Valuation Publications Frequently Asked Questions

Transcription

1 Valuation Publications Frequently Asked Questions

2 Valuation Publications Frequently Asked Questions The information presented in this publication has been obtained with the greatest of care from sources believed to be reliable, but is not guaranteed. Morningstar, Inc. and providers of data to Morningstar, Inc. make no warranties, express or implied, as to the results to be obtained by the subscriber or others from the use of the data provided hereunder, and there are no express or implied warranties of merchantability or fitness for a particular purpose or use. The accuracy and completeness of the data are not guaranteed, and Morningstar, Inc. and its data providers shall have no liability for errors or omissions with respect to the data or its delivery regardless of the cause of such error or omission. In no event shall Morningstar, Inc. or its data providers have any liability for any indirect, special, or consequential damages, including but not limited to lost profits. Data contained hereunder is proprietary to Morningstar, Inc. and its data providers and is for customers internal use only; redistribution of the data is expressly forbidden. Copyright 2012 Morningstar, Inc. All rights reserved. No part of this publication may be reproduced or used in any form or by any means graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage and retrieval systems without written permission from the publisher. To obtain permission, please call or write to Morningstar, Inc., 22 West Washington St., Chicago, IL Specify the data or other information you wish to use, the manner in which it will be used, and attach a copy of any charts, tables, or figures derived from the information. There is a minimum $1500 charge per request. There may be additional fees depending on usage.

3 New Content from 2012 Ibbotson SBBI Valuation Yearbook: Supply-Side ERP: Q: Why are the Supply-Side and Historical Equity Risk Premia only 50 basis points apart? A: In general, the supply-side equity risk premium model is an adjusted version of the historical. They both contain the same fundamental factors and are not in conflict of one another. The difference between the two is that we remove the growth of price to three-year earnings (P/3E) beginning in 1926 in the supply-side model. This is removed because beginning in the 1980s, the S&P 500 market price began to grow in an unsustainable rate and was, arguably, artificially increasing the equity risk premium. Eventually this growth is expected to subside and return to normal levels. As the market self-adjusts, the growth of P/3E should diminish and, therefore, the supply-side should shift back up to the historical ERP levels, which we have begun to see the last few years. Q: Where can I get the article mentioned in the yearbook? A: It is available online at Q: Why is a three-year average earnings being used instead of current-year earnings? A: Graph 5-9 of the 2012 Ibbotson SBBI Valuation Yearbook tracks the annual P/E and the price to three-year earnings (P/3E) since As can be seen in the graph, one-year P/E ratios can be very volatile because the earnings component can jump around from year to year. The price to three-year earnings (P/3E) ratios in the same graph tell a much less volatile story and allow for more applicability. Premia calculated using the one-year ratios would be unpredictable and too volatile for practical use. The P/3E is measured as the current S&P 500 price over its average earnings from one year prior, the current year, and the future year. The date of the three-year earnings must revolve around the current date, which is the same date the price, in order to keep the denominator and numerator comparable. This normalizes earnings while keeping the date relevant. Q: Would the supply-side ERP still make sense if it began on a different date? A: The short answer is that it depends. The growth of price to three-year average earnings aspect of the supply-side equity risk premium is very dependent on the beginning and ending date of the data set. Using a different starting point than 1926 can drastically alter the assumptions made in the model. We chose 1926 because it was during a normal market environment and it is the beginning date of the majority of our other data, making it compatible with other premia. Q: Can we see the same type of analysis for the supply-side equity risk premium as the historical has in Appendix A of the Ibbotson SBBI Valuation Yearbook? A: The supply-side ERP does not have the same wedge table analysis as the historical ERP because the supply-side model is heavily dependent on the beginning and ending date of the data set. If we allow for a variable beginning date, we could drastically alter the key assumptions of the model, depending on which beginning date is used. We agree with the model s assumption that the price to three-year average earnings level in 1926 is an appropriate base for the growth estimate. 3

