Business Performance Evaluation: Approaches for Thoughtful Forecasting Every day, fortunes are won and lost on the backs of business performance assessments and forecasts. Because of the uncertainty surrounding business performance, the man- ager should appreciate that forecasting is not the same as fortune-telling; unanticipated events have a way of making certain that specific forecasts are never exactly correct. This note purports, however, that thoughtful forecasts greatly aid managers in under- standing the implications of various outcomes (including the most probable outcome) and identify the key bets associated with a forecast. Such forecasts provide the manager with an appreciation of the odds of business success. This note examines principles in the art and science of thoughtful financial fore- casting for the business manager. In particular, it reviews the importance of (1) under- standing the financial relationships of a business enterprise, (2) grounding business forecasts in the reality of the industry and macroenvironment, (3) modeling a forecast that embeds the implications of business strategy, and (4) recognizing the potential for cognitive bias in the forecasting process. The note closes with a detailed example of financial forecasting based on the example of the Swiss food and nutrition com- pany Nestle, Understanding the Financial Relationships of the Business Enterprise Financial statements provide information on the financial activities of an enterprise. Much like the performance statistics from an athletic contest, financial statements provide an array of identifying data on various historical strengths and weaknesses This technical note was prepared by Professor Michael J. Schill. Special thangs goto Vladimir Kolcin for datacollection assistance and to Lee Ann Long-Tyler and Ray Nedzel for technical assistance. Copyright © 2015 by the University of Virginia Darden School Foundation, Charlottesville, VA. All rights reserved To order copies, send an e-mail 1 [email protected]. No part ofthis publication may be reproduced, stored ina retrieval system, used ina spreadsheet, or transmiited in any form or by any ‘means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of the Darden School Foundation.PartTwo Financial Analysis and Forecasting across a broad spectrum of business activities. The income statement (also known as the profit-and-loss statement) measures flows of costs, revenue, and profits over a defined period of time, such as a year. The balance sheet provides a snapshot of business invest- ment and financing at a particular point in time, such as the end of a year. Both state- ments combine to provide a rich picture of a business's financial performance. The analysis of financial statements is one important way of understanding the mechanics of the systems that make up business operations. Interpreting Financial Ratios Financial ratios provide a useful way to identify and compare relationships across finan- cial statement line items.' Trends in the relationships captured by financial ratios are particularly helpful in modeling a financial forecast. The comparison of ratios across time or with similar firms provides diagnostic tools for assessing the health of the vari- ous systems in the enterprise. These tools and the assessments obtained with them provide the foundation for financial forecasting We review common financial ratios for examining business operating performance. Itis worth noting that there is wide variation in the definition of financial ratios. A mea~ sure such as return on assets is computed many different ways in the business world. Although the precise definitions may vary, there is greater consensus on the interpreta- tion and implication of each ratio. This note presents one such definition and reviews the interpretation. Growth rates: Growth rates capture the year-on-year percentage change in a particu- lar Tine item. For example, if total revenue for a business increases from $1.8 million to $2.0 million, the total revenue growth for the business is said to be 11.1% [(2.0 ~ 1.8)/1.8]. Total revenue growth can be further decomposed into two other growth measures: unit growth (the growth in revenue due to an increase in units sold) and price growth (the growth in revenue due to an increase in the price of each unit). In the above example, if unit growth for the business is 5.0%, the remaining 6.1% of total growth can be attributed to increases in prices or price growth. ‘Margins: Margin ratios capture the percentage of revenue that flows into profit or, alternatively, the percentage of revenue not consumed by business costs. Business profits can be defined in many ways. Gross profit reports the gains to revenue after subtracting the direct expenses. Operating profit reports the gains to revenue after subtracting all associated operating expenses. Operating profit is also commonly re- ferred to as earnings before interest and taxes (EBIT). Net profit reports the gains to revenue after subtracting all associated expenses, including financing expenses and taxes. Each of these measures of profits have an associated margin. For example, if operating profit is $0.2 million and total revenue is $2.0 million, the operating ‘The analogy of athletic performance statistics is again useful in understanding how ratios provide additional ‘meaningful information. In measuring the effectiveness ofa batter in baseball, the batting average (number of hits + number of at bats) may be more useful than simply knowing the numberof hits. In measuring the success ofa running back in football the rato of rushing yards gained per carry may be more useful than simply knowing the total rushing yards gained,(Case 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting om margin is 10% (0.2/2.0). Thus, for each revenue dollar, an operating profit of $0.10 is generated and $0.90 is consumed by operating expenses. The margin provides the analyst with a sense of the cost structure of the business. Common definitions of mar- gin include the following: Gross margin = Gross profit/Total revenue where gross