JULY/AUGUST 2021Join the team that has taught the certified training for NACVA and built successful consulting practices to expand your business. Today, we are excited to announce a new alliance with PocketExperts.Solutions to assist you, the NACVA accredited member, to attract new business clients and expand current client services offerings by providing you with: For more information about PocketExperts.Solutions visit PocketExperts.Info • A cutting-edge method to obtaining new engagements • Education on how to address complex issues across a broad spectrum of topics and plan the scope of work • Alternatives for engagement structure and pricing • Access to a growing panel of national and international subject area experts to join your team in demonstrating a proven track record to others • Use of the collaborative effect, integrating experts that stand ready to assist your team with execution of the work • Training for your staff in value growth advisory services • Training for your staff in management consulting services • Referrals for firm coaching engagements • A voice on a global platform on topics of interest that you select • A way to differentiate your firm in the market EXPAND THE CVA & MAFF SERVICE OFFERINGSA PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES the value examiner JULY | AUGUST 2021 3 on the cover in this issue … 24 19 ValExCov_J-A_2021comps_v1.indd 67/13/21 9:47 AM 6 Letters to the Editor In the May/June 2021 issue of The Value Examiner, authors Kipp A. Krukowski, PhD, CVA, ASA (retired), and Lawrence Justin White, Jr., PhD, CVA, offered “A Call for Industry Specialization: An Academic Perspective on the Importance of Qualitative Research.” In a letter to the editor, “In Defense of Generalists,” Judith H. O’Dell, CPA, CVA, offers a different perspective. The authors’ response follows. GDPR and the Increasing Cost of Cybersecurity By Dorothy Haraminac, MBA, CFE, MAFF, PI I t is critical for business valuators to scrutinize cybersecurity as part of their risk assessments of subject companies. This article reviews an important component of cybersecurity risk: the increasing risk of fines under the EU’s General Data Protection Regulation (GDPR) and similar laws and regulations. Estimating Business Value with Bayesian Networks By Kurt S. Schulzke, JD, CPA, CFE Given wide plausible value ranges, the greatest value that a valuation expert offers a client may be the ability to persuade others (e.g., judges) to locate their preponderance of probabilities across the client’s interval within the plausible range. This article explores Bayesian networks as a platform for facilitating the probabilistic estimation, negotiation, and communication of business value. departments ACADEMIC REVIEW Academic Research Briefs Guest Reviewer: Philipp Schaberl, PhD This column provides readers with summaries of contemporary research in valuation and forensic accounting. Summarized manuscripts—selected from numerous academic research outlets— cover significant developments that affect the ever-changing valuation and forensic accounting landscape. The objective is to increase awareness of recently completed research that advances knowledge of these subjects. 28A PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES 4 JULY | AUGUST 2021 the value examiner EDITORIAL STAFF CEO & Publisher: Parnell Black Editor: Daniel Shiffrin, JD Associate Editor: Lynne Johnson EDITORIAL BOARD Chair: Lari B. Masten, MSA, CPA, ABV, CFF, CVA, ABAR, MAFF Past Chair: Michael Goldman, MBA, CPA, CVA, CFE, CFF Ashok Abbott, MBA, PhD John E. Barrett Jr., MBA, CPA, ABV, CVA, CBA Gary W. Baum, MBA, CPA, CVA Neil J. Beaton, CPA, ABV, CFF, CFA, ASA Rod P. Burkert, CPA, CVA Lorenzo Carver, MS, MBA, CVA Wolfgang Essler, CVA (Germany) Richard W. Goeldner II, ASA, CBA, CVA Dorothy Haraminac, MBA, CFE, MAFF, PI Hubert Klein, CPA, ABV, CVA, CFE, CFF Andrew M. Malec, PhD Michael J. Molder, JD, CPA, CFE, CVA, MAFF Judith H. O’Dell, CPA, CVA Michael D. Pakter, CPA, CFF, CGMA, CFE, CVA, MAFF, CA, CIRA, CDBV Danny A. Pannese, MST, CPA, ABV, CVA, CSEP Kevin A. Papa, CPA, CVA, ABV, CVGA Donald Price, CVA, ASA Angela Sadang, MBA, CFA, ASA, ABV Keith Sellers, CPA, ABV Todd Zigrang, MBA, MHA, FACHE, CVA, ASA The Value Examiner ® is a publication of: National Association of Certified Valuators and Analysts ® (NACVA ® ) 1218 East 7800 South, Suite 302 Sandy, UT 84094 Tel: (801) 486-0600, Fax: (801) 486-7500 E-mail: NACVA1@NACVA.com NACVA members are automatically provided a subscription to The Value Examiner with membership. If you do not want to receive this publication, upon request, we will reduce your annual dues by $25. The Value Examiner ® departments SUBMISSION DATES Issue Submission Publish Dates Dates Nov./Dec. Aug. 18 Dec. 1, 2021 Jan./Feb. Oct. 19 Feb. 1, 2022 Mar./Apr. Dec. 15 Apr. 1, 2022 ALL SUBMISSIONS The Value Examiner is devoted to current, articulate, concise, and practical articles in business valuation, litigation consulting, fraud deterrence, matrimonial litigation support, mergers and acquisitions, exit planning, and building enterprise value. Articles submitted for publication should range from 800 to 4,000 words. Case studies and best practices are always welcome. SUBMISSION STANDARDS Manuscripts should be submitted via the Scholastica professional journal management platform. For more information, or to submit an article, please visit: https:// www.nacva.com/tveauthors. By clicking on the "Submit via Scholastica" button, you can view detailed editorial and submission guidelines. If you have questions, please contact Dan Shiffrin, Editor, at DanS1@NACVA.com, or Lynne Johnson, Associate Editor, at LynneJ1@NACVA.com. REPRINTS Material in The Value Examiner may not be reproduced without express written permission. Article reprints are available; call NACVA at (800) 677-2009 and/or visit the website: www.NACVA.com. Production: Mills Publishing, Inc.; President: Dan Miller; Office Administrator: Cynthia Bell Snow; Art Director/Production Manager: Jackie Medina; Magazine Designer: Jackie Medina; Graphic Designers: Ken Magleby, Patrick Witmer; Advertising Representatives: Paula Bell, Dan Miller, Paul Nicholas Mills Publishing, Inc., 772 East 3300 South, Suite 200, Salt Lake City, Utah 84106, (801) 467-9419. Inquiries concerning advertising should be directed to Mills Publishing, Inc. Copyright 2021. For more information please visit millspub.com. Cover designed by: Chris Peterson, Creative Director, Digital Paint Booth, digitalpaintbooth.com 2021 AWARDS FOR PUBLICATION EXCELLENCE PRACTICE MANAGEMENT Practicing Solo: Andrew Park By Rod P. Burkert, CPA, CVA The author interviews sole practitioner Andrew Park, CPA, from New York, NY. 46 SMALL BUSINESS FOCUS Size Matters: Valuation of Small and Micro Businesses By Gregory R. Caruso, JD, CPA, CVA This column focuses on valuation issues unique to very small or “micro” businesses. These businesses often have less financial and management information available, much of which may be deficient by GAAP or other standards. Therefore, valuators must do more qualitative review and apply greater professional judgment. This issue’s installment examines “Valuing Small Businesses Using the Income Approach: Two Common Mistakes.” 40 35 HEALTHCARE INSIGHTS Valuation of Telemedicine: Introduction (Part I of V) By Todd Zigrang, MBA, MHA, FACHE, CVA, ASA, and Jessica Bailey- Wheaton, Esq. Telemedicine has rapidly advanced over the past couple of decades, and its advancement has been significantly accelerated since the COVID-19 pandemic struck the U.S. In this first installment of a five-part series on the valuation of telemedicine, the authors provide a description of telemedicine, an overview of its role during the COVID-19 public health emergency, and the potential challenges and opportunities it may face in the future.Duff & Phelps Cost of Capital Navigator: U.S. Cost of Capital Module PRICING U.S. Cost of Capital Module Basic (two most recent years) Single User . . . . . . . . . . . . . . . . . . . . .$310 . . . . . . . . .+$129 per Basic additional user U.S. Cost of Capital Module Pro (all years, 1999 to present) Single User . . . . . . . . . . . . . . . . . . . . .