Roundtable Meeting and Discussion of "Accounting for Goodwill"

Executive Summary:

• On May 31, 2000, "Accounting for Business Combinations: A Workable Solution," was presented to the FASB by a group of investment banks and accounting firms seeking to resolve the current debate over the treatment of goodwill. The team was lead by the author of this summary and included representatives of Morgan Stanley Dean Witter, Goldman Sachs, Deloitte & Touche, PricewaterhouseCoopers and ArthurAndersen. The presentation package is available on the FASB's website www.fasb.org or from the author.

• The approach presented to the FASB links the measurement of goodwill, the perceived irrelevance of a systematic amortization charge in earnings and an impairment test. We argue that goodwill should be capitalized but not amortized. A goodwill impairment test should be conducted using a residual income valuation model to prevent an overstated asset value.

• In an acquisition, goodwill is the difference between the purchase price and the value of recorded net identifiable assets. So goodwill represents the portion of the price paid for the entity's ability to generate a rate of return on invested capital (profitability or return on equity) in excess of a required rate of return (cost of equity) for some period. The amount and duration of "excess" profitability determine the size of goodwill.

• The "excess" profitability has historically been defined as residual income, and can be easily incorporated into a valuation model.

• A residual income valuation model measures an entity's value as the sum of its invested capital and discounted expected residual income. The model makes an explicit link between the accounting measures of net asset value (equity) and earnings, which more traditional corporate finance models do not.

• The valuation analysis can be performed at any time after the acquisition and so provides a relatively simple and reliable benchmark for evaluating whether goodwill has been impaired.

• In contemplating the relevance of goodwill amortization we note that the amount and duration of excess profitability implicit in the goodwill is impacted by both the ability of a firm to sustain its competitive advantage and the relative conservatism of accounting practices. We note that the goodwill will not decline in value over time unless the profitability is consistently less than originally forecasted. We argue that the excess profitability is sustained by:
  • Spending on brand and quality maintenance

  • Sustaining product differentiation through research and development, and

  • Superior management skills and other intellectual property.

• The current accounting model requires all of these costs to be expensed. Hence, if the entity were investing to sustain its profitability we would be "double-expensing" by charging amortization of goodwill. Non-amortization of goodwill increases comparability of reported income among companies, so long as profitability is sustained.

• If the company is not investing to sustain its profitability then this will become apparent in the residual income calculation and an impairment charge should be made. The impairment charge is much more relevant to investors than an amortization charge over some arbitrary period, as it provides information that is useful for forecasts of future income and returns on reinvested capital.

• We note that, while most sophisticated investors are adding back goodwill amortization in their valuation analyses, there is confusion in the investment community about the extent to which analysts are retaining or adding back amortization in their estimates. Less sophisticated investors will find the information even more confusing.

The fundamental concept underlying all valuation models is that an entity's value (to its shareholders) is the sum of all cash flows to and from the shareholders, discounted by some rate to adjust for the time value of money and risk.

The simplest version of this valuation framework is a dividend discount model [DDM] that utilizes the actual cash flows to shareholders. ₁ However, such models are impractical because dividend payment patterns are arbitrary, and in any period the best companies only return a small percentage of their earnings (or free cash) as a dividend to the owners. Consider the case of Berkshire Hathaway, run by Warren Buffett, arguably the most successful investor of our time. The company has not paid a dividend during Mr. Buffett's ownership, yet it is highly valued by its shareholders. A DDM for Berkshire Hathaway would give a strange and largely meaningless result.

The DDM is based on discounting the cash flows to shareholders. Traditional corporate finance models have adopted the framework to develop discounted cash flow [DCF] models for a company. However, in practice, these models suffer from three main concerns:

i. Measurement of relevant cash flows is difficult,

ii. Selecting a discount rate is quite arbitrary, and

iii. For most companies more than half (and often 80-90%) of the computed value is in a terminal value number which is subjective.

While DCF models are commonly used, they can be easily manipulated to provide a wide range of values for companies that are expected to operate indefinitely. As a result, it is understandable that auditors and regulators are uncomfortable relying on such valuation models to assess whether goodwill is impaired. This concern is exacerbated by the fact that a DCF valuation is difficult to link with accounting measures of shareholder's equity and goodwill.