4 10th Decile Analysis: Q: What is an acceptable probability of default, as shown in Table 7-15? A: Table 7-15 of the 2012 Ibbotson SBBI Valuation Yearbook shows the number of companies in the 10a, 10b, 10w, 10x, 10y, and 10z portfolios that fall at or above various probabilities of default. We provide these different probabilities of default in order for the reader to have a better understanding of the types of companies included in each portfolio. It is difficult to pinpoint a threshold in probability of default since it increases as the probability increases. Being able to see the degrees of probability of default provides the user with more information than a threshold would. Q: Why are there companies missing from the Company Movement table, Table 7-16? A: Table 7-16 of the 2012 Ibbotson SBBI Valuation Yearbook shows how many companies moved by decile from the December 2010 decile portfolio breakouts to the December 2011 breakouts. This table only tracks the companies that were included in the deciles for both years. Therefore, companies that were included in 2010 and not in 2011 are not included. New companies in 2011 are also not included in the analysis. Companies drop in and out of the indices for various reasons such as IPOs, delistings, acquisitions, etc. Q: Why does the company count per year per decile not match up exactly in Table 7-16? A: See the answer to the above question. Q: Why aren t the 27.4% of financial companies shown in Graph 7-4 removed from the 10z portfolio, like in Duff and Phelps s methodology? A: Graph 7-4 of the 2012 Ibbotson SBBI Valuation Yearbook shows the sector composition of the 10z portfolio. This graph shows that the most concentrated sectors are the financial services, technology, and healthcare sectors. We do not remove financial stocks from our size decile analysis. We are aware that another commercial source of U.S. size premium data, Duff & Phelps, excludes financial stocks from their size analysis. They are correct to remove financial companies in their company universe since a financial company s performance, as measured using fundamental data, is not comparable to companies in other sectors. For example, a commercial bank does not determine sales or debt in the same way that industrial and service firms do. This is why size premium that is calculated using fundamental measures of size should exclude financials from their analysis. The Morningstar size premium analysis uses market capitalization in representing company size. Financial firms should not be treated differently from firms in other industries when they are being compared only by market capitalization. If we were using other metrics to define size, such as sales, then financial firms would need to be excluded. Q: How can we actually use Table 7-7? A: In the 2012 Ibbotson SBBI Valuation Yearbook, Table 7-7 contains five measures of size as well as four ratios to present a broad look into the 10z portfolio s composition. Table 7-7 was created to provide the reader more descriptive analysis of the types of companies included in the 10z portfolio. Note that each column is calculating the percentile data independent of the others. Therefore, the 50th percentile company when sorted by sales data is not the same company that falls in the 50th percentile when sorted by market capitalization. It is meant to reflect the portfolio s composition and is not meant to be used as calculation inputs. 4

5 Q: Some of the size measures shown in Table 7-7 seem to show larger companies (over 2,000 employees, $1,175 in total assets, etc). Should these companies be included in the 10z portfolio? A: The majority of these companies with the unusual ratios or size measures belong in the Financial Services sector and therefore report fundamental data differently than companies in other sectors. These companies should still be included in the size premium calculation because, when viewed in terms of market capitalization, financial firms should not be treated differently from other sectors. Fundamental measures of size for a financial company, such as sales, are not comparable to companies in other sectors, in which case, it would be advisable to remove financials from the company universe. Although when market capitalization is used as a measure of size, it is applicable to all fairly valued companies. International Sector Betas: Q: Why didn t you use the local market benchmark? A: In our international sector analysis, we chose to use MSCI country indices instead of the local market for Australia, Canada, and the UK so that these betas can be used alongside the International Equity Risk Premium Report, which also uses MSCI indices for most countries. Also, using the same source for country index returns allows the sector betas to be comparable across countries as well as within each country. Q: Why are you publishing the beta for Australia s Communication Services and Utilities sectors since they contain a market share of less than 1% each? A: We chose to publish all available information for each country s sector, along with descriptive data, in order to provide our customers with as much information as possible. Any company set with less than 30 companies is considered a small sample and is less statistically reliable than a set of 30 or more companies; however the data can still be useful in creating analysis and comparisons across sectors and countries. 5

6 General Frequently Asked Questions: Size Premium: Q: There are four portfolios I can choose from when valuing my small company, which one of the 10 th decile options is right for me? A: Mike Barad wrote a very easy to understand article focused on choosing the most appropriate size premium. The article is located in this website: and it is called Size Matters: How to Apply Size Premium Metrics When Size-Based Category Breakpoints Overlap. Q: Do you have historical size premium data for Decile 10a and 10b for years prior to 2000? Can it be calculated? A. We first began publishing the 10a and 10b portfolios in the 2001 Valuation Yearbook. We do not have size premium data already published for Decile 10a and 10b for years prior to December 2000 but we can calculate this as a custom project. Contact our sales team to discuss timing and pricing at (888) Q: I know that Morningstar calculates U.S. size premia for domestic valuations and country risk premia for international valuations. Do you also calculate a country-specific size premia? A: We do not presently calculate size premia for any country other than the US. This is due to a lack of data availability. Equity Risk Premium: Q: Do the equity risk premia over time tables reflect arithmetic averages or geometric compounding? A: They reflect arithmetic averages. Q: Morningstar calculates the long-horizon equity risk premium as the difference between large company stocks total return and long-term government bond income return. Why doesn t the calculation subtract out the long-term government bond total return instead of income return? A: Total return is comprised of three return components: the income return, the capital appreciation return, and the reinvestment return. The income return is defined as the portion of the total return that results from a periodic cash flow or, in this case, the bond coupon payment. The capital appreciation return results from the price change of a bond index over a specific period. Bond prices generally change in reaction to unexpected fluctuations in yields. When a bond is sold and a new one is purchased to create a historical index with a consistent maturity, these price changes are incorporated in the bond index s total return. Reinvestment return is the return on a given month s investment income when reinvested into the same asset class in the subsequent months of the year. The income return is thus used in the estimation of the equity risk premium because it represents the truly riskless portion of the return and does not incorporate the price return produced through maintaining a constant-maturity bond index. 6