$595 . . . . . . . . . . +$299 per Pro additional user *The U.S. Cost of Capital Module Basic subscription has EVERYTHING contained in the regular Basic U.S. Cost of Capital Module, but limits the number of cost of capital analyses/downloads based on your subscription level. The limits are three cost of capital analyses/downloads for Silver, seven for Gold, and ten for Platinum. Data is updated quarterly. Duff & Phelps Cost of Capital Navigator: U.S. Cost of Capital Module (Basic*) NOW INCLUDED IN KeyValue Data ® Titanium continues to include the U.S. Cost of Capital Module (Pro). Silver , Gold , or Platinum subscriptions. Duff & Phelps Cost of Capital Navigator: U.S. Cost of Capital Module DATA INCLUDED IN THE U.S. COST OF CAPITAL MODULE SIZE PREMIA AND RISK PREMIA g CRSP Deciles 1–10 size premia, plus the 10th Decile split g Risk Premium Report Size Study size premia and “risk premia over the risk-free-rate” g Risk Premium Report Risk Study “risk premia over the risk-free rate” g High-Financial-Risk Study size premia and “risk premia over the risk-free rate” g Comparative Risk Study (fundamental risks of companies comprising Risk Premium Report Size Study portfolios) RISK-FREE RATES g Spot long-term risk-free rates (from the Federal Reserve, updated daily) g Duff & Phelps long-term normalized risk-free rate (adjusted for impact of flights-to-quality, actions of the Federal Reserve, and inflation expectations) U.S. EQUITY RISK PREMIA (ERPS) g 1926–present Historical (i.e., realized) ERP g 1926–present Supply-side ERP g Duff & Phelps Recommended ERP (reflecting current point in business cycle) “SIZE” TABLES Includes CRSP Deciles Size Study and Risk Premium Report size premia tables that are viewable within the Navigator. ESTIMATE WACC Develop into weighted average cost of capital (WACC) estimates. ENHANCED OUTPUTS Export more robust and comprehensive documentation of the user’s cost of capital assumptions, sources, analysis, and results as PDF or in Excel. EXCEL ADD-IN This powerful new tool enables users to directly import cost of capital data in their own Excel spreadsheets. BETAS g Industry-level betas (Full- Information, Vasicek-Adjusted, Sum, and High-Financial-Risk Betas) g Industry Risk Premia for use in the Build-up Method g Debt betas (for unlevering and relevering betas) COMPANY LISTS g List of companies used to calculate Full-Information Betas and Industry Risk Premia (updated quarterly) g List of companies used to calculate “Pure-Play” U.S. industry statistics (updated quarterly) The Duff & Phelps Cost of Capital Navigator is an interactive, web-based platform that guides valuators through the steps of calculating Cost of Capital, a key component of any valuation analysis . To learn more, visit www.NACVA.com/store_home.asp or call Member/Client Services at (800) 246-2488. NEW ENHANCED FEATURES NEW! Now Included in KeyValueData Bundles at no additional cost, OR BUY standalone! 1 1 4 5 2 3 2 3 4A PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES 6 JULY | AUGUST 2021 the value examiner I n Delaware statutory appraisal cases, the parties “have the burden of proving their respective valuation positions by a preponderance of evidence.” 1 They typically try to carry this burden by hiring a valuation expert to persuade a “law-trained,” non-valuation- expert judge 2 who alone has the authority and obligation to determine the fair value after accounting for “all relevant factors.” 3 In this endeavor, the experts frequently fail, as evidenced by courts characterizing expert valuations as “widely divergent,” “bracketing the outer limits of plausibility,” or “ridiculously biased.” 4 In one recent case involving three valuation experts, the high discounted cash flow (DCF) valuation was eight times the low. Lamenting that “an optimist might [wrongly] assume that experts hired to examine the same company, analyzing the same set of financial data, would reach similar results,” the court wrote that “the best scenario is that one expert, at the least, is wildly mistaken.” It then used its own DCF estimate to peg the statutory fair value at precisely $98,783 per share. 5 To be fair, for continuous variables, probability attaches only to intervals, not points on a line. Thus, as a point estimate with zero probability of occurrence, even the court’s own valuation was a “mistake.” 6 Given the uncertainty and subjectivity inherent in business valuation, the intervals or probability distributions are wide, and objectively verifiable or “correct” values are illusory. For this reason, noted valuation evangelist Aswath Damodaran advocates simulation- generated probability distributions, because they provide “not only everything you would get in a standard valuation (an estimated value for your company) but [also] additional output (on the variation in that value and the likelihood that your firm is under- or over-valued).” 