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.

 

Michele Miles: When Picking a Financial Expert, Beware of Hormonal Imbalances

By Tom Hagy

 

 

Battle metaphors are frequent when talking about litigation and trials.  Verbs like kill, shoot, crush, and destroy are not uncommon.  Michele G. Miles, both a J.D. and accredited business appraiser, sees a danger in taking the macho metaphors too far, especially when they affect the conduct of a financial expert. 

 

Miles said you need to avoid what she calls “testosterone poisoning,” and urges you to learn the signs.

 

First, she said during the recording of a Mealey’s Litigation Conference / BVR audio, that you do not want a witness who is going to pick a fight.  Instead, she said, you want one who is proud of their process – “so proud that he will be glad to go through it again and again.”  This will give counsel an indicator that the expert will exude credibility.

 

But what are the signs of a testosterone-poisoned witness?  Miles said such a character may refer to a place in their office as a “war room” where they store and summarize documents.  They may be inclined to place a photo of themselves in their expert report.  They tend to come to you with checklists, suggesting rigidity in their process.  They may have trademarked some of their terms or processes.

 

If you want someone to pass Daubert review, she said, you do not want to show “personally developed methods” were used in place of “real world” methods.  

 

Testosterone-engorged experts may also have a tendency to read or interpret case law for the attorney.  If case law makes its way into the expert’s report “something is badly broken,” Miles warns.  If an expert must discuss a case in his report, he should refer to a treatise written by an authority on the subject, something everyone agrees is authoritative. 

 

A testosterone imbalance also can lead to a decision to have one expert to attack another, something Miles cautions against.  “Don’t use your expert as a paid assassin,” she said.  “Your witness can’t testify on his own opinion while throwing rocks as someone else’s.”   She said to stick to the process and avoid getting into past conflicts. 

 

Not only should an expert not be an assassin, they should not be a “crutch” for an attorney making his way through new subject matter.   “Do your learning behind the scenes,” she advises.  “Ask dumb questions behind the scenes.”

 

Miles warned against appearing to be too close to your witnesses.  Never take your expert out for “show and tell,” she said.  Do not pass notes during hearings, do not take him to depositions, and do not hang out during trial.  “Don’t treat them as part of the team,” she said.   

 

“What you want is an expert that, when he leaves, prompts everyone to say ‘and we never even got a chance to thank him,’” she said, recalling romanticized heroes from the Old West. 

 

You can hear more from Michele Miles, who also is former executive director of the Institute for Business Appraisal, plus commercial litigator James Dugan of Willkie Farr & Gallagher and financial expert Steven Schroeder of Schwartz & Associates LLC,  on the CD package entitled “Effective Timing & Use of Financial Experts.”   For more information, write to me directly at tom.hagy@bvresources.com.

 

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.

Fierce Daubert Challenge Fails, Court Says Wait for Trial

A recent case out of federal court in Louisiana isn’t groundbreaking or unusual, but is illustrative of how attorneys swat at financial expert witnesses like low-hanging piñatas. 

 

In a case where the value of an engineering firm came into play, the plaintiff aimed its Daubert challenge at defense expert Kim Early CPA, MBA, calling him unqualified and his testimony irrelevant, based on the wrong methodology and not supported by facts.  The court determined, however, that even though Early had not taught in the BV field, is not published and holds no BV certifications, he is not required to have these to testify.   The plaintiff said Early didn’t know enough about the claims in the litigation, but the court said this and other issues relating to the value of stock are relevant to damages and available for scrutiny during cross-examination.  The plaintiff took issue with Early’s methodology, his lack of a publishing history and the fact that the engineering firms he used for comparisons weren’t the same size.  Based on an email exchange between counsel and the expert, plaintiff also questioned whether the defendant’s in-house counsel overly influenced the witness’ conclusions. 

 

Feelings

 

The defense even took issue with the way Early started his sentences, several of which began with phrases such as “we think” or “we feel” or “we estimated.”  

 

The court rejected the challenge and gave Early the green light to testify, saying it is required to look at a witness’ principals and methodology, not just his conclusions.  The court repeated its view that the plaintiff was free to address its concerns during cross-examination.

 

Willis v. TRC Cos., 2008 U.S. Dist. LEXIS 38573 (W.D. La. May 12, 2008).  Note:  In case you’re interested, I see that Lexis posted 10 motions and other documents leading up to this decision. Good stuff.     

‘Lost Sales’ Damages Valuation Based on Independent Data Verification Prevails

Lyman v. St. Jude Medical S.C., Inc.; 2008 WL 2224352 (U.S. Dist.) or 2008 U.S. Dist. LEXIS 42015

A case from federal court in Wisconsin, involving starkly different approaches to valuing damages in a contract dispute, stresses the importance of independent verification and objective data.

Plaintiff Lyman entered into a 10-year deal with St. Jude Medical to sell pacemakers and defibrillators (CRM products). The agreement contained a four-year guarantee period for which Lyman was to receive $3 million. After that, compensation was commission-only. St. Jude terminated the agreement after two years. Lyman sued. Daniel Gotter with Winter, Kloman, Moter & Repp, the plaintiff’s expert, used raw data to prepare five different models of damages projections based on sales quotas; projections calculated by the CFO for St. Jude’s parent company; regression analysis of the plaintiff’s company’s past sales at St. Jude; projected future sales; and projected net sales of the plaintiff’s company. The court allowed testimony on all five calculations, saying they would assist the jury in assessing damages.

Winter, Kloman, Moter & Repp, the plaintiff’s expert, used raw data to prepare five different models of damages projections based on sales quotas; projections calculated by the CFO for St. Jude’s parent company; regression analysis of the plaintiff’s company’s past sales at St. Jude; projected future sales; and projected net sales of the plaintiff’s company. The court allowed testimony on all five calculations, saying they would assist the jury in assessing damages.

The defendant’s expert, Randall D. Wilson with RGL Forensic Accountants & Consultants prepared two sets of projections: one weighted and one unweighted. Both sets were derived from the plaintiff’s “alleged sales” during the approximately two years it sold CRM products to St. Jude, the numbers were obtained from one document provided to the expert by St. Jude’s counsel. Taking into account the plaintiff’s decreased sales performance between 2003 and 2004, St. Jude’s expert’s original regression resulted in a “negative sloping trendline.” Because the plaintiff’s sales varied widely from month-to-month, the expert weighted the data to obtain a positive trendline for years five through 10 of the agreement.

The court focused not only on the source of the defendant’s expert’s data, but what he failed to do with it. Contrasted with Gotter’s independent valuation of raw data, the court said “the bottom line is that [the defense expert] never talked to anyone at St. Jude to verify the accuracy of the information in any of the documents he reviewed.” Finding the basis for his projections unreliable, the court excluded Wilson’s projection testimony. The case settled on the eve of trial.

Plaintiff attorneys: Todd R. Seelman and Leslie E. Miller of Grimshaw & Harring PC, Denver, CO; Michael J. Cohen and Thomas M. Hruz, Meissner Tierney Fisher & Nichols SC, Milwaukee, WI. Defense counsel: Brian G. Cahill, David J. Turek and Paul F. Heaton, Gass Weber Mullins LLC, Milwaukee, WI.