Summary
denying a motion to certify class, indicating that liability would depend on "a file-by-file review of all class members' transactions," that such "individualized inquiry is incompatible with a class action," and that recent cases "trend sharply against class certification in actions like this one" (citing Randleman v. Fidelity Nat'l Title Ins. Co., 646 F.3d 347 (6th Cir. 2011); Benavides v. Chicago Title Ins. Co., 636 F.3d 699 (5th Cir. 2011); Corwin v. Lawyers Title Ins. Co., 276 F.R.D. 484 (E.D. Mich. 2011); Scott v. First Am. Title Ins. Co., 276 F.R.D. 471 (E.D. Ky. 2011))
Summary of this case from Smith v. First Am. Title Ins. Co.Opinion
CASE NO. C10-199RAJ
07-24-2012
THOMAS BOUCHER, et al., Plaintiffs, v. FIRST AMERICAN TITLE INSURANCE COMPANY, Defendant.
HONORABLE RICHARD A. JONES
ORDER
I. INTRODUCTION
In the wake of the court's May 2, 2011 order (Dkt. # 125) declining to certify a class, the court now considers a renewed motion to certify a class from Plaintiffs Thomas and Carol Boucher. The parties requested oral argument, but the court finds oral argument unnecessary in light of the parties' extensive submissions and the voluminous evidentiary record. For the reasons stated herein, the court DENIES the Bouchers' motion to certify a class. This order concludes with a deadline for a joint statement on a schedule for trying the Bouchers' individual claims.
II. BACKGROUND & ANALYSIS
The Bouchers, like hundreds of thousands of Washingtonians, bought title insurance from Defendant First American Title Company when they refinanced their home. More accurately, they bought title insurance for their lender, who probably required that they buy a policy to protect the lender's collateral. The Bouchers claim that First American overcharged them. More importantly, for purposes of this motion, they hope to bring a class action to vindicate the claims of a (possibly) large number of other Washingtonians who they allege First American overcharged in home refinance transactions since November 2003.
When the court addressed the Bouchers' class claims in May 2011, it found that the Bouchers had failed to show that issues capable of classwide resolution predominated over individualized issues and had failed to show that a class action was superior to other ways of resolving the class claims. The court observed that in years of litigation, the Bouchers had failed to conceive of a manageable way to analyze the hundreds of thousands of refinance transactions to determine who might be a class member, much less a manageable way to prove each class member's claim. Instead, four years of litigation had yielded evidence of perhaps five customers who First American had overcharged and no viable plan for systematically identifying other customers. Still, because nothing in the record ruled out the possibility that First American had overcharged Washington customers, the court gave the Bouchers another chance to show that they could identify those customers and resolve their claims on a classwide basis. The renewed motion for class certification now before the court, which the Bouchers filed more than seven months after the court's prior order, is their attempt to capitalize on that chance. As the court will explain later in detail, it concludes that although the Bouchers have made progress, they have not convinced the court that they can efficiently and reliably identify class members or prove their claims. To understand that conclusion, the court first reviews the underpinnings of the Bouchers' claims, then examines their recent efforts to show that they can prove claims on behalf of similarly-situated First American customers in Washington.
A. First American's Washington Rates for Title Insurance
Washington, like many states, regulates the prices title insurers can charge for their policies. First American could charge only those prices that it disclosed in rate manuals that it filed with Washington's Insurance Commissioner. A rate manual typically sets prices for one or more counties. From the early 2000s through late 2007, First American had 17 rate manuals covering Washington's 39 counties. Stuczynski Decl. (Dkt. # 173) ¶ 4. It has had just five rate manuals since then. Id. ¶ 7.
Every First American rate manual describes a "reorganization rate" that applies to many refinance transactions. The reorganization rate invariably applies when a "lender's policy insuring the lien of a deed of trust has been issued at the scheduled rate, and a new deed of trust on the same property is" executed. E.g., Stuczynski Decl. (Dkt. # 173), Ex. A at 9. Some rate manuals impose the additional requirement that the "same borrower" execute the new deed of trust on the property. Compare id. at 9 ("same borrower" requirement) with id. at 47 (no "same borrower" requirement). In all cases, the reorganization rate is 50% of the "general schedule rate," and the court will sometimes refer to it as the "reorganization discount." The general schedule rate escalates based on the limit of the policy, which in the case of a lender's policy is the amount of the home loan.
