Opinion
CV-04-4775 (NGG) (VVP), CV-04-5620 (NGG) (VVP).
October 5, 2007
DECISION AND ORDER
The motion now before the court, described as a motion in the nature of a motion in limine, seeks to establish parameters for the completion of discovery concerning the pattern and practice allegations in these two actions, which have been consolidated for discovery purposes. The motion is made by Leo White, the third-party plaintiff in CV-04-4775, and by Linda and Kimberly Council, the plaintiffs in CV-04-5620 (collectively hereinafter the "plaintiffs"). They seek rulings (1) to determine the universe of transactions from which evidence may be adduced to prove, or disprove, the pattern and practice allegations, and (2) to determine the minimum number of non-party purchasers in those transactions that the plaintiffs will be required to call to satisfy their burden of proof with respect to the pattern and practice allegations.
Broadly speaking, the plaintiffs allege that they are the victims of a widespread, fraudulent scheme engineered by the defendants. The scheme allegedly targeted first-time minority home buyers who were sold homes at grossly inflated prices, financed with inflated mortgages, and induced by misrepresentations concerning the value of, and repairs to be made to, the homes. The principal architects of the scheme were the defendants Better Homes Depot, Inc., Madison Home Equities, Inc. and their respective principals, Eric Fessler and Nadine Malone (hereinafter "the defendants"). The plaintiffs also allege that an essential part of the scheme was the endorsement of the inflated mortgages by the Department of Housing and Urban Development ("HUD"), which permitted Madison to obtain mortgage insurance from the Federal Housing Authority ("FHA") that indemnified the lender in the case of a default by the purchaser. With such insurance in place, Madison was able to sell the mortgage loans it made to other lenders and thus divorce itself from any risk.
The motion has been made largely at the instigation of the court in an effort to establish a reasonable plan for discovery concerning the plaintiffs' pattern and practice claims. To prove their pattern and practice allegations, the plaintiffs contemplate using testimony by other home buyers who had purchased homes from Better Homes with financing by Madison. To meet that proof, the defendants also contemplate using testimony from home buyers, presumably buyers who were satisfied with their experiences with the defendants. As the defendants have participated in hundreds, if not thousands, of home sales together, providing temporal and perhaps other limits on the universe of sales about which non-party witnesses could be called to testify would assist the parties in focusing their efforts in discovery and marshaling their proof. Similarly, setting limits on the number of non-party witnesses who could be called at trial, and identifying such witnesses in advance, would permit the parties to decide which non-party witnesses they wished to depose. The court thus recommended that the plaintiffs make the instant motion to determine (1) the universe of sales about which proof would be offered and (2) the number of non-party homebuyers the parties would be permitted to call as witnesses. The court addresses each of these questions in turn.
The Universe of Home Sales
In their moving papers, the plaintiffs have identified three criteria that a home sale should meet to fall within the universe of home sales subject to their pattern and practice allegations: (1) Better Homes was the seller named in the contract of sale; (2) Madison provided the mortgage to purchaser; and (3) the mortgage was insured by FHA. They contend that any home sale meeting those three criteria that occurred within the period from January 1, 1997 through May 31, 2000 falls within the pattern and practice allegations in their complaints and therefore may be considered at trial. In their opposition papers, the defendants objected to the plaintiffs' limiting criteria and contended instead that a much broader range of sales should be considered. The universe they proposed would have included all of Better Homes's transactions and all of Madison's transactions from January 1, 1997 through the present.
At oral argument, however, the defendants withdrew their objections to the plaintiffs' proposal in part. Specifically, they agreed that the universe of home sales should be limited to those where the three criteria above were met. They parted ways with the plaintiffs, however, with respect to the temporal limitation on transactions to be included. Although in their papers the defendants had endorsed a time period dating back to January 1, 1997, at oral argument they sought a shorter period from January 1, 1999 through May 31, 2000, a period that corresponds to the time period for which the court had required the defendants to produce documents during the documentary phase of discovery. The plaintiffs continue to press for a universe of sales dating back to January 1, 1997.
The plaintiffs have the better of the argument. To begin with, the time period selected by the court for documentary discovery was not intended to be a limitation on the plaintiffs' proof at trial. The limitation was a response to the defendants' argument that discovery over the full range of time sought by the plaintiffs was too burdensome. In an effort to meet that concern, the court narrowed the time period of sales for which the defendants were required to search for and produce their own documents. The limitation was thus meant only to relieve the burden of production by the defendants; it was never explained as, or meant to be, a limitation on the proof the plaintiffs would be allowed to present at trial.
The plaintiffs' claims allege a pattern and practice of fraudulent conduct targeting minority home buyers. The complaints themselves contain specific allegations about the defendants' conduct dating back to August 1998 when the plaintiffs allege they first made contact with the defendants concerning their interest in purchasing homes. Thus, notwithstanding that their own purchases occurred in 1999, they are entitled to seek to prove that the defendants' pattern of conduct dates back at least to the point where they themselves first began dealing with the defendants. Beyond that, there is nothing in their complaints that limits them to only the time period when they were dealing with the defendants. They allege an ongoing pattern of behavior. If they wish to prove that the pattern preceded their own contact with the defendants by at least a year and a half, that is, as far back as January 1, 1997, the court sees no reason why they should not be permitted to do so, particularly since the defendants themselves advocated a time period going back to that date. Indeed, it would appear that the longer the time period, the greater the burden on the plaintiffs to show that instances of the misconduct they allege are part of a pattern rather than isolated, unconnected occurrences. The court thus concludes that the appropriate universe of home sales for examination concerning the plaintiffs' pattern and practice allegations are the sales meeting the three criteria set forth above that occurred within the time period from January 1, 1997 through May 31, 2000.
