Jeffrey Dorman: Statistical Evidence is a Powerful Weapon, Not a Magic Bullet

 Written by Tom Hagy

 

There are endless uses for statistical evidence.  It is great for estimating damages, providing quantitative proof of the extent to which conduct had an impact, or rebutting or cross examining an expert.  But, attorney Jeffrey Dorman says, “Just as important to knowing how to use statistical evidence is knowing when not to use it.”

 

Speaking on a live recording for Mealey’s Litigation Conferences and Business Valuation Resources LLC, Dorman said statistical evidence is not a magic bullet that should be used in every case.  “Sometimes it’s inappropriate, irrelevant or even inadmissible.” 

 

And if anyone should know it would be Dorman.  In addition to being a 30-year litigator and a partner with Freeborn & Peters LLP, he has extensive training and experience in statistical estimation, mathematical modeling, parametric estimating, system dynamics programming, and financial analysis. 

 

As a general rule it is only necessary to use damages experts when it is necessary to estimate damages, Dorman said.  In many cases there is no need to estimate, such as in cases of liquidation.  When a contract is broken, the damages generally comprise the cost of covering the contract.  In the case of theft, generally speaking it is the cost of the item stolen, unless it is something rare or unique.  In both cases, damages can usually be computed directly, without the need for estimation, and can be presented through a fact witness, as opposed to an expert witness.

 

In contrast, where damages depend upon future economic variables like profits, sales, demand and cost, or future commissions, competitive harm, or patent infringement, there is no way to avoid the use of a damages expert, he said.

 

In between these categories you have the cusp, Dorman said, like a class action where back pay is involved and data are available to directly calculate the lost back pay of each class member.  When in this situation, he said, if feasible, you should perform the direct calculation of damages; this will be better than a class-wide estimate based on regression sample because even a good approximation will miss the true value in virtually every instance.  A properly chosen random sample, while methodologically acceptable, will have sampling errors. 

 

Dorman warned:  If you chose to base your damages calculation on a sample and the opposing party correctly performs a direct calculation, there will be credibility problems because “an estimate is just that – an estimate.”  However, if the class members are in the tens of thousands so that it is impractical to perform a direct calculation, the use of a sample might be the only practical way to assess damages. 

 

It is a judgment call as to whether to perform a direct calculation or estimation, he went on.  You have to look at what data are needed, whether the data are available, and whether there is someone who can perform the direct calculation.  You also need to examine the cost and the likelihood that the opposing party will perform a direct calculation. 

 

Besides expenses there are other tactical reasons to consider when employing a financial expert.  You need to weigh the necessity against the impact on discovery  – the subject of another post.   

 

Dorman spoke for Mealey’s Litigation Conferences and BVR along with Dr. G. William Kennedy PhD, CPA/ABV with Anders Minkler & Diehl, LLP.   This is an excerpt from the session.  To receive more information about how to receive a copy of the recording, the materials and transcript of the presentation, contact Customer Service at (888) BUS-VALU, (503) 291-7963 or write to me directly at tom.hagy@bvresources.com.  I also own a phone and know how to use it:  610-312-4754.

 

Dr. Kennedy Says Statistical Models Are Great Liability & Damages Tools – In the Right Hands

Written by Tom Hagy

 

Experienced statistical modeling expert Dr. G. William Kennedy says the analytical tools he uses can be “extremely powerful” in establishing or disproving liability or determining damages if applied by a properly trained expert.   However, he says, in the hands of a novice they can be dangerous. 

 

Dr. Kennedy, Ph.D. and CPA/ABV with Anders Minkler & Diehl, LLP in St. Louis spoke on a live recording for Mealey’s Litigation Conferences and Business Valuation Resources LLC entitled “Compelling Statistical Evidence: Mining, Modeling, and Presenting Quantitative Financial Evidence to Juries.” 

 

If paired with a well-trained professional, Kennedy is extremely confident in the accuracy of the tools he discussed on the recording.  “Most of the methods I am relying on are in the textbooks and offer a great deal of certainty,” Kennedy said.  “The benefit of these statistical tools, the benefit to the bar, is that they all meet the criteria established under Daubert.”   He said he has told juries he is 99.999% confident in his estimations.