profit equals total revenue less the cost of goods sold. Operating margin = Operating profit/Total revenue where operating profit equals total revenue less all operating expenses (EBIT). NOPAT margin = Net operating profit after tax (NOPAT)/Total revenue where NOPAT equals EBIT multiplied by (1 — ), where r is the prevailing marginal income tax rate. NOPAT measures the operating profits on an after-tax basis without accounting for tax effects associated with business financing. Net profit margin = Net income/'Total revenue where net income or net profit equals total revenue less all expenses for the period. A business that has a high gross margin and low operating margin has a cost structure that maintains high indirect operating expenses such as the costs associated advertising or with property, plant, or equipment (PPE). Turnover: Turnover ratios measure the productivity, or efficiency, of business assets. The turnover ratio is constructed by dividing a measure of volume from the in- come statement (i.e., total revenue) by a related measure of investment from the balance sheet (ice., total assets). Turnover provides a measure of how much business flow is generated per unit of investment. Productive or efficient assets produce high levels of asset turnover. For example, if total revenue is $2.0 million and total assets are $2.5 million, the asset-turnover measure is 0.8 times (2.0/2.5). Thus, each dollar of total asset investment is producing $0,80 in revenue or, alternatively, total assets are turning over 0.8 times a year through the operations of the business Common mea- sures of turnover include the following: Accounts receivable turnover = Total revenue/Accounts receivable Accounts receivable turnover measures how quickly sales on credit are collected. Busi- nesses that take a long time to collect their bills have low receivable turnover because of their large receivable levels. Inventory turnover = Cost of goods sold/Inventory Inventory turnover measures how inventory is working in the business, and whether the business is generating its revenue on large levels or small levels of inventory. For inven- tory tumover (as well as payable turnover) it is customary to use cost of sales as the ‘volume measure because inventory and purchases are on the books at cost rather than at the expected selling price. PPE turnover = Total revenue/Net PPE~ 2 Part Two Financial Analysis and Forecasting PPE turnover measures the operating efficiency of the fixed assets of the business. Businesses with high PPE turnover are able to generate large amounts of revenue on relatively small amounts of PPE, suggesting high productivity or asset efficiency. Asset turnover = Total revenue/Total assets Total capital turnover = Total revenue/Total capital Total capital is the amount of capital that investors have put into the business and is defined as total debt plus total equity. Since investors require a return on the total capital they have invested, total capital turnover provides a good measure of the productivity of that investment. Accounts payable turnover = Cost of goods sold/Accounts payable Accounts payable turnover measures how quickly purchases on credit are paid. Businesses that are able to take a long time to pay their bills have low payable turnover because of their large payables levels. An alternative and equally informative measure of asset productivity is a “days” ‘measure, which is computed as the investment amount divided by the volume amount multiplied by 365 days. This measure captures the average number of days in a year that an investment item is held by the business. For example, if total revenue is $2.0 million and accounts receivable is $0.22 million, the accounts receivable days measure is calcu- lated as 40,2 days (0.22/2.0 x 365). The days measure can be interpreted as that the average receivable is held by the business for 40.2 days before being collected. The lower the days measure, the more efficient isthe investment item. If the accounts receiv- able balance equals the total revenue for the year, the accounts receivable days measure is equal to 365 days as the business has 365 days of receivables on their books. This means it takes the business 365 days, on average, to collect their accounts receivable. While the days measure does not actually provide any information that is not already contained in the respective turnover ratio (as it is simply the inverse of the turnover measure multiplied by 365 days), many managers find the days measure to be more in- tuitive than the turnover measure, Common days measures include the following: Accounts receivable days = Accounts receivable/Total revenue x 365 days nventory/Cost of goods sold x 365 days Inventory days Accounts payable days = Accounts payable/Cost of goods sold x 365 days Return on investment: Return on investment captures the profit generated per dollar of investment. For example, if operating profit is $0.2 million and total assets are $2.5 million, pretax return on assets is calculated as operating profit divided by total assets (0.2/2.5), or 8%. Thus, the total dollars invested in business assets are generating pretax operating-profit returns of 8%, Common measures of return on investment in- clude the following: Return on equity (ROE) = Net income/Shareholders’ equity where shareholders’ equity is the amount of money that shareholders have put into the business. Since net income is the money that is available to be distributed back to equityCase 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting 93 investors, ROE provides a measure of the return the business is generating for the equity investors. Return on assets (ROA) = NOPAT/Total assets where NOPAT equals EBIT x (1 ~ 1), EBLT is the earnings before interest and taxes, and 1 is the prevailing marginal income tax rate. Like many of these ratios, there are many other common definitions. One common alternative definition of ROA is the following: Return on assets (ROA) = Net income/Total assets and, lastly, Return on capital (ROC) = NOPAT/Total capital Since NOPAT is the money that can be distributed