Note the differences between net income and cash flow from operations over the last one to five years. If net income closely approximates cash flow from operations, then there is less likelihood that it is being manipulated. However, if net income is consistently higher (more than 30 percent) than cash flow from operations, this may be a sign that management is managing earnings.
Evaluating Whether Management Manipulates Earnings
One objective way to gain insight into the integrity and character of a business’s management team is to learn whether they employ conservative or liberal accounting practices. A company’s accounting books are written by its managers, and the accounting practices they employ are, to a degree, a reflection of the kind of people they are. If the accounting is too complicated to understand, this may be a signal that management does not really want you to understand how the business makes money or how well they are running the company.
Management can manipulate earnings in several ways. The degree of manipulation varies from minor changes to outright fraudulent transactions. Earnings are most often manipulated to cover up deteriorating earnings within the core business. Other times, management wants to meet the quarterly expectations of Wall Street analysts by shifting earnings from good years to bad years or by shifting future earnings to the present.
There are several ways that management manipulates earnings despite being within the technical bounds of what GAAP accounting standards consider acceptable. Look for any (or all!) of the following:
Improperly inflating sales
Under- or over-stating expenses
Manipulating discretionary costs
Changing accounting methods
Using restructuring charges to increase future earnings
Creating reserves by manipulating estimates
Let’s take a look at each of these manipulations.
Inflating Sales
Management can book a sale before revenue is actually earned by incentivizing customers to take more product than they need. This is most likely to happen when the business sells its products through distributors or resellers, when it has a few large customers, or when a sales representative needs to hit a quarterly sales number.
For example, Telecom equipment company Riverstone Networks inflated its sales numbers in 2006 by improperly recognizing revenues. In this case, Riverstone had side agreements with its customers saying payments to Riverstone were contingent upon resale of Riverstone’s products, but Riverstone booked them as sales even though they didn’t actually receive payment. The Securities and Exchange Commission’s (SEC) complaint against Riverstone cited almost $30 million in fraudulent revenues over several quarters, with overstatements ranging from 14 percent to more than 20 percent of revenue during each quarter.1
You can identify methods used to accelerate revenues by looking for large increases in accounts receivable growth compared to sales growth. When accounts receivable are growing faster than sales consider it a warning sign. Growth rates in sales and in accounts receivable should be roughly equal. If sales grow by 10 percent, then accounts receivable should grow by 10 percent.
For example, the sales of contact-lens manufacturer Bausch and Lomb grew 12 percent in 1992 and 10 percent in 1993, while receivables grew 35 percent and 39 percent in those same years. The reason for the difference was that Bausch and Lomb changed the way it accounted for revenues. It previously recognized sales when made to end customers, but it changed to recognizing sales when made to distributors. A sale to a distributor is not the same as selling to an end customer, because selling to distributors doesn’t guarantee that the end customer will actually buy the product, and the distributor can return any unsold goods.
By the end of 1994, the SEC began a formal investigation into Bausch and Lomb’s accounting practices. The SEC investigators concluded that during 1993, executives artificially boosted the company’s earnings by wrongly recognizing revenue from the sale of contact lenses. This accounting scandal led to an earnings restatement and the departure of executives.2
There may be legitimate reasons for differences in accounts receivable and sales growth, although any discrepancies should be carefully investigated. For example, sometimes receivables growth exceeds sales growth, and this may happen for several reasons:
First, the disparity in growth rates might reflect a deliberate change in sales terms designed to attract new customers. For example, instead of requiring payment within 30 days of shipment, a business might allow customers to pay in 45 days.
Second, it can be due to deteriorating creditworthiness among existing customers, which would represent another problem.
Third, a business could have changed its financial-reporting procedures, which determine when sales are recognized.
Under- or Over-Stating Expenses
Management sometimes shifts current expenses to later periods in order to boost short-term earnings. Management can do this by capitalizing an expense item over several periods. This way, it can deduct the expense over many years, instead of all at once when it is incurred. Capitalized costs end up on the balance sheet as assets, which are then amortized over future periods.
Common types of expenses that are capitalized and later depreciated include:
Start-up costs
Research and development (R&D) expenses
Software development
Maintenance costs
Marketing
Customer-acquisition costs
You can find out if a business routinely capitalizes its costs by reading the footnotes to the financial statements.
For example, Internet service provider America Online (AOL) disclosed in its footnotes to the 1994 10-K that it was classifying marketing costs as balance sheet assets rather than operating expenses, naming them Deferred Subscriber Acquisition Costs. By capitalizing these expenses, AOL was able to overstate its earnings for several years.