The Number Necessary to Establish a Pattern and Practice
The plaintiffs also ask the court to set the minimum number of non-party home buyers whom they will need to call to satisfy their burden of proof on their "pattern and practice" claims. Such a ruling, however, would constitute a ruling on the sufficiency of evidence, a matter best left to the trial court, and the court accordingly declines to make such a pronouncement.
Some direction, however, must be provided to the parties at this stage of the litigation to permit them to conduct discovery adequate for trial preparation, including the opportunity to depose the witnesses who will appear. The Federal Rules of Civil Procedure afford the court wide latitude in controlling discovery, see, e.g., Will v. Amerada Hess Corp., 379 F.2d 32, 41 (2nd Cir. 2004), guided by the admonition that the rules "be construed and administered to secure the just, speedy, and inexpensive determination of every action." Fed.R.Civ.P. 1. The number of potential non-party witnesses, i.e., home buyers whose transactions fall within the universe above, is well over a hundred, far too many for the parties to depose and far more than the trial court would permit to testify. To focus the parties' deposition efforts, then, the court will require the plaintiffs, collectively, and the defendants, collectively, to identify the non-party homebuyers they may wish to call at trial, and will set a limit on the number that may be identified. In setting that limit, the court will take into account the number of witnesses other courts have found sufficient to establish a "pattern and practice" of behavior, as that will undoubtedly influence the trial courts' decisions concerning how many such witnesses will be permitted to testify.
The plaintiffs have cited cases in a variety of contexts in which anecdotal evidence offered by victims of misconduct was accepted by courts as proof that the perpetrators were engaged in a pattern of misconduct. Although none of the cases specifically addressed the question of how many witnesses are necessary to establish a pattern and practice claim, they are nevertheless instructive. In United States v. Damti, 109 F. App'x 454, 456 (2nd Cir. 2004), a criminal case involving allegations of consumer fraud, the government was permitted to argue that evidence of ten fraudulent transactions out of more than four-thousand transactions conducted by the defendants were representative of the fraudulent manner in which the defendants conducted business. In FTC v. Kitco of Nevada, Inc., 612 F. Supp. 1282 (D. Minn. 1985), testimony by eight victims, when coupled with other evidence, was sufficient to establish that the defendants had engaged in a "widespread pattern and practice of deception." 612 F. Supp. at 1294. In United States v. Oloyede, 982 F.2d 133, 136 (4th Cir. 1992), the court concluded that testimony by eight witnesses demonstrated a "distinct pattern" of fraudulent conduct by an attorney in connection with immigration applications. Finally, in Garcia v. Victoria Independent School District, No. 75-V-11, 1978 WL 97, at *12 (S.D. Tex. July 24, 1978), the court concluded that the testimony of ten witnesses, coupled with statistical and other evidence, established a pattern and practice of employment discrimination.
The defendants have not cited cases concerning the number of witnesses deemed appropriate to establish a pattern of behavior. Rather, their attack on this aspect of the plaintiffs' motion focuses on the plaintiffs' failure to identify what would constitute a statistically significant sample of transactions about which testimony would be offered. They cite various cases in which courts concluded that the samples about which evidence was presented were too small. See Haskel v. Kaman, 743 F.2d 113, 119 (2d Cir. 1984) and cases cited therein at 121. Their argument would require more detailed analysis if the court was making a ruling like the one requested by the plaintiff, i.e., a determination as to the minimum number of witnesses necessary to establish the plaintiffs' pattern and practice claim. Because the court is declining to do so, however, and is merely setting a limit on the maximum number of non-party witnesses who may be identified as potential witnesses for trial, the court need not require proof concerning statistically significant sample size. Of course, if either the plaintiffs or the defendants believe that the limit set by the court places them at risk with respect to this issue, they are free to seek reconsideration by this court upon a proper showing.
The parties offer little guidance on the issue of sample size, and it is not clear that statistical significance is a particularly relevant consideration for the type of proof the plaintiffs wish to offer given the issues to be decided. The question of sample size is clearly of importance in employment discrimination cases where a plaintiff wishes to offer statistical evidence to prove discriminatory intent. E.g., Haskell, 743 F.2d at 119-20. But where a plaintiff is seeking to establish a pattern of fraud, rather than raise an inference of discrimination, the sufficiency of a plaintiff's proof would appear to depend less on the statistical significance of the number of incidents examined, and more on the similarity of the fraudulent means employed in the incidents examined. See, e.g., Damti, 109 F. App'x at 456; Oloyede, 982 F.2d at 135-36; Kitco of Nevada, Inc., 612 F. Supp. at 1294.
In view of the foregoing, on or before November 15, 2007 the plaintiffs and the defendants shall each identify the non-party homebuyers, not to exceed 20 in number on each side, whom they may call at trial. No non-party homebuyer will be permitted to testify at trial unless his or her name is identified in this manner. This limitation is not to be construed as a ruling that the district court will accept testimony by all or any witnesses so identified, nor is it a ruling that testimony by this or any other number of witnesses is sufficient to establish the plaintiffs' pattern and practice claims.
A conference will be held on October 18, 2007 at 10:00 a.m. A schedule for completing discovery and for making dispositive motions, if any, will be set at the conference.
SO ORDERED: