Stop using P/E ratios! Use EV / Unlevered FCF.
Apologies to readers for not completing OA’s tutorial on P/E ratios sooner. I promised Part 3 last summer, and here it is March…But this is a good thing folks, because the subject is now far more topical.
I hope readers don’t mind the length of this post. Can’t cut corners trying to explain the proper way to value a company. I think you’ll find more explanation is worth it.
Here’s a video from Yahoo’s TechTicker. In it, a hedge fund manager argues that, even at $7, GE still isn’t cheap. He notes that if you add total debt to the market value of the company’s equity, GE is still valued at 15x-16x more than its earning power. But wait a minute—a novice investor might interject—the company earned $1.72 per share last year. With the stock at $7.00, the P/E ratio is 4x. Which is it, 4x or 16x? What gives?
The problem with using P/E is that it is a totally incomplete way to measure the value of an entire company. The method alluded to by the hedgie in the video is what I will spend the rest of this post explaining.
The “P” in P/E is just the stock price, which is the value of one share of the company’s equity. Equity and nothing else. What about the rest of the balance sheet? (unless you’re totally familiar with the concept of a company’s equity value, please read the prior 2 parts of this tutorial series before moving on.)
We’ll be using GE’s financial statements as part of this example. I recommend readers open page 72 of the recently published 10-k in another browser window so you can toggle back and forth with this post.
Regular readers of OA know how much I’ve been focused on The Accounting Equation lately. Assets = Liabilities + Equity. For heavily-indebted companies whose assets are declining in value very quickly, equity can in fact be worth less than zero. That is, a company can owe more debt than its assets are actually worth. Just like a homeborrower can be upside-down on his mortgage.
So a stock that is around $1 may appear cheap when it really isn’t. If the company’s equity value is below $0, as is the case for all the major banks and insurance companies, then the stock is worthless. Citi’s stock only has positive value to the extent that investors believe there’s still a possibility the government will absorb the bank’s toxic assets. But the prospect of a government “Bad Bank” is growing more remote by the day—thank heavens—so the stock is bleeding all the way to $0.
So if the “P” in P/E ratios is an incomplete, is there a better way to measure company valuation? Yes, it’s called “Enterprise Value,” or “EV” for short. Again, the problem with “P” is that it measures equity value only. When you buy a company, you aren’t just buying it’s equity. You’re buying the whole balance sheet. And that includes the company’s obligations. And its assets….
[Aside: If you multiply the whole ratio, Price / Earnings, by total shares outstanding, we get Total Market Capitalization / Total Net Income.]
The idea behind EV is to measure the true value of an entire company, what a buyer would pay to own the whole shebang. Here, an example will make it pretty simple: a house. Say I want to buy that nice 3 Bedroom/2 Bath house on the corner. The price is $450,000 and the owner still has $400,000 left on his mortgage. The cost to buy the house isn’t just the value of the owner’s equity, it’s also the value of the mortgage. It’s the same for a company…
House’s Value = Owner’s equity + Mortgage
Company’s Enterprise Value = Market value of shareholder’s equity + Obligations
But wait! Our formula isn’t complete. When you buy a company, it also comes with a chunk of assets, and those assets have value. Going back to our house example: what if the seller is a bit foolish and decides to leave a very valuable antique collection worth $50,000 in the basement, along with a $25,000 pile of cash on the floor in the living room. You find all this when you move in. In other words, you may have paid $450k for the house, but there’s $75k of assets that come as part of the deal.
So you’re all-in cost to buy the house isn’t $450k. You decide to sell the antiques and pocket the cash, which reduces the cost of the deal by $75k. So our revised formula is this:
House’s Value = Owner’s Equity + Mortgage – Cash – Antiques. When you purchase the house, the lower it’s value, the less you pay. So the bigger the cash pile, the more you subtract from your upfront cost.
Companies have bank accounts and other assets that have value independent of the business itself. We deduct these when calculating EV.
EV = Market Value of Shareholder’s Equity + Company Obligations – Cash – Long-Term Investments. The more cash and investments on the balance sheet, the lower my cost to buy the company, which is a good thing.