Beginning in 2001 and continuing through May 2002, telecom company WorldCom capitalized more than $9 billion of ordinary expenses in order to mask its deteriorating financial condition and increase the price of WorldCom’s stock. On July 21, 2002, WorldCom filed for Chapter 11 bankruptcy protection and on March 15, 2005, CEO Bernard Ebbers was convicted of fraud and sentenced to 25 years in prison.3
Manipulating Discretionary Costs
Discretionary costs include advertising, R&D expenses, and maintenance costs. These costs can also be manipulated in order to smooth earnings. Management can curtail spending in any of these areas to meet an earnings goal. This reduction in expenses may compromise the long-term viability of the business. You should monitor these expenses on a quarterly basis and look for any irregular patterns. If you discover that in the fourth quarter, R&D expenses dropped by a material amount compared to the same quarter a year ago, this may indicate that management is attempting to smooth earnings.
Manipulating discretionary expenses is quite common. Professors Campbell Harvey and John Graham at Duke University and Shiva Rajgopal at the University of Washington surveyed 401 financial executives at U.S. companies in 2003 asking them what actions they might take to smooth earnings and meet analyst estimates. Close to 80 percent of the financial executives surveyed responded that in order to meet an earnings goal, they would reduce spending on discretionary items like maintenance, research, and advertising.4
Changing Accounting Methods
Look for changes in the accounting methods the company uses. A business must always disclose in a footnote if it changes accounting methods, and it must disclose any impact on earnings.
One of the more popular methods used to defer expenses is to extend the useful life of assets in order to reduce depreciation expenses. A business must disclose in the footnotes to the financial statements whether it has extended the useful life of an asset or changed depreciation methods. Also, when the book value of the asset is more than the present value of projected cash flows from the asset, some companies simply write down the value of the asset.
Management may be tempted to write down the value of assets substantially during years when earnings are lean. The effect of writing down the asset is that depreciation and amortization expenses are also reduced for all periods in the future, causing earnings to automatically increase during future periods.
For example, Waste Management artificially boosted earnings by extending the useful life of assets and using high residual values. The alleged fraud unraveled in 1997 when new interim CEO Robert Miller ordered a review of Waste Management’s past accounting practices. Waste Management issued a restatement of $1.7 billion, which was the largest in corporate history at that time. Stockholders lost more than $6 billion in market value when, after the restatement, the price of the stock fell by 33 percent.5 In its 1997 annual report, the company described how it changed to more conservative accounting:
Effective October 1, 1997, the Board of Directors approved a management recommendation to revise the company’s North American collection fleet management policy. Front-end loaders will be replaced after 8 years, and rear-end loaders and rolloff trucks after 10 years. The previous policy was not to replace front-end loaders before they were a minimum of 10 years old and other heavy collection vehicles before they were a minimum of 12 years old. Also effective October 1, 1997, the company reduced depreciable lives on containers from between 15 and 20 years to 12 years, and ceased assigning salvage value in computing depreciation on North American collection vehicles or containers.
Restructuring Charges or One-Time Expenses
If the business is reporting a large restructuring loss, management may add extra expenses in that restructuring charge in order to decrease future expenses. Later, management can reverse these restructuring charges to increase future earnings.
You can find these reversals in the footnotes to the financial statements by looking for liability reserves. Typically, a liability reserve is set up by writing off the expected cost of the restructuring as an expense and crediting a restructuring reserve or other accrued liabilities and payables account that is recorded as a liability on the balance sheet. Later, as the costs of the restructuring are paid in cash, the restructuring reserve liability is reduced. If management initially overestimates the amount of the restructuring expense, it can reverse the liability and add the amount to earnings.
For example, when Albert Dunlap became CEO of Sunbeam in July 1996, he took several restructuring charges that reduced net income from $50 million in 1995 to a loss of $228 million in 1996. By recording such unusually high restructuring costs, Sunbeam was able to move future-year expenses into its 1996 results. As a result, by 1997, net income jumped to $109 million, more than two times the amount of net income Sunbeam reported only two years earlier in 1995.
It was later discovered that Dunlap was manipulating earnings by taking large restructuring charges and then reversing these charges in later years. In 2001, the SEC sued Dunlap for many instances of fraud and barred him from serving as a public company executive.6
Using Reserves
Reserves are often called the cookie jar of accounting manipulation because they require a large degree of judgment to estimate. The term cookie jar refers to the ability to artificially store earnings in the balance sheet, so that dishonest management can draw on them in unprofitable future years and lessen the impact of the negative times on its financial statements. This is not a desirable occurrence, and the result can be financial statements that are intentionally misleading. Management can overestimate the size of a reserve account, which it can then later draw on to increase future earnings. Reserves can be booked for:
Bad debts
Sales returns
Inventory obsolescence
Warranties
Product liability
Litigation
Environmental contingencies
To create reserves, management can take a charge against current earnings for one of the above-listed reasons. Sometimes management can use unrealistic assumptions when they estimate reserve charges. They can over-reserve in good times, then cut back on or even reverse charges in bad times. This is what makes the reserve (the cookie jar) such a convenient income-smoothing mechanism for management. Look in the footnotes at the allowance for doubtful accounts and then compare the provision for doubtful accounts to actual charge-offs. If the provision account estimates increase or decrease by a material amount over a three-year period while charge-offs remain constant, this may be a sign that management is using reserves to manipulate earnings.
