You will be able to choose a valuation method for banks and REITs and explain why free cash flow does not fit them.
Marcus tried his Larkspur DCF on his local bank and gave up within ten minutes. When he tried the same method on his S-REIT, he hit the opposite problem.
A standard DCF based on free cash flow fails for both banks and REITs, for opposite reasons. At a bank, operating cash flow swings by billions from year to year as deposits come in and loans go out, and capital spending is tiny, a rounding error. A REIT pays out almost everything it earns, so there is little to forecast except the payout itself.
An industrial company borrows to fund its operations, and its debt sits apart from the business. At a bank, borrowing is the business itself. A bank takes deposits, which are debts it owes to customers, and lends them out at a higher rate, so debt is its raw material. Deposits, wholesale funding and loans all move with the bank's operations. There is no clear line between operating cash flow and financing cash flow, so free cash flow to the firm stops meaning anything for a bank.
Analysts value bank equity directly instead, from what shareholders receive or earn. They discount it at the cost of equity rather than at WACC.
Say Marcus's bank has a book value of S$20 a share, an ROE of 12%, a cost of equity of 9% and long-term growth of 3% (all example figures).
The dividend discount model, in its simplest version, sets value per share equal to next year's dividend divided by (cost of equity - long-term dividend growth). It suits banks better than most companies for two reasons. Banks pay steady dividends, and regulators require banks to hold capital in proportion to their loans, which ties growth to retained profit.
Start with the retention ratio, which is growth divided by ROE: 3% / 12% = 25%, so the bank keeps a quarter of its profit. The payout ratio is 1 minus the retention ratio, which is 75%. Earnings per share are ROE times book value per share: 12% x S$20 = S$2.40. A 75% payout gives a dividend of S$1.80, and the value is S$1.80 / (9% - 3%) = S$30.
Then rerun the model with the cost of equity one point lower and one point higher. At 8% the value is S$36. At 10% it is about S$25.70. Because the dividend is assumed to grow for ever, the answer is very sensitive to the cost of equity: a one-point change moves the value by roughly a fifth either way.
The excess return model values the bank as its book value plus the present value of the profits it earns above what shareholders require. Excess return per share is (ROE - cost of equity) x book value per share. For Marcus's bank, 12% - 9% is 3 points on S$20, which comes to S$0.60 a share, growing at 3%. Value that stream as excess return / (cost of equity - growth): S$0.60 / 6% = S$10. Add book value and you get S$20 + S$10 = S$30.
That is the same answer as the dividend discount model, but it shows where the value comes from: S$20 of book value and S$10 from earning more than shareholders require. It also explains why bankers and analysts quote price to book so often, and why price to book misleads when used on its own. A bank whose ROE equals its cost of equity earns nothing above what shareholders require, so it is worth about book value, a price to book of 1. A bank whose ROE is above its cost of equity is worth more than book.
Justified price to book is (ROE - growth) / (cost of equity - growth). For Marcus's bank, 12% - 3% is 9 points and 9% - 3% is 6 points, which gives 1.5 times book. Multiply 1.5 by the S$20 book value and you get S$30 again.
Marcus's bank trades at S$32, which is 1.6 times book. If the bank can keep its ROE at 12% or more, that price is cheap. If ROE drifts down to 9%, the fair price to book falls to 1 and the value falls to S$20, which makes S$32 expensive. Price to book only means something when you read it next to ROE. The question to ask about any bank is whether it can keep earning its ROE through a credit cycle. Module 9, lesson 9.6, "Industry frameworks: banks, insurers, commodity producers and software", lists the measures that show this.
S-REITs distribute most of their taxable income to unitholders, which lets them avoid tax at the REIT level. Dividend stocks and S-REITs, lesson 4.2, "The payout rule and tax transparency", explains how this works. Because a REIT retains so little, there is little to forecast except the payout. REIT growth comes mainly from rent increases, new acquisitions funded by new units or new debt, and changes in property values. Free cash flow to the firm still exists for a REIT in theory, but in practice two methods do most of the work: valuing the distributions, and comparing the price with NAV.
The steps here add only the valuation. Before you trust any distribution you discount, check that it is safe: gearing, interest cover, debt maturity profile, occupancy, rent reversions and weighted average lease expiry. Dividend stocks and S-REITs, module 5, "Read a REIT's numbers like a lender would", teaches each of these and gives you the scorecard to run first.
Distribution valuation uses the same formula as the bank dividend discount model: next distribution / (required return - growth). Marcus's REIT pays S$0.12 a unit and trades at S$2.00, a yield of 6%. He expects distributions to grow 1% a year and wants a 7% return. The next distribution is 0.12 x 1.01, about S$0.121. Divide that by 7% - 1%, which is 6%, and the value is about S$2.02. At a required return of 7.5% it is about S$1.86, and at 6.5% it is about S$2.20.
Price to NAV is the unit price divided by net asset value per unit, and NAV is built from property valuations. Dividend stocks and S-REITs, lesson 5.4, "NAV per unit and what price to NAV tells you", explains how. Marcus's REIT has a NAV of S$2.30 per unit, so at S$2.00 it trades at about 0.87 times NAV.
Use the two methods together. A REIT on a low price to NAV whose distributions do not support even the current price usually has a reason for its discount, and the module 5 scorecard is where you look for it.
For the activity, value one STI bank with the simple dividend discount model above, using its latest dividend, book value and return on equity, then rerun it at a cost of equity one point higher and one point lower.
Value one STI bank with a simple dividend discount model and note how much the result changes when you change the discount rate.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).