7 Given wide plausible value ranges, the greatest value that a valuation expert offers a client may be the ability to persuade others (e.g., judges) to locate their preponderance of probabilities (evidence) across the client’s interval within the plausible range. Accomplishing this feat is a function of technical valuation expertise, as well as communication tools and techniques. This article explores Bayesian networks (BNs) as a platform for facilitating the probabilistic estimation, negotiation, and communication of business value. What Are Bayesian Networks and Why Use Them? The author’s November/December 2019 piece, Estimating Economic Damages with Linear Regression and Bayesian Networks (Part II of II)—on which this article builds— provides essential background on the “what and why” of BNs. 8 Briefly, a BN is a “directed acyclic graph” (DAG) that VALUATION /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////// Estimating Business Value with Bayesian Networks By Kurt S. Schulzke, JD, CPA, CFE “Far better an approximate answer to the right question, which is often vague, than the exact answer to the wrong question, which can always be made precise.” - John Tukey 1 M.G. Bancorporation, Inc. v. Le Beau, 737 A.2d 513, 520 (Del. 1999). 2 “[T]he past and current members of this court are ‘law-trained judges,’ not valuation experts. Ironically for the petitioners' position that valuation is exclusively a matter for experts, the appraisal statute mandates that a lay individual express the final conclusion on the fair value of the petitioners' shares.” In re Dole Food Co., 114 A.3d 541, 555 (Del. Ch. 2014). 3 8 Del. C. § 262(h), https://delcode.delaware.gov/title8/c001/sc09/#262. 4 See Dole Food Co., 114 A.3d 541, 557–558 at n. 10 (string-citing opinions expressing skepticism regarding “widely divergent, litigation-driven expert valuations”). 5 In re ISN Software Corp., C.A. No. 8388-VCG (Del. Ch. 2016). 6 Mathematically, the probability associated with any point estimate is zero. In theory, a valuation with zero probability cannot meet the “preponderance of the evidence” threshold. See Kurt S. Schulzke, “Estimating Economic Damages with Linear Regression and Bayesian Networks (Part II of II),” The Value Examiner (November/December 2019): 22 (noting that the zero probability of point estimates should render them inadmissible in court). 7 Aswath Damodaran, Valuation: Lecture Note Packet 1, Intrinsic Valuation, slide 354 (updated January 2021), http://people.stern.nyu.edu/adamodar/ pdfiles/eqnotes/valpacket1spr21.pdf. 8 Schulzke, “Estimating Economic Damages” (see n. 6).A PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES the value examiner JULY | AUGUST 2021 7 visualizes probabilistic relationships between the variables, called “nodes,” (e.g., DCF value, cash flows, and discount rate) included in the BN. A BN can learn inputs from data (e.g., EBIT multiples database). However, where data are scarce or suspect, BNs readily incorporate judgmentally chosen variables, structure, and value inputs. Despite “big data” hype, judgmental models often outperform data-centric ones. Einstein published his world-altering Special Theory of Relativity with no data, in 1905. In 1986, NASA brass blew up the Challenger shuttle by using data (or lack thereof) to override the advice of Morton Thiokol’s rocket engineers, whose judgmental model predicted that the infamous O-ring seals would fail. 9 Whether the model is built around judgment, data, or a blend, the BN pours the model’s structure and input values into a transparent and thoroughly vetted sequence of mathematical operations 10 that typically include a junction-tree algorithm, an algebra of node probability tables, and dynamic discretization or Monte Carlo simulation. 11 The end result is a graphical network that clearly shows (visualizes) the probability distributions for each node (e.g., DCF value) conditioned on the other nodes (e.g., cash flow and discount rate). Thus, a BN portrays a joint probability distribution over plausible value ranges 9 See Challenger: The Final Flight (2020), https:// www.netflix.com/title/81012137. 10 A Google Scholar search of “Bayesian network” produced 312,000 pieces, 133,000 of which are dated 2000 or later. 