Here are the four components of the EV calculation:
- Market Value of Shareholder’s Equity = Total Shares Outstanding * Share Price. Also called the “market capitalization” or “market cap”
- Company Obligations. This includes short-term and long-term debt, pension obligations and minority interest. Look at the liabilities section of the balance sheet and add all of these up in order to get total company obligations.
- Cash. This one is easy. Just go to the asset section of the balance sheet and look at the top for “cash” and/or “short-term investments.” Add these up and subtract them.
- Long-Term Investments. The company may have long-term investments in which it has a stake. (Yahoo, for instance, owns large chunks of Yahoo Japan and Alibaba Group in China. Like the antiques in the house example above, these investments also have value for an acquirer. Appraise their current value and determine what they can be sold for.) As with cash, add up the appraised value of a company’s long-term investments and subtract them.
Again, we have:
Enterprise Value = Market Cap + Debt/Oligations – Cash/ST Investments – LT Investments.
Let’s wrap up this section using GE as an example (see pages 72 and, for total shares outstanding page 97, of the 10-k):
- GE’s market value = (10.1 billion shares outstanding * $7.00 price per share) = $70.7 billion
- GE’s obligations = $193.7 billion short-term debt + $330.1 billion long-term debt + $8.9 billion minority interest = $532.7 billion
- GE’s Cash = $48.2 billion cash + $41.4 billion short-term investments + $0 long-term investments = $89.6 billion
GE’s Enterprise Value = $70.7 billion + $532.7 billion – $89.6 billion = $513.8 billion
Compare that figure to the company’s market cap of $70.7 billion. Using just market cap to measure the value of the entire company would be like valuing a $517,000 house at only $71,000 because that’s the value of the existing owner’s equity. Like I said above: incomplete…
UNLEVERED FREE CASH FLOW
Now we’ve established EV as a more complete formula to replace the “P” in “P/E.” But what about the “E?” The reason for computing P/E in the first place is to get a feel for how expensive the company is relative to the profits it generates. The higher the P/E, the more I’m paying per dollar of earnings. (again, see Part 2)
It makes sense to measure earnings against the market cap. If I own the company’s equity, I own whatever profits are left AFTER the company has met its obligations. (again, see Part 1)
But if I own the whole company, I also own its obligations, which means I may have some control over the mix of equity and debt used to finance operations. If readers will pardon the use of a technical term, what I need to replace the “E” in the denominator is a measure of earnings that is capital-structure neutral. In other words, what is the cash being generated by the business before I have to pay things like interest on my debt?
That’s easy, it’s the company’s unlevered free cash flow. Despite the ugly name, it’s not hard to calculate. Just go to the cash flow statement in the company’s financials (page 74 in GE’s 10-k). This statement is separated into three parts, cash flow from operating activities, cash flow from investing activities, and cash flow from financing activities. For now, we’re only worried about the first two.
What is the free cash flow generated by a company’s operations? Easy, just look at the last line of the “Cash Flow From Operating Activities” section. That total is, wouldn’t you know, cash flow from operations listed at the bottom. This number is very easy to understand conceptually. We start with a company’s net income, which is the same as “E”, and add back all of the non-cash charges that have been deducted. Stuff like depreciation and amortization. Luckily, you don’t have to do this because it’s all done for you on the cash flow statement.
But there are two adjustments you have to make to cash flow from operations in order to get a complete, and capital-structure neutral, picture of the company’s cash earnings power. First you have to deduct capital expenditures. CapEx is what the company spends on large equipment that is depreciated over time. It’s okay to back out depreciation from net income to determine cash flow, but it’s not okay to ignore costs associated with running the business.
A quick anecdote. I remember reading an investor newsletter in which a “value-oriented” manager argued that Home Depot was cheap at around $40 per share. He said it was only trading at 6x cash flow. That WOULD have been cheap. If only it were true. Trouble is, to calculate his valuation ratio of 6x, he used an incomplete figure in the denominator. EV was in the numerator and Cash Flow from Operations was in the denominator. His problem was that this method totally ignored Home Depot’s cost for building new stores, which is considered a “capital expenditure” not an “expense.” The difference is how the cost is treated for accounting purposes.