 

Once you have gathered the data Kennedy recommends that you draw a simple plot, which, he says, “helps avoid traps of using data that may have errors or need serious diagnostic work, or that the conclusions you are drawing are really not valid.” 

 

Then begin performing analysis.  “This process is iterative,” he says.  “There is not a eureka moment.”  The answers should be very clear at this point and you should only modify the analysis if necessary.  “I don’t mean to get to the answer counsel wants.  I mean if there are errors in the modeling, in the data, etcetera.  Anything that would make us go back and make corrections to make sure the results are credible.”  It is a quality control process, he said, that once complete offers data that are ready for forecasting.

 

These kinds of tools and techniques might be used in a productive way in litigation settings, both for damages and liability estimations. 

 

There are several statistical tools one can use in establishing liability or in damages quantification:  statistical sampling, correlation analysis, analysis of variance, time-series analysis, regression analysis, event studies and Monte Carlo simulation.  

 

Statistical sampling.  Take a sample but make sure it is representative of the population.  You can use it where the total population isn’t available or if it is impractical to obtain.  In a recent case where insurance allocation was based on addresses, he said he had to use sampling because it was impossible to get information on all of the locations.

 

Correlation analysis.  This looks at the relationship of two variables, for example height and weight, analysis of which would show that the taller a person is the more he will weigh.   This analysis will show the strength and direction of that correlation.  Correlation analysis is an effective tool to use in lost profits cases, he said, because it can test whether a factor is significant to a company’s sales and profitability.

 

Analysis of variance.  This not a method Kennedy uses a lot because it is imbedded in other tools like regression analysis.  He used it in a damages-for-lost-profits case where the question was whether national publicity about a product was the cause of the damages.  “We tested whether after the event the sales month to month changed and if there was an adverse reaction by consumers to the product problem.   If so there would be an increased volatility.  We tested this in various ways.  We compared that to the immediate period after the event and news publicity and came to an objective and quantifiable conclusion as to whether sales volatility increased.  In this case we actually were working on a liability determination.”  

 

Time series analysis.  This is a method that includes a regression-based calculation displaying a single series of dates.  A popular example is in graphical updates on the stock market.  What this analysis shows is whether the variable movements have a pattern.  This is one tool that can help forecast future sales, observe seasonality, or allow adjustments for seasonal or cyclical trends.   

 

Regression analysis.  This is broader than time series analysis, although it can be based on time.  Regression analysis permits two or more variables.  You can have variables X or Y or a number of variables.  In statistical analysis X tends to be time, but in regression X can be anything, like height or weight.  A business model might be that company’s sales move in sync with overall retail sales or with macroeconomic or internal factors.  Regression analysis can be used with cross sections of data.  It is important when diagnosing errors in data to know whether something is cross sectional or time sensitive.  Examples are lost profit calculations, which are influenced by a multitude of internal and external factors, “a situation well suited for regression analysis,” Kennedy explained. 

 

Event studies.  Event studies have been used in securities fraud cases, subprime lending cases, and in accounting liability action.  Conducting event studies is similar to constructing market models where you regress a company’s returns against market indices and come up with relationship about how stock prices might be performing across time against particular indices.  An event is something defined by the pleadings in the case, a particular adverse behavior by a board of directors or management, or an external event on a particular date.   In a classic event study you go back to pre-event time windows, look at daily returns against market indices or, in a more refined model, at daily stock returns or industry indices.  Then you test around the event window:  was there a change in the relationship between your company’s stock price and the overall market?  If so, how much?  You then need to quantify the change around the event date.  You can use this model to answer questions like:  what would the stock price have been had the event not occurred? 

 

Monte Carlo simulation.  How do you get around issues of quantifying uncertainty of an unestablished business where there is no commercialization yet?  The Monte Carlo simulation is a good tool for this, he said.  It is good for construing something as simple as “price x quantity” in software commercially available.  You can specify a range of prices and ranges of quantities and specify how you think the prices will likely stack – at higher or lower ranges.  Allowing you to show the variables will behave the Monte Carlo simulation will let you run and re-run the model, change the numbers and track the answer.  It allows you to do this thousands of times, he said, taking forecasting uncertainty out of the analysis.  “I use it almost every time I conduct an IP valuation.”

