Not all earnings and book values are created equal. Certain companies and industries use much more aggressive or much more conservative accounting practices than others.
A given company can show substantially different earnings for a given year depending on the depreciation methods (accelerated vs. straight-line depreciation, 30-year vs. 5-year depreciation), inventory accounting methods (LIFO vs. FIFO vs. others), and other arbitrary accounting practices. Also, one-time factors (such as an asset sale or acquisition) can temporarily boost or suppress earnings.
In theory, book value is supposed to represent the company's liquidation value. In practice, book value doesn't mean much, and if the company is prosperous, book value is meaningless. Book value is subject to the same arbitrary accounting practices as earnings. If a company goes bankrupt, the assets usually sell for far, far less than book value. Book value tells you nothing about what assets a company has or even how leveraged it is. Some companies are rich in liquid assets and need very little property, plant, and equipment to operate while other companies have most of their assets tied up in property, plant, and equipment.
While a company cannot fake a dividend, the dividend it pays may not be sustainable and can be cut or eliminated at any time. Dividends are NOT free money that falls out of the sky - they come from the company. Dividends are just a taxable liquidation of a piece of the company. John Train calls unsustainable dividends a "taxable return of capital".
In theory, return on equity (earnings divided by book value) shows the quality of a company. In practice, it suffers from the same problems as the earnings and book value yardsticks it's based on and many more. Using return on equity makes leveraged companies look better at the expense of liquid companies.
In theory, return on assets (earnings divided by assets) puts leveraged and liquid companies on a level playing field. In practice, it's not much better. Using return on assets does not distinguish between companies that have most of their assets tied up in property, plant, and equipment vs. companies that are rich in liquid assets and only need a modest amount of property, plant, and equipment to operate.
I have devised a better method of appraising companies. If you are a computer geek, think of it as Graham-and-Dodd 2.0. If you are a weather weenie, think of it as Doppler Value Investing.
Please note that my method has its limitations. My method is for industrial companies and only shows what happened in the past, not what will happen in the future.
The formula for cash flow is:
Cash Flow
= Net Revenue
- Cost of Goods Sold
- Selling/General/Administrative Expenses
+ Depreciation/Amortization
+ Adjustment for decrease in receivables
+ Adjustment for decrease in inventory
+ Adjustment for decrease in prepaid expenses
+ Adjustment for decrease in other current assets
Cash flow is independent of depreciation/amortization practices. Since depreciation/amortization is part of "Cost of Goods Sold" or "Selling/General/Administrative Expenses", we're simply taking this out of the equation.
Cash flow does not reward companies for letting the receivables grow. If a company sells a product for $X, the net revenue increases by $X, and this boosts earnings. If the company does not receive cash for it right away, then $X is added to receivables. Cash flow is not affected, because any increase in receivables is deducted when reconciling net revenue to cash flow. When the company does receive the $X payment in cash for the sale, $X is added to the cash flow and subtracted from receivables but earnings are not affected. In extreme cases, companies have gone under as a result of extending excessive credit (receivables) to customers and then being unable to collect on it. In some cases, companies have concealed problems by doing this to boost earnings. Looking at cash flow instead of earnings would have revealed this hidden time bomb.
Cash flow does not reward companies for letting the inventory pile up. If it costs a company $X to buy or produce a product to sell, the cash flow is reduced by $X, but nothing is added to the "Cost of Goods Sold" until AFTER the product is sold, and that means that earnings remain the same even though $X of cash was spent. Depending on the method of inventory accounting (LIFO, FIFO, average cost, etc.), the amount of money added to the "Cost of Goods Sold" when the product finally does sell can be substantially different from $X. Cash flow is independent of the inventory accounting method, because any increase in inventory is subtracted in the process of reconciling revenue to cash flow, and any decrease in inventory is added instead. In extreme cases, companies have gone under as a result of letting the inventory pile up and then being unable to sell it at a good price. In some cases, companies have concealed problems by doing this to maintain earnings. Looking at cash flow instead of earnings would have revealed this hidden time bomb.
Cash flow does not reward companies for letting the prepaid expenses pile up. If a company prepays $X in expenses to a vendor or supplier, the cash flow is reduced by $X, but nothing is added to the "Cost of Goods Sold" or "Selling/General/Administrative Expenses" UNTIL the company receives the product or service it paid for early. This means earnings stay the same even though $X of cash was spent. If the supplier/vendor goes under before the company receives the product/service, then the company suffers from a setback. While I haven't heard of companies that went under because of this, I'm sure it could happen.
I exclude nonoperating expenses and income when calculating cash flow. Nonoperating earnings are usually part of the company's investing or financing activities.
I exclude interest (whether the company pays it or receives it), because I consider this part of the financing activities.
I exclude income taxes, because the same one-time non-operating events that can temporarily boost or suppress earnings do the same thing with income taxes. It's difficult to figure out what part of the income taxes is due to the one-time non-operating event and what part is due to the cash flow.
I exclude changes in current liabilities. I consider this part of the financing activities. Whether the company owes money to a supplier/vendor, a bank, or bondholders, it owes money. If a company borrows lots of money from suppliers/vendors and avoids paying them back, cash flow (according to the classic definition) is boosted. However, this is not a sustainable situation. Thus, I don't add increases in current liabilities to net revenue when calculating cash flow.
You could subtract actual capital spending figure from cash flow to get free cash flow, but that would make contracting companies look better at the expense of growing companies.
Suppose a company begins the year with A dollars in fixed assets. It makes B dollars in capital expenditures, C dollars in capital divestitures, and ends the year with D dollars in fixed assets. If we had magical fixed assets that never wore out and never became obsolete, we would end the year with A+B-C dollars in fixed assets. In reality, some of the fixed assets do wear out or become obsolete, and we only have D dollars (which is less than A+B-C dollars) in fixed assets. The difference, or $(A+B-C-D) represents normalized capital spending. The formula is:
NCAP(Y)
=PPE(Y-1)
+CAP_EXP(Y)
-CAP_DIV(Y)
-PPE(Y)
where NCAP is normalized capital spending, PPE is plant/property/equipment at cost, CAP_EXP represents capital expenditures, and CAP_DIV represents capital divestitures. Y represents the current year, and Y-1 represents the previous year.
The problem with the above formula is that companies make acquisitions or divestitures, and you don't know exactly how much PPE was obtained or shed as a result of the transaction.
Typically, a company has to replace about 5%-10% of its PPE per year to maintain its operations. To be conservative and simple, I assume that a given company has to replace 10% of its PPE each year. Thus, the formula for normalized capital spending is: NCAP(Y)=.1*PPE(Y-1)
Remember to use the PPE figure to use is AT COST. Please ignore the depreciation, because companies all use different methods and schedules.
I got this idea from Warren Buffett and John Train, who have written at length about why capital-intensive businesses are bad and why companies that can generate lots of cash from a modest capital investment are wonderful.
To smooth out the returns, just use a moving average of the Return on PPE to get the Smoothed Return on PPE.
Think of the Smoothed Free Cash Flow as the Alternate Earnings.
I got this idea from Peter Lynch and John Train. They said that if you buy a stock that sells for $10/share, and the company has $10/share of cash sitting in the treasury, then you're getting the business for free.
Use the Alternate Earnings Yield as a substitute for the conventional earnings yield. Note that if the company is priced at the estimated intrinsic value, the Alternate Earnings Yield will be 10%. If the Alternate Earnings Yield is 15% or greater, the stock is a screaming bargain.