7 Industry Risk Premium: Q: Which beta is used in the calculation for Industry Premia estimates? A: We use Full Information Betas for the Industry Risk Premia and that methodology is described in Chapter 6 pages Q: Why is the industry risk premium different when I calculate it myself using the adjusted beta from Morningstar s Cost of Capital Industry Reports? A: We calculate the industry risk premium using full information betas instead of adjusted betas. Actually, the full information beta is incorporated in the adjusted beta calculation. The adjusted beta is a standard-error (of the raw beta) weighting of the raw and peer group beta. The higher the standard error of the raw beta is, the more the adjusted beta will look like the peer group beta. The less the standard error of the raw beta, the more the adjusted beta will look like the raw beta. Of course, the raw and peer group betas in this case are derived from market weighted excess returns (on a monthly basis); the peer beta is the one that has the full information beta values as additional inputs in order to come up with peer group weights. Q: Is the Full Information Beta levered or unlevered and/or raw? A: It is a Levered Raw Beta. International Reports: Q: Why are some cost of capital measures not available for all countries? A: The short answer is that we don t have the necessary data for the calculations. For a few models, the country indexes we use are MSCI (Morgan Stanley Country Index) or S&P/IFCG, which don t supply country-returns for every country. We have a pdf on the corporate site that lists what data we have and for which country so our customers are not surprised after buying the report. The link is We have a list for both the International Cost of Capital Report and the International Cost of Capital Perspectives Report. Q: Does the International Cost of Capital Report publish cost of capital figures? What is the targeted capital structure used in the WACC calculation? A: The International Cost of Capital Report uses 6 models in calculating the cost of equity of each country. While the report is named the Cost of Capital report, the published data are the cost of equity inputs to calculate the cost of capital. Therefore, we do not publish the cost of debt data nor the country s or a company s targeted capital structure. 7

8 Q: Is the Country Risk Rating from the International Cost of Capital Report a Cost of Equity? A: Page 3 of the report clarifies that the Country Risk Rating Model is a cost of equity model along with all the models used in the report. This model uses a regression of country credit ratings and market data to determine the forward looking cost of equity based off the most recent country credit ratings. Most of the countries we have listed do not have enough data to determine a meaningful traditional cost of equity model. Q: Does Morningstar have historical international equity risk premium data prior to 1970? A: We have data available prior to 1970 for only a few countries. The majority of our underlying data only goes back to 1970 so we would not be able to create an International ERP for those countries. To see how far our country data goes back, please refer to the reports methodology on this website: Q. Is currency risk and inflation incorporated in the international CAPM calculation? A: Pages 6-7 of the International Cost of Capital report are dedicated to the International CAPM calculation. For the Int l CAPM calculation specifically, we don t adjust due to currency risk or inflation. Instead the country beta is calculated with the MSCI World index as the benchmark and it uses a World ERP rather than US. So we don t incorporate the country risk rating into it either. Therefore the risk being calculated in this CAPM is the volatility of the country s returns compared to the world with the World ERP and the US risk free rate (because it s still from the US investor perspective). Q: Is inflation adjusted for in the country spread model and the country risk rating model? A: The country spread model uses the difference between the sovereign yield and the US LT Government Bond Yield with a date that s closest to the sovereign yield (so not necessarily the risk free rate). Then we add together the yield spread, the US ERP, and the US Risk-Free Rate to get the Cost of Equity for each country. So we don t make a direct adjustment for inflation. The country risk rating model is based off of Country Credit Risk ratings we get from Institutional Investor and the available country raw total returns. Then we determine the log and linear models of cost of equity using betas based of those returns and ratings and the logarithm of those ratings. So we don t include an inflation adjustment in this model either. 8