back to both debt and equity inves- tors and total capital measures the amount of capital invested by both debt and equity investors, ROC provides a measure of the return the business is generating for all investors (both debt and equity). It is important to observe that return on investment can be decomposed into a margin effect and a turnover effect, That relationship means that the same level of business profitability can be attained by a business with high ‘margins and low turnover, such as Nordstrom, as by a business with low margins and high turnover, such as Wal-Mart. This decomposition can be shown algebraically for the ROC: ROC = NOPAT margin x Total capital turnover NOPAT NOPAT _ Total revenue Total capital ~ Total revenue ™ Total capital Notice that the equality holds because the quantity for total revenue cancels out across the two right-hand ratios. ROE can be decomposed into three components: ROC = Net profit margin x Total capital turnover x Total capital leverage Netincome ___Netincome | Total revenue Total capital Shareholders’ equity Total revenue Total capital “ Shareholders’ equity This decomposition shows that changes in ROE can be achieved in three ways: changes in net profit margin, changes in total capital productivity, and changes in total capital leverage. This last measure is not an operating mechanism but rather a financing mech- anism, Businesses financed with less equity and more debt generate higher ROE but also have higher financial risk. Using Financial Ratios in Financial Models Financial ratios provide the foundation for forecasting financial statements because fi- nancial ratios capture relationships across financial statement line items that tend to be preserved over time. For example, one could forecast the dollar amount of gross profit for next year through an explicit independent forecast. However, a better approach is to forecast two ratios: a revenue growth rate and a gross margin. Using these two ratios in4 Part Two Financial Analysis and Forecasting combination one can apply the growth rate to the current year’s revenue, and then use the gross margin rate to yield an implicit dollar forecast for gross profit. As an example, if we estimate revenue growth at 5% and operating margin at 24%, we can apply those ratios to last year's total revenue of $2.0 million to derive an implicit gross profit fore cast of $0.5 million [2.0 x (1 + 0.05) x 0.24]. Given some familiarity with the financial ratios of a business, the ratios are generally easier to forecast with accuracy than are the expected dollar values. The approach to forecasting is thus to model future financial statements based on assumptions about future financial ratios. Financial models based on financial ratios can be helpful in identifying the impact of particular assumptions on the forecast. For example, models can easily allow one to see the financial impact on dollar profits of a difference of one percentage point in op- erating margin. To facilitate such a scenario analysis, financial models are commonly built in electronic spreadsheet packages such as Microsoft Excel. Good financial fore- cast models make the forecast assumptions highly transparent. To achieve transparency, assumption cells for the forecast should be prominently displayed in the spreadsheet (eg,, total revenue growth rate assumption cell, operating margin assumption cell), and then those cells should be referenced in the generation of the forecast, In this way, it becomes easy not only to vary the assumptions for different forecast scenarios, but also to scrutinize the forecast assumptions. Grounding Business Forecasts in the Reality of the Industry and Macroenvironment Good financial forecasts recognize the impact of the business environment on the per- formance of the business. Financial forecasting should be grounded in an appreciation for industry- and economy-wide pressures, Because business performance tends to be correlated across the economy, information regarding macroeconomic business trends should be incorporated into a business’s financial forecast. If, for example, price in- creases for a business are highly correlated with economy-wide inflation trends, the financial forecast should incorporate price growth assumptions that capture the avail- able information on expected inflation. If the economy is in a recession, then the fore- cast should be consistent with that economic reality. Thoughtful forecasts should also recognize the industry reality. Business prospects are dependent on the structure of the industry in which the business operates. Some in- dustries tend to be more profitable than others. Microeconomic theory provides some explanations for the variation in industry profitability. Profitability within an industry is likely to be greater if (1) battiers to entry discourage industry entrants, (2) ease of indus try exit facilitates redeployment of assets for unprofitable players, (3) industry partici- pants exert bargaining power over buyers and suppliers, or (4) industry consolidation reduces price competition.” Table 5.1 shows the five most and the five least profitable industries in the United States based on median pretax ROAs for all public firms from Michael E. Porter, “How Competitive Forces Shape Strategy,” Harvard Business Review 7, n0.2 (March-April 1979): 137-4.Case.S_ Business Performance Evaluation: Approaches for Thoughiful Forecasting 95 TABLE 5.1 | Most profitable and least profitable US. industries: 2005-2014. ee Most Profitable Industries Median Firm ROA —_Least Profitable industries Median Firm ROA. Tobacco Products 18% Chemicals and Allied Products ~25% Building Materials, Retail 16% Metal Mining 14% Leather and Leather Products 13% Mining and Quarrying ~4% Apparel and Accessory Stores 11% Building Construction -2% Apparel 10% Oil and Gas Extraction ~1% a 2005 to 2014. Based on the evidence, firms operating in the apparel and accessory retail industry should have systematically generated more profitable financial forecasts over that period than did firms in the metal-mining industry. One explanation for the differ- ences