 

When presenting analysis based on Monte Carlo, Kennedy quotes authorities that have commented on the simulation.  The Litigation Services Handbook has many flattering things to say about Monte Carlo tool, calling it “The most flexible method of calculating an expectation when there are multiple potential outcomes or when the outcomes depend to varying degrees on the different inputs.”  The ABA treatise Fundamentals of Intellectual Property Valuation, a primer for identifying and determining value, also has good things to say about Monte Carlo simulations, he said.

 

Kennedy summarized by saying that statistical models are a very powerful way to present damages or proof or lack of liability; that because of the complex nature of the tools the expert should be thoroughly trained and knowledgeable of the tools; that attorneys should know the limitations of the tools; and, even though the tools are complex, experts and attorneys must present them to the trier of fact in a way that is easy to understand. 

 

Dr. Kennedy presented along with Jeffrey Dorman, Esq. with Freeborn & Peters, LLP of Chicago.  To receive more information about how to receive a copy of the recording, the materials and transcript of the presentation, contact Customer Service at (888) BUS-VALU, (503) 291-7963 or write to me directly at tom.hagy@bvresources.com.  I also own a phone and know how to use it:  610-312-4754.

 

Statistical models can make all the difference

Just how important are statistical models?   Just take a look at some recent cases where effective statistical analysis, including market event studies, became the “make or break” evidence, especially in cases concerning securities litigation and lost profit/economic damages:

• Expert’s market event study helps support loss causation claims and class action certification in large securities litigation; In re Flag Telecom Holdings, Ltd. Securities Litigation, (U.S. District Court, S.D.N.Y., 2007)
• Sophisticated event study shows efficient market and convinces court to certify class of litigants in a “fraud on the market” theory; In re Xcelera.com Securities Litigation (1st Cir., 2005)
• Delaware Chancery Court rejects lost profits computation based on an event study analysis that failed to consider all possible events that may have contributed to the loss; In Penn Mart Supermarkets, Inc. v. New Castle Shopping, LLC, (2005)
• Delaware Chancery Court favors “shared synergies” and actuarial analysis in the fair value appraisal of an insurance conglomerate’s merger; Highfields Capital, Ltd. v. AXA Financial, Inc. (2007)
• Federal court defines market efficiency, the five factors that prove it—and the statistical financial analysis that does (and does not) meet the burden of proof; In re Polymedica Corporation Securities Litigation (U.S. District Court, Mass., 2006)
• In context of shareholder seeking access to company books, Delaware Chancery Court approves unique statistical methodology for testing whether stock option backdating has occurred; Louisiana Municipal Police Employees’ Retirement System v. Countrywide Financial Corp. (2007)
• Court reverses $1.45 billion jury verdict based on lack of “fraud-free” valuation of securities; Morgan Stanley & Co. Inc. v. Coleman Holdings Inc.(Fla. App., 2007)

The last case was—until its reversal for lack of reliable statistical evidence, the largest jury award for economic damages ever.  All of these case—especially in light of the securities and loss litigation that is certain to come out of the current economic spiral—demonstrate just how important it is for valuation analysts to present compelling and understandable statistical models to a judge or jury (and assist their attorneys in preparing the evidence for deposition and trial). Reminder: Copies of all cases are available at BVLaw™, and the case abstracts are available by subscription at BVLibrary™.)

LEARN MORE.   To refresh and refine your use of statistical models, regression analysis, and market event studies, be sure to tune into BVR’s next teleconference, “Compelling Statistical Evidence: Mining, Modeling, and Presenting Quantitative Financial Evidence to Juries,” featuring William Kennedy, Ph.D, CPA/ABV of St. Louis’ Anders Minkler & Diehl, LLP, and Jeffrey Dorman, Esq. with the Chicago office of law firm Freeborn & Peters, LLP.  The teleconference takes place Wednesday October 22, 2008; to register, click here. 

Compiled and written by the fine staff at Business Valuation Resources in the latest issue of their free eZine, the BV Wire.

