| EXECUTIVE SUMMARY 
Return-on-equity (ROE) is the correct profit metric to evaluate the performance of a business. However, the primary emphasis on financial ratio analysis must be on operating performance.
The “advanced” version of the DuPont model remedies the original model’s failure to cleanly separate the effects of operating and financing decisions. It introduces the concept of return on net operating assets (RNOA) as the core measure of operating performance and clearly separates the effects of leverage and operating decisions.
The advanced model does not change the result of the ROE calculation. However, the elements underlying the ROE ratio are different and provide a clean separation of operating and financing decisions.
RNOA is effectively insulated from financing decisions. Changing the amount of debt does not affect the operating assets or the profit before interest expense and, therefore, does not affect RNOA. It also permits straightforward computation of the impact on ROE of alternative financing decisions. Changes in the interest rate affect the spread, while changes in the amount of debt affect financial leverage in a transparent manner. David C. Burns, CPA, DBA, J. Timothy Sale, CPA, Ph.D., and Jens A. Stephan, Ph.D., are accounting professors at the University of Cincinnati College of Business. Their e-mail addresses, respectively, are david.burns@uc.edu , tim.sale@uc.edu and jens.stephan@uc.edu . R atios provide a concise and systematic way to organize the enormous quantity of data contained in financial statements into a framework that creates meaningful information. Financial managers use ratios to benchmark their firm’s performance against that of their competitors and set goals for future performance. Financial advisers use ratios to identify underpriced or overpriced stocks and make recommendations to investors.
Public accountants use ratios in nearly every service they offer to their clients. In the independent auditing arena, analytical procedures, which include ratio analysis, have effectively become a generally accepted auditing procedure since SAS 56 was issued in April 1988. Return-on-equity (ROE) is the ratio most commonly used to analyze the profitability of a business. The “original” DuPont ROE model, which was created in 1919 by a finance executive at E.I. du Pont de Nemours & Co., breaks ROE into three fundamental drivers of accounting returnon-equity: net profit margin, asset utilization and financial leverage. An “advanced” version of the DuPont model, which has found its way into accounting textbooks over the past several years, remedies the original model’s failure to cleanly separate the effects of operating and financing decisions. It introduces the concept of return on net operating assets (RNOA) as the core measure of operating performance and clearly separates the effects of leverage and operating decisions. This article demonstrates the benefits of the advanced DuPont model and uses the model to compare the financial performance of two popular U.S. retailers: Target and Costco. THE ORIGINAL DUPONT MODEL OF FINANCIAL ANALYSIS
The original DuPont model can be illustrated with a simple example using two savings banks. Assume that Bank A and Bank B are, respectively, willing to pay 10% and 6% for deposits and are also willing to lend at the same rates. An all-equity firm with $1,000 in assets to invest would clearly choose Bank A to earn an ROE = Return on Assets (ROA) = 10%. Assume the firm pays out all profits in the form of a cash dividend. Using the original DuPont model to examine the effects of financial leverage on ROE and ROA, assume a firm borrows $800 from Bank B at 6%, leaving an equity investment of $200. The reader can verify that: Interest revenue = $100
Net income (NI) = $52
Assets = $1,000
Interest expense = $48
Equity = $200
ROE = $52 / $200 = 26.0% The model disaggregates ROE into two components: ROA (NI / Assets) and financial leverage (Assets / Equity). 
This demonstrates the fundamental shortcoming of the original DuPont model. The decline in ROA from 10% to 5.2% could easily be interpreted as lower operating performance. However, in both examples, the firm’s managers have invested all of the firm’s assets in Bank A, still earning a 10% return. Therefore, many seasoned analysts feel that the decline in ROA is a false signal of a decline in the firm’s operating performance. The model’s failure to insulate the ROA ratio from the effects of financing decisions is a major shortcoming in the model and prompted the development of a variation of the DuPont model. Here’s another way of illustrating the significance of the problem: Consider an operating manager who is entrusted with $1,000 of assets and has been earning a 10% return on assets. The CFO then decides to finance some of the assets with debt instead of equity. The ROA used to assess the operating manager’s performance drops from 10% to 5.2%. In this case, the operating manager does not have control over capital structure decisions, yet his or her performance metric (ROA) is being affected by the financing decision. Alternatively, a financial analyst or auditor using the model might fail to take into account the separate effects of operating and financing decisions on a firm’s ROA and might reach erroneous conclusions about the relative operating performance of levered and un-levered firms. Finally, an auditor applying the model might reach unwarranted conclusions about a client’s operating performance if the capital structure changed over time. These simple examples illustrate the need for an ROE model that distinguishes between the effects of operating and financing decisions. The advanced DuPont model has been developed to meet that need.