Most business expenses—salaries, utility bills, cost of goods sold—are expensed as they are incurred. These are deducted, dollar-for-dollar, from net income. But cash spent on a “capital asset” like a new building is considered an investment that will pay off over time. So accountants came up with a concept, depreciation, by which the cost of the asset is expensed over its useful life, not all at once when the asset is purchased.
If a new Home Depot supercenter costs HD $100 million to build, but lasts 20 years, then HD’s “depreciation expense” is $4 million per year over 20 years. The actual cash to buy the store is spent up front. “Depreciation expense” is just a way of allocating that investment over the time during which the asset will be in use. In other words, depreciation is a “non-cash expense.” We deduct it from earnings to determine accounting profits, yes. But form a free cash flow perspective it’s meaningless.
Recall above how cash flow from operations adds back non-cash expenses in order to determine the cash flow generated by the business? As it should be. But we still have to account for the money HD spends building new stores.
And this was the problem with the hedge fund manager’s valuation technique. His free cash flow number added back non-cash depreciation expense, but it did NOT deduct capital expenditures. That makes no sense of course. For HD to increase earnings for shareholders, it builds new stores. The cost of building those stores has to be accounted for. And it is, as a “capital expenditure” on the CF statement under “cash flow from investing activities.” (It might also be called “additions to propert and equipment”)
So this is the first adjustment to our formula for FCF:
FCF = Cash Flow from Operations – Capital Expenditures
We need to make one more quick adjustment. I mentioned above that we’re interested in valuing the whole company from an owner’s perspective, which means I may have some control over the mix of equity and debt used to finance operations.
Pretend I’m buying that $450k house in the example above and I have enough cash to pay for the whole thing. I can use all cash if I want, or I can mix in some debt by taking on a mortgage. Similarly, a big corporation might be able to buy out a small business with all cash, or a mix of cash and debt if that is preferable. (Adding debt gives the company more leverage, which gooses return on equity.) Anyway, the point is, interest expense is something that happens AFTER choosing to fund part of the business with debt. We want a “capital-structure neutral” measure of cash flow because we’re valuing the business BEFORE determining whether to fund our acquisition with debt.
That’s easy. Just add back interest expense to FCF. Don’t forget to adjust for taxes. Because interest is tax deductible, if we stop paying interest, we stop saving money on our taxes. The formula is now complete:
Unlevered FCF = Cash Flow from Operations – CapEx + [Interest Expense * (1 - company's tax rate)]
As we did with the first section, let’s wrap up using GE as an example (see page 74 of the 10-k):
- Cash Flow from Operations = $48.6 billion
- CapEx = $16.0 billion
(see page 69 for the next two figures)
- tax rate = ($1.1 billion taxes paid) / ($19.1 billion pre-tax income) = 6%
- After-tax interest expense = $26.2 billion interest expense * (1 – .06) = $24.6 billion
Unlevered FCF = $48.6 billion – $16.0 billion + $24.6 billion = $57.2 billion
Compare $57.2 billion of unlevered free cash flow with net income of $17.3 billion…
Now for the payoff of this ridiculously long blog post:
- P/E ratio = 4.2x ($70.7 billion market cap / $17.3 billion net income = 4.1)
- (divide by 10.1 billion shares outstanding yields same result = $7.00 per share / $1.72 = 4.1x
- EV/Unlevered FCF = 9x ($513.8 billion / $57.2 billion = 9)
As you can see, when measured properly, GE is twice as expensive as its P/E ratio would appear to indicate.
As you can see, it’s worth putting pen to paper to do this calculation. When considering an investment, the company’s STOCK value is less important than the company’s ENTERPRISE value.
The hedgie in the video referenced at the top says GE’s valuation is 15-16x cash flow. I suspect he’s excluding after-tax interest expense. ($513.8 billion / $32.6 billion = 15.8x)