For example, food distributor Sysco uses a conservative method for its allowance for doubtful accounts. In 2009, the company charged $74 million to costs and expenses (provision for bad-debt expense). In 2009, it wrote off $72 million in customer accounts (charge-offs), net of recoveries.7 Therefore, it correctly matched its provision for doubtful accounts to actual charge offs. You can therefore assume that Sysco conservatively estimates the charge-offs. Table 6.1 illustrates.
Table 6.1 Sysco: Allowance for Doubtful Accounts
Source: 2010 Sysco 10-K.
On the other hand, Krispy Kreme Doughnuts manipulated earnings by failing to book adequate reserves for doubtful accounts. The provision for doubtful accounts and the actual charge-offs are shown in Table 6.2.
Table 6.2 Krispy Kreme: Allowance for Doubtful Accounts Related to Trade Receivables
Source: 2007 Krispy Kreme 10-K.
As you can see in Table 6.2, the provision for doubtful accounts was $12.7 million in 2005, which management overstated because actual charge-offs were $2.6 million in 2005. By over-estimating the provision in 2005 and under-estimating the provision in 2006 and 2007, the management of Krispy Kreme was able to increase earnings in 2006 and 2007.
28. Does the business generate revenues that are recurring or from one-off transactions?
Businesses that earn their revenues from recurring sources are easier to value when compared to those that generate revenue from one-off transactions. Businesses that have recurring revenue include:
Subscription-based businesses, such as a cable television business.
Razor/razorblade-type business models, such as Immucor, which sells blood-bank equipment and consumable reagents to be used with the equipment.
Franchisors such as Choice Hotels, which earns license fees on hotel brands such as Comfort Inn, Rodeway Inn, and Sleep Inn.
Service businesses, such as Fiserv, which provides integrated data processing and information-management systems to financial-services providers and earns more than 95 percent of its revenue from recurring service contracts.
It is easier to forecast the future revenues for these types of businesses because a recurring revenue business’s starting base isn’t zero, but a certain percentage of last year’s level of sales. Suppose a business earns $1.00 per share in revenue and $0.90 is from recurring revenue, then you know that $0.10 is at risk. Recurring revenue allows new sales to add to the revenue base, rather than simply replace lost revenues. For example, Praxair provides hydrogen and other gases to industrial clients under 15-year contracts, which produces an extremely predictable revenue stream. Every new customer Praxair is able to add then provides additional revenue.
Other advantages of businesses that earn recurring revenues include:
Less dependence on new products. A business does not have to continually come up with new products or services to replace the prior year’s revenues. On the opposite side of the spectrum from recurring-revenue businesses are those that generate revenues from one-off transactions or those that depend on continually selling the same number of products to maintain their sales from the prior year. These types of businesses range from consumer-products businesses, such as navigation-equipment maker Garmin, to equipment manufacturers that secure large individual contracts, such as Caterpillar. These types of businesses depend on orders that come in one at a time, with no guarantee or predictability in the revenue stream, and are therefore more difficult to value.
Greater predictabili
ty. If there is a lack of visibility into a business, it is difficult for management to know how much to invest or how much to budget for expenses, which increases the odds of making mistakes. For example, if management makes a mistake in its sales forecast for the year and budgets its expenses to that level of sales, the business will likely generate a loss. Therefore, recurring revenues help put a limit on the downside earnings of the business.
You can typically find information on recurring revenues in the 10-K under the Management, Discussion and Analysis (MD&A) section. For example, the 2010 10-K for hotel franchisor Choice Hotels states, “Our company generates revenues, income, and cash flows primarily from initial, relicensing, and continuing royalty fees attributable to our franchise agreements (20-year agreements on average).” This indicates that Choice Hotels generates the majority of its revenue from recurring sources.
29. To what degree is the business cyclical, countercyclical, or recession-resistant?
As the economy speeds up or slows down, industries and individual businesses may either go along with the economic cycle, go in the opposite direction, or not be affected at all. There is tremendous variation in how businesses react to economic cycles. In general, if a business’s earnings decline during an economic downturn, the business is classified as cyclical. If earnings increase during an economic contraction, we call it countercyclical. It is important to determine the degree to which a business is cyclical or countercyclical when forecasting earnings. Earnings that fluctuate up and down with the business cycle are more difficult to forecast than earnings that display greater stability over the economic cycle.
For example, it is difficult to know what the future economics will be for a homebuilder. Demand for homes is driven by many macro-economic factors such as interest rates, the cost of materials, the employment situation (people don’t buy houses if they are worried about losing their jobs), and the general health of the economy. It is difficult, if not impossible to forecast how these multiple macroeconomic factors will change and what the cumulative effect on homebuilders will be.
The Investment Checklist Page 18