toward Cost of Equity)
≈ 47%
Stable FCFE = Terminal FCFE × (1 − Stable Reinvestment Rate)
= $278.9M × (1 − 0.47)
= $278.9M × 0.53
= $147.8M
Now apply the Gordon Growth Model — the same formula from IV Part 7:
Terminal Value = Stable FCFE ÷ (Stable Cost of Equity − Terminal Growth Rate)
= $147.8M ÷ (7.6% − 4.25%)
= $147.8M ÷ 3.35%
= $4,412M
Discount that terminal value back five years to get its present value:
Terminal Value PV = $4,412M ÷ 1.076⁵
= $3,070M
Putting It All Together
Now we have all the pieces.
Equity Value = PV of 5-Year FCFEs + PV of Terminal Value
= $957.7M + $3,070M
= $4,027.7M
Divide by shares outstanding. First River Bank has 200 million shares outstanding.
Intrinsic Value per Share = $4,027.7M ÷ 200M shares
= $20.14 per share
First River Bank’s intrinsic value is approximately $20.14 per share.
Now let’s say First River Bank is currently trading at $15.50 per share on the open market.
Margin of Safety = Intrinsic Value − Market Price
= $20.14 − $15.50
= $4.64 per share
Margin of Safety (%) = 1 − ($15.50 ÷ $20.14)
= 23%
A 23% margin of safety means you would be buying the bank at a 23% discount to what it appears to be worth based on the model. That gap — between what the model says something is worth and what the market is currently charging — is the investor’s cushion. If the model is a little too optimistic, a 23% buffer means you may still have not overpaid.
What the Model Tells You — and What It Doesn’t
The bank DCF is a rigorous process. But it is only as good as the net income number at its foundation.
Here’s the risk: banks are cyclical. A bank’s earnings in a boom year — when credit is easy, loan losses are low, and spreads are wide — can look dramatically better than its earnings during a credit crunch. If you plug peak-cycle ROE into this model, you will get an inflated intrinsic value that will look wrong the moment the cycle turns. If you plug trough-cycle ROE from a recession year, you will get a deflated value that understates the bank’s normal earning power.
The model we built assumes First River Bank’s current earnings are representative of its normalized earning capacity. That is a reasonable starting point — but it requires judgment. Before trusting the output, an analyst should ask: is this year’s ROE typical? Or are we near a credit peak, or recovering from a recent shock?
For most banks, this is manageable. You can look at ROE over a full economic cycle — say, 10 years — and average it. That smoothed ROE is a better input than any single year.
For insurance companies, the problem is more acute. A hurricane year, a pandemic year, or a catastrophic event can wipe out several years of underwriting profits in one quarter. You can’t meaningfully value an insurer using last year’s net income if last year included a $2 billion catastrophe loss. That requires a different kind of normalization — which is exactly what we’ll tackle in Part 3.
A Note to Luca and Lili
There’s something I want you to notice about this post that goes beyond the numbers.
We built a model. We ran the math. We got a number: $20.14. And then I immediately told you the things that could make that number wrong.
That is not a flaw in the process. That is the process.
A valuation model is not a truth machine. It is a structured way of making your assumptions explicit. When the model says $20.14, what it really says is: if the bank earns around 10% ROE going forward, retains about 60% of earnings, and grows at around 6% per year, then based on a 7.6% cost of equity, the bank is worth about $20.
Every one of those “ifs” is an assumption. The model forces you to name them. And once you’ve named them, you can ask whether they’re reasonable.
The investors who get into trouble aren’t the ones who use models. They’re the ones who forget that the model is only as good as the assumptions inside it — and who stop asking “what if I’m wrong?”
I want you to always ask that. The margin of safety is your partial answer: even if you’re a little wrong, you haven’t overpaid by much.
— Papa
The One-Sentence Summary
To value a bank, project net income forward using ROE and the retention ratio as growth drivers, discount each year’s FCFE back at the cost of equity, add a terminal value for all the years beyond the projection, and divide by shares outstanding — the gap between that number and the market price is your margin of safety.
Next: Financial Firms, Part 3 — Normalizing Earnings for Insurance Companies. Insurance firms face a unique challenge: a single bad year — a hurricane, a pandemic, a catastrophe — can dwarf years of normal profits. We can’t use last year’s earnings as our baseline. We have to normalize. Part 3 shows you how.
— Jim