in industry profitability is the ease of industry exit. Inthe retail industry, unprofit- able businesses are able to sell their assets easily for redeployment elsewhere. In the mining industries, where asset redeployment is much more costly, industry capacity may have dragged down industry profitability. Being within a profitable industry, however, does not ensure superior business performance, Business performance also depends on the competitive position of the firm within the industry. Table 5.2 shows the variation of profitability for firms within the U.S. apparel and accessory stores industry from 2005 to 2014. Despite being one of the most profitable industries as shown in Table 5.1, there is large variation in profit- ability within the industry. All five firms at the bottom of the profitability list generated median ROAs that were actually negative (Delia’s, Frederick's, Bakers Footwear, Pacific Sunwear, and Coldwater Creek). Good forecasting considers the ability of a business to sustain performance given the structure of its industry and its competitive position within that industry. Abnormal profitability is difficult to sustain over time. Competitive pressure tends to bring abnormal performance toward the mean. To show that effect, we can sort all USS. public companies for each year from 2005 to 2015 into five groups (group 1 with low profits through group 5 with high profits) based on their annual ROAs and sales growth, We then follow what happened to the composition of those groups over the next TABLE 5.2 | Most and least profitable firms within the apparel and accessory stores retail industry: 2005-2014. Rankings in Tables 5.1 and 5.2 are based on all firms from Compustat organized into industries by 2-digit SIC codes. ee Most Profitable Firms Median Firm ROA. Least Profitable Firms Median Firm ROA Francesca’s 53% Delia's ~ ~29% Buckle 38% Frederick's 17% TWX 27% Bakers Footwear 10% Ross Stores 26% Pacific Sunwear 6% J.Crew 26% Coldwater Creek -5%96 Par Two FIGURE 5.1 | Financial Analysis and Forecasting Firm-ranking annual transitions by profitability and sales growth. Firms are sorted for each year into five groups by either annual pretax ROA or sales growth. For example, in the ROA panel, group 1 comprises the firms with the lowest 20% of ROA for the year; group 5 comprises the firms with the highest 20% of ROA for the year. The figure plots the mean ranking number for all U.S. public firms in the Compustat database from 2005 to 2015. ROA Sales Growth High ROA Group High Growth Group * Ranking Quintite Low ROA Group Low Growth Group f L L L 1 L YearO Yeart. Year? ~~ Year3 YearO Year! Year? Year3 Year After Ranking Year After Ranking three years. The results of this exercise are captured in Figure 5.1. The ROA graph shows the mean group rankings for firms in subsequent years. For example, firms that ranked in group 5 with the top ROA at year 0 tend to have a mean group ranking of 4.5 in year 1,4.3 in year 2, and 3.7 in year 3. Firms that ranked in group I with the low- est ROA at year 0 tend to have a mean group ranking of 1.5 in year 1, 1.7 in year 2, and 2.2 in year 3. There is a systematic drift toward average performance (3.0) over time. The effect is even stronger vis-a-vis sales growth, Figure 5.1 provides the transition Iatrix for average groups sorted by sales growth. Here we see that, by year 2, the aver- age sales growth ranking for the high-growth group is virtually indistinguishable from that of the low-growth group. Figure 5.1 illustrates that business is fiercely competitive. Itis naive to assume that superior business profitability or growth can continue unabated for an extended period. Abnormally high profits attract competitive responses that eventually return profits to their normal levels. Modeling a Base-Case Forecast that Incorporates Expectations for Business Strategy With a solid understanding of the business’s historical financial mechanics and of the environment in which the business operates, the forecaster can incorporate the firm’s operating strategy into the forecast in a meaningful way. All initiatives to improve revenue growth, profit margin, and asset efficiency should be explicitly reflected in the‘Case 5 Business Performance Evaluation: Approaches for Thoughiful Forecasting ” financial forecast, The forecast should recognize, however, that business strategy does not play out in isolation. Competitors do not stand still. A good forecast recognizes that business strategy also begets competitive response. All modeling of the effects of busi- ness strategy should be tempered with an appreciation for the effects of aggressive competition, One helpful way of tempering the modeling of business strategy’s effects is to complement the traditional bottom-up approach to financial forecasting with a top- down approach. The top-down approach starts with a forecast of industry sales and then works back to the particular business of interest. The forecaster models firm sales by modeling market share within the industry. Such a forecast makes more explicit the challenge that sales growth must come from either overall industry growth or market share gain. A forecast that explicitly demands a market share gain of, say, 20% to 24%, is easier to scrutinize from a competitive perspective than a forecast that simply projects sales growth without any context (¢.g., at an 8% rate). Another helpful forecasting technique is to articulate business perspectives into a coherent qualitative view on business performance. This performance view encourages the forecaster to ground the forecast in a qualitative vision of how the future will play out, In blending qualitative and quantitative analyses into a coherent story, the fore- caster develops a richer understanding of the relationships between the financial fore- cast and the qualitative trends and developments in the enterprise and its industry. Forecasters can better understand their models by identifying the forecast’s value drivers, which are those assumptions that strongly affect the overall outcome. For example, in some businesses the operating