THE ADVANCED DUPONT MODEL OF FINANCIAL ANALYSIS
The advanced DuPont model does not change the result of the ROE calculation. However, the elements underlying the ROE ratio are different. The following examples demonstrate how the advanced DuPont model allows for a clean separation of operating performance—return on net operating assets (RNOA)—and financing decisions—financial leverage (FLEV) and spread—and helps managers, analysts and CPAs avoid erroneous conclusions from their financial statement analyses. Using the earlier set of facts to illustrate the use of the advanced DuPont model, net income is computed as follows: 
Note that the $200 in equity earns 10% while the $800 in debt earns 4% net of the interest cost. ROE is defined as net income divided by equity, so dividing by $200 yields the following expression for ROE: 
The notation of the advanced DuPont model yields the following expression: 
RNOA = $100 / $1,000 = 10%
Spread = RNOA – Interest Rate = 10% – 6% = 4%
FLEV = Interest-Bearing Debt / Equity = $800 / $200 = 4.0 RNOA is the appropriate metric to assess management’s operating decision to invest the assets in Bank A because it is effectively insulated from the financing decision. Changing the amount of debt does not affect the operating assets or the profit before interest expense and, therefore, does not affect RNOA. This remedies one of the primary problems with the original DuPont model. It also permits straightforward computation of the impact on ROE of alternative financing decisions. For example, changes in the interest rate affect the spread while changes in the amount of debt affect financial leverage in a transparent manner. A PPLICATION
We now apply the advanced DuPont model to analyze the discount retailers Target Corp. and Costco Wholesale Corp. Yahoo! Finance provides financial statements in a standardized format facilitating our analysis. See Exhibit 1 for the balance sheets and income statements for both firms in GAAP format. (Target officials reviewed this article. A Costco investor relations official could not be reached.) 
Note that in our example, we include Target’s credit card revenues/expenses and Costco’s membership fees in the analysis. (Yahoo! Finance provides only the total revenue number.) Some financial analysts would exclude these items to get a better picture of operating performance. Both firms have nonsales related revenue of approximately equal magnitude, so including the figures should not affect the comparison in a significant way. First, we calculate ROE from the financial statements in GAAP format using average stockholder equity in our calculation, because the net income was earned evenly throughout the year and the investment by owners, therefore, changed evenly throughout the year. The same rationale is applied for all balance sheet items. The averages of beginning and ending balances are used in the subsequent analyses. It is worthwhile to note that both companies have highly seasonal revenues and inventories. Using average inventory levels based on beginning and ending fiscal year values (when such levels are unusually low), therefore, understates the true average investment in inventories (operating assets) throughout the year. 
Next, we reformulate the financial statements to distinguish between operating and financing activities (see Exhibit 2 ). Note that the balance sheet calculates operating assets, operating liabilities (the difference between these two is net operating assets), financial assets (Target does not report any financial assets in the form of investments in marketable securities), and financial liabilities (interest-bearing debt). 
The difference between financial assets and financial liabilities is net financial obligations. Also, a firm with financial assets greater than financial liabilities would effectively have negative financial liabilities (financial assets) and negative interest expense (interest income). Similarly, the income statement calculates net operating profit after tax (NOPAT) and then subtracts after-tax interest to arrive at net income, which is the same as in the GAAP format. We are now ready to apply the advanced DuPont model formulation to Target and Costco. From Exhibit 2 , we reproduce the income statement and balance sheet items shown in Exhibit 3 (all Target and Costco financial statement numbers are in millions and balance sheet numbers are averages of values for the beginning and end of the year): 
These variables are combined to verify that the advanced DuPont model calculates the same ROE as in the original DuPont model using the GAAP formulation of the financial statements. 