9 Q: I recently purchased the International Cost of Capital Perspectives Report. In this report, you suggest equity rates for investors from 6 countries investing in many countries. What risk premiums are taken into account in your rate models? A: The International Cost of Capital Perspectives report uses country credit ratings to estimate cost of equity. These cost of equity numbers are calculated using a regression, but does not use a risk-free rate or equity risk premium specifically. A better explanation of what s included in the regression is written on page 3 under Country Risk Rating Model. Since most countries do not have enough data for an equity risk premium, this Country Risk Rating Model is very helpful. While the model we use for the International Cost of Capital Perspective report provides a cost of equity for a typical company within the country, it is not recommended to use the cost of equity to derive an equity risk premium. It s ideal to either use our model as your base cost of equity, or find other supporting data (like Rf + ERP) to build up a cost of equity. Mixing and matching can create problems. Note that our model does not adjust for size, industry, or other specific risk premiums, it is just general equity risk for a typical company. At this point, we don't have any additional data to help make further size or industry adjustments in other countries. Some folks have tried to use U.S. data for this although we don t publish much to support this. It is up to the practitioner to make subjective adjustments for industry or size, if they are supportable. Q: Where in the model is the tax rate of each country taken into account. Is it in the credit rating model? A: Tax rates are not included in the calculations. The credit rating is derived from a rating system on a scale from 0 to 100 and we use that as one of the contributions in determining the cost of equity. The second paragraph under the Country Risk Rating Model on page 3 explains our approach in more detail. Q. Can you update the international cost of capital as of May? A. Yes. It would constitute as a custom project however only 4 of the 5 models can be updated as of the most recent month-end. The Country Spread, Int l CAPM, Globally Nested CAPM, and the Relative Standard Deviation Model can all be updated. The Country Credit Ratings can only be updated for March and September. 9

10 Composition of our Series: Q: Could you provide the definition or methodology for the IA SBBI US Inflation? A: The Consumer Price Index for All Urban Consumers (CPI-U), not seasonally adjusted, is used to measure inflation, which is the rate of change of consumer goods prices. Unfortunately, the CPI is not measured over the same period as the other asset returns. All of the security returns are measured from one month-end to the next month-end. CPI commodity prices are collected during the month. Thus, measured inflation rates lag the other series by about onehalf month. Prior to January 1978, the CPI (as opposed to CPI-U) was used. For the period 1978 through 1987, the index uses the year 1967 in determining the items comprising the basket of goods. Following 1987, a three-year period, 1982 through 1984, was used to determine the items making up the basket of goods. All inflation measures are constructed by the U.S. Department of Labor, Bureau of Labor Statistics, Washington. For an easier breakout: From 1926 thru 1987 US DOL BLS Series ID: CUUR0000AA0 Description: CPI for All Urban Consumers: All Items (Not Seasonally Adj; Old Base 1967=100) From 1988 thru Present US DOL BLS Series ID: CUUR0000SA0 Description: CPI for All Urban Consumers: All Items (Not Seasonally Adj; Base =100) When CPI changed their default series, Ibbotson decided to alter theirs on an on-going basis. At the time, they were hesitant to change their data series so as not to alter past calculations and such. Now Ibbotson s SBBI US Inflation series is a combination of the most relevant inflation data for the specific time periods. This way, if we were to value a company in 1980, we would be able to do it using the same data that would have been available during Q: Are taxes are being considered in calculating excess return, as shown in Chapter 4 of the Classic Yearbook, or is it free of taxes? A: Excerpt from the Ibbotson SBBI Classic Yearbook: This chapter presents the returns for the seven basic asset classes and describes the construction of these returns. More detail on the construction of some series can be found in the January 1976 Journal of Business article, referenced at the end of the Introduction. Annual total returns and capital appreciation returns for each asset class are formed by compounding the monthly returns that appear in Appendix A. Annual income returns are formed by summing the monthly income payments and dividing this sum by the beginning-of-year price. Returns are formed assuming no taxes or transaction costs, except for returns on small company stocks that show the performance of an actual, tax-exempt investment fund including transaction and management costs, starting in Cost of Capital Yearbook: Q: What is the cut off between the Large Composite and the Small Composite? A: The Large and Small Composites are the composed of either the largest 3 companies and the smallest 3 companies, respectively, or the top (or lowest) 10% of the companies during the most recent fiscal year. So basically if there are 100 companies in the industry, the largest composite would be composed of the 10 largest companies. The composites are only published if there are more than 10 companies within the industry, otherwise we don t consider it meaningful. 10

11 Q: Why does the company set for the SIC industries vary between the Cost of Capital Yearbook and the Industry Risk Premia in the Ibbotson SBBI Valuation Yearbook? A: The biggest difference is that the Cost of Capital yearbook s industries only include companies that have at least 75% of their sales whereas the industry premia in the Valuation yearbook include any company that has any sales linked to that SIC. We only include companies with 75% of their sales in the specific industry in the Cost of Capital yearbook to remove companies that could skew the data. A good example of this would be a large company like Kraft Foods, which has sales linked to multiple industries. Kraft is a major player within some of these industries with its contributed sales making up almost 50% of the total industry sales but, as a percent of total company sales, it may not be that much. In this case, Kraft is such a big player that it should be included in the industry risk premium, although information from Kraft s financial statements would skew the industry multiples and margins that are listed in the Cost of Capital industry pages since the majority of Kraft s sales are not attributed to that industry. An example is further explained in Table 6-9 in the 2012 Ibbotson SBBI Valuation Yearbook. 11