Courts Concerned Over Veracity of Electronic Records

 This is an excerpt from “Is It Real or Is It Memorex?  The Looming Crisis of the Authenticity of Electronically Stored Information in Commercial Litigation,” written by Thomas E. Hilton with the Forensic & Valuation Services Group at Anders Minkler & Diehl LLP in St. Louis.  Mr. Hilton prepared the article and will be presenting at the Law Education Institute’s National CLE Conference to be held in Vail, Colorado, January 3-7 with tracks covering several practice areas and business valuation.  BVR Legal is a proud sponsor of the event.

  

An increasing number of federal judges are concerned that electronic records produced for use in litigation may have been altered or manipulated in some way prior to presentation to the court. Additionally, these judges are concerned that the programs or procedures used to create and maintain ESI [electronically stored information] cannot be relied upon to protect the information from alteration by corporate insiders. To lay a proper foundation for the admission of ESI, the court must be able to understand the genealogy of the data: how was the data created, to what extent does it contain computer-generated data as opposed to computer-stored data, who has responsibility for maintaining the integrity of the data, how has the data been maintained, what opportunities have existed for the data to be corrupted or manipulated, how does the specific evidence for admission compare to the underlying data it purports to represent, and whether the ESI satisfies the basic tenets of trustworthiness, accuracy, relevance, and probative value?

 

With this level of uncertainty surrounding the various aspects of ESI, is it any wonder the admissibility of ESI has become a core issue in litigation? Is it any wonder judges have become increasingly hesitant to admit evidence lacking a traceable pedigree? A relatively recent ruling demonstrates the level of doubt in the minds of judges with respect to the authenticity of proffered ESI in litigation and the resulting impact on its admissibility.

 

In May 2007, Chief U.S. Magistrate Judge Paul W. Grimm of the District of Maryland rendered an extensive and detailed memorandum opinion addressing the “many interrelated evidentiary issues associated with ESI” in the case of Lorraine v. Markel American Insurance Co., 241 F.R.D. 534, 538 (D. Md. 2007). In this case, Judge Grimm denied both party’s motions for summary judgment because the ESI evidence in support of the motions (particularly e-mails) were not properly authenticated. This opinion stands as the most comprehensive judicial analysis to date on the evidentiary issues affecting ESI and provides useful information to the forensic accounting expert and litigator alike.

 

Judge Grimm’s opinion sets forth a series of evidentiary hurdles in the form of the Federal Rules of Evidence that must be overcome for ESI to be admissible by the court. Specifically, the judge focused on FRE 901 and 902, which requires a proponent of evidence to establish its authenticity; that the evidence is what the proponent claims it to be, through either extrinsic evidence or self-authentication. FRE 901 provides 10 illustrations of methods of authentication that comply with the rule, and FRE 902 provides 12 methods of “self-authentication” in which extrinsic evidence of authenticity is not required. Judge Grimm’s opinion provides a stern warning to practitioners desiring to proffer ESI to the court, “Unless counsel knows what level of scrutiny will be required, it would be prudent to analyze electronic business records that are essential to his case or her case by the most demanding standard.”

 

In March 2008, The Sedona Conference Working Group on Electronic Document Retention & Production issued The Sedona Conference Commentary on ESI Evidence & Admissibility (Commentary). This document and a wealth of others can be found at thesedonaconference.org. This document is replete with references to Lorraine and quotes the opinion of Judge Grimm throughout its 21 pages of text. This  publication, while not authoritative, provides the thinking of some of the best legal minds in the nation on the issue of authentication of ESI for purposes of litigation.

 

The Commentary explains that various types of ESI require different approaches for authentication, and explains how the various illustrations under FRE 901 and 902 may assist in the authentication process for the various types of ESI, such as e-mail, website postings, text messages, chat room content, and computer-stored records. Examples of a universally applicable authentication method would include a witness with personal knowledge (FRE 901). Absent such evidence, authentication would result from distinctly different methods that run to the distinctive characteristics of the given data. The appendix provides helpful guidance in the form of checklists, decision trees and practice tips to assist practitioners in determining whether the proper authentication that might be deemed conclusive by the court can be achieved.

 

Again, this is only an excerpt.