RNOA = NOPAT / NOA
NBC = net borrowing cost = NFE / NFO
SPREAD = RNOA – NBC
FLEV = NFO / SE Both firms have virtually identical operating performance (RNOA). Target’s higher ROE is almost entirely a result of financing effects. The firm uses more financial leverage and has a higher spread, which makes ROE almost 50% greater than without financial leverage (Target’s credit card business helps explain a portion of the financial leverage difference). We now understand how operating performance and nonoperating performance combine to create return on equity. To gain a better understanding of operating performance, we can disaggregate it into profit margin and asset turnover components. This is accomplished in the following manner (note the algebraic identity because revenue (REV) cancels out): 
NOPM = net operating profit margin
NOAT = net operating asset turnover Target achieves its operating performance with higher profit margins but lower operating asset turnover. How do we know this is strategy and not happenstance? Firms that choose a cost leadership strategy will typically sell at lower prices but with higher volume, while firms that choose a differentiation strategy will sell smaller quantities but at higher prices. Looking at NOPM and NOAT for six firms in the discount variety store industry— Target, Costco, Dollar General Corp., Wal-Mart Stores Inc., 99¢ Only Stores, and BJ’s Wholesale Club Inc.—we see evidence of just such a tradeoff between NOPM and NOAT (see Exhibit 4 ). 
However, providing actionable advice to clients or achieving specific performance goals for your firm requires even more detail. This leads to a closer examination of expense and turnover ratios for specific assets. Using the financial statements in GAAP format, we can calculate the ratios shown in Exhibit 5 . 
The large differences between cost of goods sold (CGS) to sales and selling, general and administrative expense to sales suggest that Target and Costco classify their costs into each major expense category differently. A review of the footnotes to the financial statements did not allow us to make the classifications comparable. Unfortunately, this problem also precludes unambiguous inferences about number of days inventory and number of days payables because both metrics include CGS. This is one of those examples where more detailed information from management is necessary to recast the income statement and balance sheet numbers on a comparable basis. Your clients should be able to provide this information. Property, plant & equipment (PP&E) turnover is much larger for Costco. However, an accounting issue that might affect this is the degree to which each company uses operating leases to finance stores and warehouses. The footnotes reveal that total minimum operating lease payments for Target and Costco are $3,325 and $1,894, respectively (in millions). Capitalizing these lease payments would make the PP&E turnover ratios converge somewhat. We conclude that Costco generates significantly more revenue per dollar of fixed assets than Target and that asset utilization appears to be an important value driver in its business model. A related question is why Target’s spread is so much larger than Costco’s. It is unlikely that the cost of borrowing is dramatically different for the firms given their similar operating performance (and presumably similar risk of default). There is, unfortunately, insufficient disclosure in the financial statements and the footnotes to resolve this issue—reinforcing the adage that ratio analysis does not give you answers—it tells you what questions to ask. DIFFERENT CONCLUSIONS ?
Would our conclusions have been different if the original DuPont model of financial analysis had been applied to these companies? The answer is likely yes. Computing several key ratios using the original DuPont model yields the following: 
ROA = net income ÷ average total assets
NPM = net profit margin = net income ÷ revenue
TAT = total asset turnover = revenue ÷ average total assets
LEV = financial leverage =average total assets ÷ average stockholders equity Using the original DuPont model, we would infer that Target’s operating performance (ROA) is 33% better than Costco’s—a much different conclusion than was reached using RNOA. And we would gather that Target uses somewhat more leverage than Costco (2.42 vs. 2.09), but the financial leverage for the firms (0.667 vs. 0.050) paints a much more radical picture of Target’s use of debt (leverage in the original DuPont model mixes operating and financial liabilities We would not know the spread between operating performance (RNOA) and the cost of debt (NBC). For example, we don’t know whether more interest-bearing debt would increase or decrease ROE. ROE is accepted as the best profit metric to evaluate the performance of a business. However, the primary emphasis on financial ratio analysis must be on operating performance. That is, after all, where value creation takes place. The advanced DuPont model permits a clean separation of operating and nonoperating performance and represents a state-of-the-art tool available for systematic analysis of company performance. Accountants must understand their employer’s or clients’ performance, both with respect to its own history and with respect to competitors in order to provide competent service. An understanding of the advanced DuPont model is a critical skill set in this endeavor  AICPA RESOURCES CPE
Financial Statement Analysis: Basis for Management Advice , a CPE self-study course (#731249) Publications
Guide to Financial Statement Analysis: Basis for Management Advice (#091022)
Analyzing Financial Ratios Practice Aid—Forensic and Valuation Services Practice Aid 06-3 (#055302)
Building Financial Models with Microsoft Excel: A Guide for Business Professionals (#WI661031) For more information or to place an order, visit www.cpa2biz.com or call the Institute at 888-777-7077. | 