margin assumption may have a dramatic impact on overall business profitability, whereas the assumption for inventory turnover may make little difference, For other businesses, the inventory turnover may have a tremendous impact and thus becomes a value driver, In varying the assumptions, the forecaster can better appreciate which assumptions matter and thus channel resources to improve the forecast’s precision by shoring up a particular assumption or altering the business strategy to improve the performance of a particular line item. Lastly, good forecasters understand that it is more useful to think of forecasts as ranges of possible outcomes rather than as precise predictions. A common term in fore- casting is the “base-case forecast.” A base-case forecast represents the best guess out- come or the expected value of the forecast’s line items, In generating forecasts, itis also important to have an unbiased appreciation for the range of possible outcomes, which is commonly done by estimating a high-side and a low-side scenario. In this way, the forecaster can bound the forecast with a relevant range of outcomes and can best appreciate the key bets of the financial forecast. Recognizing the Potential for Cognitive Bias in the Forecasting Process - A substantial amount of research suggests that human decision making can be system- atically biased. Bias in financial forecasts creates systematic problems in managing and investing in the business. Two elements of cognitive bias that play a role in financial forecasting are optimism bias and overconfidence bias. This note defines optimism bias98 Part Two Financial Analysis and Forecasting FIGURE 5.2 | Optimism and overconfidence biases in forecasting the sales growth rate, Forecast Distribution Probability True Distribution, co) Sales Growth Rate asa systematic positive error in the expected value of an unknown quantity, and defines the overconfidence bias as a systematic negative error in the expected variance of an unknown quantity. The definitions of those two terms are shown graphically in Figure 5.2. The dark curve shows the true distribution of the sales growth rate. The realization of the growth rate is uncertain, with a higher probability of its being in the central part of the distribution. The expected value for the sales growth rate is g*; thus, the proper base-case forecast for the sales growth rate is precisely g*. The light curve shows the distribution expected by the average forecaster. This distribution is biased for two reasons. First, the expected value is too high. The forecaster expects the base-case sales growth rate to be g’, rather than g*. Such positive bias for expected value is termed optimistic. Second, the dispersion of the distribution is too tight. This dispersion is captured by the variance (or standard deviation) statistic. Because the forecast dispersion is tighter than the true dispersion, the forecaster exhibits negative variance bias, or overconfidence— the forecaster believes that the forecast is more precise than it really is. ‘A description and the implications of an experinent on forecasting bias among MBA. students is provided in an Appendix to this note. Nestle: An Example 1n 2013, Nestle was one of the world’s largest food and health companies, Headquartered in Switzerland, the company was truly a multinational organization with factories in 86 countries around the world. Suppose that in early 2014, we needed to forecast the financial performance of Nestle for the end of 2014, We suspected that one sensible place to start was to look at the company’s performance over the past few years, Exhibit 5.1 provides Nestle’s income statement and balance sheet for 2012 and 2103.Case 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting «99, One approach to forecasting the financial statements for 2014 is to forecast each line item from the income statement and balance sheet independently. Such an ap- proach, however, ignores the important relationships among the different line items (e.g. costs and revenues tend to grow together). To gain an appreciation for those relationships, we calculate a variety of ratios (Exhibit 5.1). In calculating the ratios, we notice some interesting patterns. First, sales growth declined sharply in 2013, from 7.4% to 2.1%. The sales decline was also accompanied by much smaller decline in profitability margins; operating margin declined from 14.9% to 14.1%. Meanwhile, the asset ratios showed modest improvement; total asset turnover improved only slightly, from 0.7x to 0.8x. Asset efficiency improved across the various classes of assets (¢.g., accounts receivable days improved in 2013, from 53.0 days to 48.2 days; PPE turnover also improved, from 2.8x to 3.0x). Overall in 2013 Nestle’s declines in sales growth and margins were counteracted with improvements in asset efficiency such that return on assets improved from 6.9% to 7.1%. Because relurn on assets com- prises both a margin effect and an asset-productivity effect, we can attribute the 2013 improvement in return on assets to a denominator effect—Nestle’s asset efficiency improvement. The historical ratio analysis gives us some sense of the trends in busi- ness performance. A common way to begin a financial forecast is to extrapolate current ratios into the future. For example, a simple starting point would be to assume that the 2013 financial ratios hold in 2014. If we make that simplifying assumption, we generate the financial forecast presented in Exhibit 5.2. We recognize this forecast as naive, but it provides a straw-man forecast through which the relationships captured in the financial ratios can be scrutinized. In generating the forecast, all the line-item figures are built on the ratios used in the forecast. The financial line-item forecasts are com- puted as referenced to the right of each figure based on the ratios below. Such a forecast is known as a financial model. The design of the model is thoughtful. By linking the dollar figures with the financial ratios, the model preserves the existing relationships across line items and can be easily adjusted