11 See, e.g., Norman Fenton and Martin Neil, Risk Assessment and Decision Analysis with for all nodes in the model. This is no “black box.” Everything is on the table, open for debate. BNs are not a new valuation method. Rather, they act as a clarifying lens, quantifying and visualizing the impact of uncertainty on business value and updating that uncertainty in response to new information. Far from replacing existing market, asset, and DCF methods, BNs increase transparency and expand the value-related information delivered by them. Simple Illustration Figure 1 shows an illustrative BN, using purely hypothetical inputs, in which the target node is Guideline Public Co Value (Node C) and the explanatory nodes are EBIT (Node A) and EBIT Multiple (Node B). Figure 1: Market-Approach Valuation Using a Bayesian Network EBIT and guideline public company value are stated in millions of U.S. dollars. EBIT Multiple shows the prior distribution (blue), plus a vertical line marking the “posterior” observation of an EBIT Multiple of three, which shifts the plausible range of posterior values to the left. Bayesian Networks, 2nd ed. (Boca Raton, FL: CRC Press, 2018), Appendices B–D (detailing the algorithms and BN propagation process implemented in AgenaRisk). A C BA PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES 8 JULY | AUGUST 2021 the value examiner Because the inputs (Nodes A and B) are stochastic (i.e., randomly distributed), the target (Node C) is, as well. Node B also reflects the “entry” of evidence that the EBIT multiple is three. After entering “3” in Node B, the posterior probability distribution for Node C (green histogram) narrows markedly and shifts left, reflecting a lower, more precise range of plausible values. In Bayesian parlance, the “posterior distribution” is what you get after entering evidence into the BN. Entering evidence transforms a stochastic node into a deterministic one, where “deterministic” means that we claim or assume knowledge (i.e., 100 percent certainty) of that node’s value or state. Nodes are either “parents” or “children,” where a child is on the receiving end of an “arrow” (a.k.a., “directed edge”). A parent without a parent (e.g., Node A) is a “root” node. Inside each node resides a “node probability table” (NPT) that is either “conditional” or “marginal.” For child nodes, a conditional NPT specifies the probabilistic, conditional relationship between the child and parent(s). For root nodes, the NPTs are marginal. Whether conditional or marginal, the NPT might be viewed as the node’s DNA, borrowing from gene terminology. NPTs for Nodes A and C appear in Figures 2 and 3, respectively. In Figure 1, priors for all nodes are portrayed as histograms, showing the distribution of probabilities across the plausible values of each node prior to observation of evidence, with Nodes A and C stated in millions of dollars. The probability of each value interval equals the area of the histogram bin(s) located over the interval. Thus, in Figure 4, we see that by entering “3” in Node B, the posterior probability (green bars) of a firm value between $16.514 and 17.263 million = 0.085112 * (17.263 - 16.514) = 0.06374889, or about 6 percent. Figure 2: Marginal Node Probability Table for Node AFigure 3: Conditional Node Probability Table for Node C Showing prior mean (5), variance (2), lower bound (0.01), and upper bound (10.0) Showing C as the arithmetic product of parent Nodes A and B .A PROFESSIONAL DEVELOPMENT JOURNAL for the CONSULTING DISCIPLINES the value examiner JULY | AUGUST 2021 9 Figure 4: Probability of Value between $16.514 million and $17.263 Million 0.085112 * (17.263 - 16.514) = 0.06374889 Figure 5 shows Figure 1 with Node C formatted as a smooth line. The small boxes mark the location of the Node C “bins” in the Figure 1 histogram. In Node C, the vertical blue (prior) and green (posterior) median lines and contrasting dispersion of plausible values are more clearly distinguishable. With the EBIT multiple at three, the smooth green line in Node C tells us we can rule out values greater than $28 million, but values between $10 and 19 million are much more probable than before. Entering evidence transforms a stochastic node into a deterministic one, where “deterministic” means that we claim or assume knowledge.Next >