A practical solution to the problem has evolved in the late 1990's with the revival of an old concept from management accounting known as residual income. The basic concept is that an investment that earns a return above the required (discount) rate generates positive residual income (adds to the economic value of the investment) and so is worth more than the original cash invested. The concept was popularized and re-introduced into management incentive programs by a variety of consulting firms, including Stern Stewart Inc., who branded their version as EVAŽ or Economic Value Added.

Simultaneously academic researchers showed that, with an accounting system that equates changes in shareholders equity to income and dividends, it is not difficult to link the DDM to the actual accounting numbers and derive a value for the firm.₂

Simply stated the value of a firm is the sum of its net asset value (invested capital) and the discounted value of expected "excess" profitability (residual income). The valuation approach can be directly linked to the reported measures of earnings and equity and so can be used to explicitly identify the premium or goodwill (excess future profitability) that is paid for in an acquisition. Moving forward, the achievement of these goals and realizations of income can be evaluated so that a natural benchmark exists for assessing impairment of goodwill with greater reliability than a traditional DCF approach.

The economic logic and simplicity of the residual income valuation approach has lead to its adoption for a wide range of uses and users. For example, sell-side analysts use it for their valuations (e.g. CS First Boston and Goldman Sachs report EVA measures and several Morgan Stanley Dean Witter analysts use residual income). Investment firms such as CREF, Oppenheimer and Putnam use variants of the model, and many corporations have adopted the model for internal performance measurements. Corporate users include BankAmerica, The Coca-Cola Company, Eli Lilly, Monsanto and Sprint. Residual income valuation is also commonly found in the most popular financial statement analysis textbooks.

One feature that makes a residual income valuation model appealing as a reliable benchmark for testing impairment of goodwill is that residual income is easily related into traditional profitability ratios, including profit margins, asset turnovers and leverage. Consequently, auditors can gain comfort as to the achievability of the target levels of performance required to justify the goodwill acquired.

In sum, a residual income valuation approach, while not flawless, has a strong conceptual foundation as well as proven practical application, and it provides an explicit link to the goodwill recorded on acquisition. This link allows for both an understanding of the expected profitability implied in the acquisition price and a reliable post-acquisition impairment test.

Goodwill is impaired if the company's rate of profitability is permanently below the growth rates that were incorporated into the original acquisition value.

• In our proposal, the impairment test is performed using the residual income valuation model that uses expected profitability to generate a net asset value that can be used to assess the value of recorded goodwill.

• The first step in the impairment review is to obtain forecasts of future profitability from company business plans, including expected: revenue, operating costs, net financial costs, returns of capital, including dividends and share repurchases, and financing needs. These measures are used to compute annual profitability measures, e.g. return on equity, return on net operating assets, operating margins, operating asset turnover and leverage.

• The results can be compared to historical performance and industry or competitive company benchmarks to evaluate the reasonableness of the forecasts.

• The next step is to compute an implied value using a residual income model and the forecasts from the business plans. To obtain the value it is necessary to evaluate the sustainability of the (positive) residual income over a competitive advantage period [CAP]. This period is determined by a number of factors - such as the competitive landscape, barriers to new entrants, reinvestment of free cash flows and investment opportunities for these investments, and conservatism of accounting. Then at any stage after the forecast period, adjust the rate of profitability towards the required rate of return. The speed of this adjustment will be dependent on the CAP.

• When residual income reaches a level growth rate, compute the terminal value.

• At this point, the implied revenue and earnings growth should be assessed for reliability, once again using industry norms, business plans, or past realization targets for comparability.

• Finally, by comparing the residual income value to net asset value, goodwill can be assessed for impairment.

• Additional sensitivity analysis can be can be performed to ensure the reliability of the estimates.

₁ - For simplicity, we can think of share repurchases as equivalent to dividends.

₂ - See Ohlson, J.A. "Earnings, Book Values and Dividends in Equity Valuation." Contemporary Accounting Research, (1995). Vol. 11. pp. 661-87.

Home | Menu | Links | Info | Chairman's Page