to accommodate different ratio assumptions. Based on the naive model, we can now augment the model with qualitative and quantitative research on the company, its industry, and the overall economy. In early 2014, Nestle was engaged in important efforts to expand the company product line in foods with all-natural ingredients as well the company presence in the Pacific Asian region. These initiatives required investment in new facilities. It was hoped that the initiatives would make up for ongoing declines in some of Nestle’s important prod- uct offerings, particularly prepared dishes. Nestle was made up of seven major busi- ness units: powdered and liquid beverages (22% of total sales), water (8%), milk products and ice cream (18%), nutrition and health science (14%), prepared dishes and cooking aids (15%), confectionary (11%), and pet care (12%). The food process- ing industry had recently seen a substantial decline in demand for its products in the developing world. Important macroeconomic factors had led to sizable declines in demand from this part of the world. The softening of growth had led to increased competitive pressures within the industry that included such food giants as Mondelez, Tyson, and Unilever.400 Part Two Financial Analysis and Forecasting Based on this simple business and industry assessment, we take the view that Nestle will maintain its position in a deteriorating industry. We can adjust the naive 2014 forecast based on that assessment (Exhibit 5.3). We suspect that the softening of demand in developing markets and the prepared dishes line will lead to zero sales growth for Nestle in 2014, We also expect the increased competition within the industry will increase amount spent on operating expenses to an operating expense-to-sales ratio of 35%, Those assumptions give us an operating margin estimate of 12.9%. We expect the increased competition to reduce Nestle’s ability to work its inventory such that inventory turnover returns to the average between 2012 and 2013 of 5.53. We project PPE turnover to decline to 2.8% with the increased investment in new facilities that are not yet operational. Those assumptions lead to an implied financial forecast. The result- ing projected after-tax ROA is 6.3%. The forecast is thoughtful. It captures a coherent view of Nestle based on the company’s historical financial relationships, a grounding in the macroeconomic and industry reality, and the incorporation of Nestle’s specific busi- ness strategy. We recognize that we cannot anticipate all the events of 2014. Our forecast will : inevitably be wrong. Nevertheless, we suspect that, by being thoughtful in our analysis, our forecast will provide a reasonable, unbiased expectation of future performance. Exhibit 5.4 gives the actual 2014 results for Nestle. The big surprise was that the effect of competition was worse than anticipated. Nestle’s realized sales growth was actually negative, and its operating margin dropped from 14.9% and 14.1% in 2012 and 2013, respectively, to 11.9% in 2014. Our asset assumptions were fairly close to the outcome, although the inventory turnover and PPE turnover were a little worse than we had expected. Overall, the ROA for Nestle dropped from 7.1% in 2013 to 5.3% in 2014. Although we did not complete a high-side and a low-side scenario in this simple example, we can hope that, had we done so, we could have appropriately assessed the sources and level of uncertainty of our forecast. Appendix To test for forecasting bias among business school forecasters, an experiment was per- formed in 2005 with the 300 first-year MBA students at the Darden School of Business at the University of Virginia, Each student was randomly assigned to both a U.S. public company and a year between 1980 and 2000.’ Some students were assigned the same company, but no students were assigned the same company and the same year. The students were asked to forecast sales growth and operating margin for their assigned company for the subsequent three years. The students based their forecasts on the >More precisely, the population of sample firms was all U.S. firms followed by Compustat and the Value Line Investment Survey. To ensure meaningful industry forecast data, we required that cach firm belong to a ‘meaningful industry, which isto say that multiform, industrial services, and diversified industries were not considered). We also required that Value Line report operating profit for each firm. To maintain consistency in the representation of firms overtime, the sample began with a random identification of 25 firms per year ‘The forecast data were based on Value Line forecasts during the summer of the frst year of the forecast. All historical financial data were from CompustatCase 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting 101, following information: industry name, firm sales growth and operating margin for the previous three years, historical and three-year prospective industry average growth and margins, and certain macroeconomic historical and three-year forecast data (real gross national product [GNP] growth, inflation rates, and the prevailing Treasury bill yield). To avoid biasing the forecasts based on subsequent known outcomes, students were given the name of their firm's industry but not the firm’s name. For the same reason, the students were not given the identity of the current year. The responses were submitted electronically and anonymously. Forecast data from students who agreed to allow their responses to be used for research purposes were aggregated and analyzed. Summary statistics from the responses are presented in Figure 5.3. ‘The median values for the base-case forecast of expected sales growth and operat- ing margin are plotted in Figure 5.3. The sales growth panel suggests that students tended to expect growth to continue to improve over the forecast horizon (years 1 through 3). The operating margin panel suggests that students expected near-term per- formance to be constant, followed by later-term improvement. To benchmark the fore- cast, we can compare the students’ forecasts with the actual growth rates and operating margins realized by the companies. We expect that if students were unbiased in their forecasting, the distribution of the forecasts should be similar to the distribution of the actual results. Figure 5.3 also plots the median value for the actual realizations. We observe that sales growth for these randomly selected firms did not improve but stayed fairly constant, whereas operating margins tended to decline over the extended term. The gap between the two lines represents the systematic bias in the students’ forecasts. Because the bias in both cases is positive, the results are consistent with systematic optimism in the students’ forecasts, By the third year, the optimism bias is a large 4 per- centage points for the sales growth forecast and almost 2 percentage points for the margin forecast. FIGURE 5.3 | Median expected and actual financial forecast values for a random sample of U.S. companies. This figure plots the median forecast and actual company realization for sales growth and operating margin over the three-year historical period and the three-year forecast period based on the responses from MBA students in an experiment. OO Sales Growth Operating Margin 10% 15% 9% aa oe 8% aa 14% we % 22. Forecast +? Forecast 6% 13% ee 5% Actual 4% 2% 3% ~ % 1% — % on 10% ey eas Forecast Year Forecast Year ——_}.:.$ $A EEO einio2 Pan Two Financial Analysis and Forecasting FIGURE 5.4 Although the average student tended to exhibit an optimistic bias, there was varia- tion in the bias across different groups of students. The forecast bias was further exam- ined across two characteristics: gender and professional training. For both sales growth and operating margin, the test results revealed that males and those whose professional backgrounds were outside finance exhibited the most optimistic bias. For example, the bias in the third-year margin forecast was 0.7% for those with professional finance backgrounds and 1.9% for those outside finance; and 2.6% for the male students and just 0.8% for the female students. In generating forecasts, itis also important to have an unbiased appreciation for the precision of the forecast, which is commonly done by estimating a high-side and a low-side scenario. To determine whether students were unbiased in appreciating the risk in forecast outcomes, they were asked to provide a high-side and a low-side sce- nario, The high-side scenario was defined explicitly as the 80th percentile level. The low-side scenario was defined as the 20th percentile level. Figure 5.4 plots the median high-side and low-side scenarios, as well as the expected base-case forecast presented in Figure 5.3. For the three-year horizon, the median high-side forecast was 4 percent- age points above the base case and the low-side forecast was 4 percentage points below the base case. The actual 80th percentile performance was 8 percentage points above the base case and the actual 20th percentile was 12 percentage points below the base case. The results suggest that the true variance in sales growth is substantially greater than that estimated by the students. The same is also true of the operating margin. The estimates provided by the students are consistent with strong overconfidence (negative variance bias) in the forecast. | Median base-case, high-side, and low-side forecasts versus the actual 20th and 80th performance percentiles for sales growth and operating margin, This figure plots the median base-case, high-side, and low-side forecasts for sales growth and operating margin over the three-year forecast period based on the responses from MBA students in an experiment. The low-side and high-side performance levels were defined as the students’ estimate of the 20th and 80th percentile levels, The actual company 20th and 80th performance percentiles for sales growth and operating margin are also plotted. Sales Growth Operating Margin cael — ‘Actual 80th Percentile ae : = ‘Actual 80th Percentile 15% oom High-side Forecast 7 waeee-- 18% ares e Case Forecast Base Case Forecast 5% oe Low-side Forecast ee ae 7 10%, Actual 20th Percentile Fetagy 20th Percent a 8% 5% m : 1 2 3 7 2 3 Forecast Year Forecast YearCase 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting 103 EXHIBIT 5.1. | Financial Statements for Nestle SA (in billions of Swiss francs) 2012 2013 (1) Sales 899 924 (2) Cost of sales 47. 48.4 (3) Gross proft 424 443 (4) Operating expenses 290 312 (5) Operating profit 13.4 134 (6) Net interest expense 07 06 (7) Proft before taxes 127 124 (8) Taxes 33 33 (9) Profit after taxes 94 92 (10) Accounts receivable 134 122 (11) Inventory 89 84 (12) Other current assets 120 95 (13) Net property plant, and equipment 327 310 (14) Other non current assets 59.2 593 (15) Total assets 1259 1208 (16) Accounts payable 146 16.4 (17) Short-term debt 184 114 (18) Other current lables 56 55 (19) Curent liabilities 386 329 (20) Long-term debt 9.0 10.4 (21) Non current labities 156 130 (22) Book equity 627 (23) Total liabilities and equity 1258 1208 Sales growth 7.4% 27% Gross margin (3/1) 47.2% 47.9% Operating exp/Sales (4/1) 323% 33.8% Operating margin 5/1) 149% 14.1% Interest expense/Debt (6(17+20)) 26% 29% Tax ate (8/7) 25.7% 26.2% Receivable turnover (1/10) 69 76 ‘Accounts receivable days (10/1°365 days) 53.0 432 Inventory turnover (2/11) 53 57 lventory days (11/2°365 days) 687 636 Other current assets/Sales (12/1) 13.4% 10.3% PPE turnover (1/13) 28 30 Other noncurrent asset turnover (1/14) 15 16 Total asset tumover (1/15) o7 o8 Return on assets (5*(1~35)15) 6.9% 74% Accounts payable days (16/2"365 days) 112.4 1219 Other curr liab/Sales (18/1) 62%. 5.9% Non curr iab/Sales (21/1) 17.8% 14.1% emer ay Note: Although including both turnover and days ratios is redundant, doing solustrates the two perspectives104 Part Two Financial Analysis and Forecasting EXHIBIT 5.2 | Naive Financial Forecast for Nestle SA (in billions of Swiss francs) 201220132014 (1) Sales 299-924 949 Salest3*(1 + Soles growth) 2) Costof soles 475 48.1484 Sales t4- Gross proft (2) Grass prot 424 443 45S Sales t4* Gross margin (8) Operating expenses 280 312 320 _Salest4"[Operatng exprSales] (5) Operating promt 134 131134 —_ Gross proft— Operating expenses (6)Net interest expense 07 —_08 —_08 _[IntExpense/Debi}" (STD +LTD) (7) Pott betore taxes 127 124426 Opprofl- Netinterest exp (8) Taxes. 33 33 33 Tax rate * Profit before tax (9) Proft ater taxes 94 92 -93._—_—Proftbelore taxes - Taxes, (10) Accounts recelvable 131122, 125—_Salest a" AR doys/365 (14) venton 89 84 BG_—_Costofsaesinv turnover (12) Other current assets 1209597 Sales14* [Other cur assetsSsles} (13) Net property, plant, and equipment 327-310 31.9_—_Salest4/PPE turnover (14) Other non current assets 592 593 61.0 __ Sales 4/Other NC asset turnover (15) Total assets 139 iz05 1237 (16) Accounts payable 146 164.165 CostofSles14* AP days/365 (17) Short-term debt 184 1417.2. Pug=TAAP-OCALTD-NCLBE (18) Otner current abies 56 _55 _86 _Solest4* [Other curlab/Saes] (19) Curent bites 336029393 (20) Longterm debt 90 104 = 10.4 Maintain 13 (21) Non cunt tabtes 156 130-56 _Sales14* Non curr labSales] (22) Book equiy 627 641685 «BETS +PAT-SDIV (23) Tota fables ond equity 158 12051237 Sales growth 7% 27% 27% Malntain'13 Gross margin (3/1) 472% 479% 479% Maintain’13 Operating exp/Saes 4/1) 323% 338% 338% Maintan"12 Operating margin (5/1) 149% 141% 14.1% Interest expensesDebt (617420) 266 29% 29% —Maintain"13 Tax ate (8/7) 257% 262% 26.2% —Maintain’13 Recelvable tunover (1/10) i is ‘Accounts receivable days (10/1*365 days) 53.0 48.2 48.2 Maintain "13, Inventory turnover (2/11) 530 57) 87—Malnain'13. Inventory days (11/2"365 days) 687 636.638 Other curent assets Sales (12/1 134% 103% 10.3% Maintain'13 PE turnover (1/13) 28 303.0 Maiman 43 Othernoncurent asset turnover (1/14) 15 18 16 Maintain'13 Tota asset turnover (1/15) 07 = 08 OB Maintain 13 Return on assets (5*(1~.35/15) 6.9% 71% TN% Accounts payable days (16/2"365 days) 112412191219 Maintain'13 Other cur iab/Sales (18/1) 62% 59% 59K Maintin'13 Non curr lisbSales (21/1) 174% 144% 14.1% —Maintain'13Case 5 Business Performance Evaluation: Approaches for Thoughtful Forecasting 105, EXHIBIT 5.3 | Revised Financial Forecast for Nestle SA (in billions of Swiss francs) 2012 20132014 (1) Sales 899 924 924 Salest3* (1 + Sales growth) (2) Cost of sales 475, 48.1 48.1 Salest4 — Gross profit (3) Gross proft 424 443-443 Sales14*Gross margin (4) Operating expenses 29.0 312 32.3 Sales4 * [Operating exp/Sales} (6) Operating proft 13.4 13.4 11.9 Gross profit ~ Operating expenses (@)Net interest expense o7 Os 0.8 nt. Expense/Debt]* (STD + LTD) (7) Profit before taxes 127 124 11.4 Op profit Net interest exp (@) Taxes 33 33 29 Tax rate * Profit before tax (9) Proft after taxes 94 92 &2 Profit before taxes - Taxes (10) Accounts receivable 13.4 122 122 Salest4* AR days/365 (11) Inventory 89 a4 8.7 Cost of sales/inv tumnover (12) Other current assets 120 95 95 Sales14 [Other curr assets/Sales) (13) Net property, plant, and equipment 327 31.0 33.6 Sales 4/PPE turnover (14) Other non current assets 502 59.3 593 Sales 4/Other NC asset turnover (15) Total assets 1259 12051233 (16) Accounts payable 146 164 161 Cost of Salest4* AP days/365 (17) Short-term debt 18.4 14 186 Plug = TA-AP-OCALTO-NCL-BE (18) Other current liabilities 56 55 5.5 Sales14* [Other cur liab/Sales) (19) Current liabilities 386 329 402 {20} Long-term debt 90 10.4 104 © Maintain'13 (21) Non current labiities 156 130 5.5 Sales14 *[Non curr liab/Sales) (22) Book equity 627 64.1 673 BEI3+PAT-SDIV (23) Total abilties and equity 125912051233 Sales growth 74% 27% 0.0% —Nogrowthin “14 Gross margin (3/1) 472% 47.9% 47.9% —Maintain'13 Operating exp/Sales (4/1) 323% 33.8% 35.0% __ Increased competition Operating margin (5/1) 149% 141% 129% Interest expense/Debt (617+20)) 26% = 2.9% 2.9% — Maintain "13 Tax rate (8/7) 25.7% 26.2% 26.2% — Maintain’13 Receivable turnover (1/10) 69 16 76 ‘Accounts receivable days (10/1"365 days) $3.0, 48.2 48.2 Maintain'13, Inventory turnover (2/11) 53 87 5.53 Average turnover more appropriate Inventory days (11/2°365 days) 687 636 660 Other current assets/Sales (12/1) 134% 10.3% — 10.3% —Maintain"13 PPE turnover (1/13) 28 30 2.8 Average turnover more appropriate Other noncurrent asset turnover (1/14) 15 16 1.6 Maintain'13 Total asset turnover (1/15) 07 os 0.7 Maintain 13 Retum on assets (51-3515) 6% 71% 63K Accounts payable days (16/2"365 days) 1124 121.9 121.9 Maintain'13 Other curr liab/Sales (18/1) 62% 59% 5.9% — Maintain'13 Non curr liab/Sales (21/1) 17.4% = 14.1% — 14.1% — Maintain "13 ee —106 Part Two Financial Analysis and Forecasting EXHIBIT 5.4 | Actual Financial Performance for Nestle SA (in billions of Swiss francs) RET OE eee ee Non curr liab/Sales (21/1) 17.4% 14.1% 17.7% 2012 2013 20140 (1) Sales 899 924 919 (2) Cost of sales 478 48.4 416 (3) Gross proft 424 443 443 (4) Operating expenses 290 312 334 (6) Operating prof 13.4 134 109 {6) Net interest expense 07 06 06 (7) Profit before taxes 127 124 103 (8) Taxes 33 33 (9) Profit after taxes 94 92 (10) Accounts receivable 13.1 122 (14) Inventory 89 a4 (12) Other current assets 120 95 (13) Net property plant, and equipment 327 310 (14) Other non current assets 59.2 59.3 (15) Total assets 1259 1208 (16) Accounts payable 146 164 (17) Short-term debt 18.4 114 (18) Other current iabilties 56 55 (19) Current fabities 386 329 (20) Long-term debt 90 10.4 (21) Non current liablties 156 130 (22) Book equity 627 64.4 n9 (23) Total liabilities and equity 1259 1205 1335 Sales growth 7.4% 27% 0.6% Gross margin (3/1) 47.2% 47.9% 48.2% Operating exp/Sales (4/1) 323% 33.8% 36.4% Operating margin (5/1) 149% 141% 11.9% | Interest expense/Debt (6(17 + 20) 26% 29% 3.0% Tax rate (8/7) 25.1% 26.2% 328% Receivable turnover (1/10) 69 76 68 ‘Accounts receivable days (10/1"365 days) 53.0 48.2 535 Inventory turnover (2/1 1) 53 87 52 Inventory days (11/2°365 days) 687 636 70.4 Other current assets/Sales (12/1) 13.4% 10.3% 123% PPE turnover (1/13) 28 30 27 Other noneurrent asset tumover (1/14) 15 16 14 Total asset turnover (1/15) 07 o8 o7 Return on assets (5(1-.35)/15) 69% 74% 5.3% Accounts payable days (16/2°365 deys) 1124 1219 1339 Other curr liab/Sales (18/1) 6.2% 59% 72%