This case has always been about First American's alleged failure to extend the reorganization discount to qualified customers. First American's rate manuals contain a variety of exceptions, surcharges, and convoluted provisos that will, in some circumstances, lead to a customer paying something other than the general schedule rate or the reorganization rate. Until very recently, the Bouchers never mentioned these possibilities. Each rate manual plainly discloses these other possibilities, however, including the "blanket exception mortgagee's policy," a set of rates that applies to policies that contain "four special blanket exceptions." E.g., Stuczynski Decl. (Dkt. # 173), Ex. A at 55-56. First American refers to policies sold at these rates as "Super Eagle" policies. Id. ¶ 34. The rate manuals also describe a "Simultaneous Issue" rate that applies when First American issues two policies in the same transaction. E.g., id., Ex. A at 14, 33, 52. The Simultaneous Issue rate would apply, for example, where the customer refinanced her home with two loans rather than one. Id. ¶ 29. Although the Super Eagle and Simultaneous Issue rates receive the most attention, the court observes that the rate manuals reflect a variety of surcharges that apply in a wide range of circumstances. The record reflects that First American frequently imposes those surcharges.
B. The Bouchers' Individual Claims and Their First Effort to Show that They Can Certify A Rule 23(b)(3) Class of Similarly-Situated Washingtonians
The Bouchers believe that they should have paid $367.50 for the policy they bought when they refinanced in 2005. That price reflects the application of the reorganization rate to the general schedule rate of $735. If this were a simple matter of First American declining to extend the Bouchers the reorganization discount, one would expect that the Bouchers paid $735, the general schedule rate. Instead, they paid $821. The court explored the peculiarities of the Bouchers' individual claim in its May 2011 order, and declines to repeat the discussion. May 2011 ord. at 2-3, 13-14. The court focuses on whether First American improperly denied the reorganization discount to other Washingtonians.
Through May 2011, the Bouchers had done little more than speculate about the number of Washingtonians to whom First American failed to extend the reorganization discount. Rather than articulate how they might assess how many of First American's hundreds of thousands of Washington customers had overpaid or potentially overpaid, they invented an "extremely conservative overcharge rate of 20%." Dkt. # 81 at 8. Their evidence, by contrast, revealed only five other customers who First American might have overcharged.
The court could not certify a class on the evidence the Bouchers put before it in May 2011. The court will not repeat its summary of the standards applicable to a class certification motion or its analysis of the Bouchers' efforts to meet the four prerequisites of Federal Rule of Civil Procedure 23(a) (numerosity of the class, commonality of legal and factual issues, typicality of class representative's claims, and adequacy of class representation). The court found that the Bouchers had presented common questions, and Ms. Boucher was an adequate class representative. The court could not conclude that the Bouchers had proven even that their proposed class was sufficiently numerous or that they had claims that were typical of other class members.
Although the court had concerns about the Bouchers' inability to meet the four prerequisites of Rule 23(a), it declined to certify a class primarily because of Rule 23(b)(3). A plaintiff seeking to certify a class for money damages must show that "questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3). The court explained that the "primary obstacle to both a finding of superiority and a finding of predominance in this case is that the Bouchers have yet to demonstrate that they can efficiently discover the evidence necessary to prove their claims." May 2011 ord. at 16. The court permitted the Bouchers to conduct additional discovery and to renew their motion for class certification. The court cautioned them that their renewed motion "must (at a minimum) demonstrate to the court that they have developed a means of identifying who belongs to the class, and they must show that they have used that methodology to actually identify at least some class members." Id. at 16.
C. The Bouchers Have Obtained Evidence Since the Court's Last Order, But that Evidence Only Convinces the Court That It Should Not Certify a Class.
The Bouchers have renewed their motion for class certification. Whereas the court was disappointed in May 2011 in the Bouchers' lack of progress toward demonstrating they could manage this case as a class action, they have made substantial progress since then. The court will soon describe their efforts.
1. The Bouchers Must Show That They Can Identify Class Members and That Their Claims Do Not Overly Depend on Individualized Evidence.
The Bouchers must accomplish two different, but related, tasks. They must demonstrate that they can identify class members and they must demonstrate that they can prove their claims on a classwide basis. Initially, the Bouchers had conflated those tasks by proposing a "fail-safe" class definition that essentially defined the class as all of First American's Washington customers who deserved the reorganization discount but did not receive it. Fail-safe classes are impermissible because they make it impossible for a defendant to prevail against the class. See May 2011 ord. at 9 (rejecting fail-safe class definition); Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 825 (7th Cir. 2012) (explaining that a fail-safe class definition "is improper because a class member either wins or, by virtue of losing, is defined out of the class and is therefore not bound by the judgment").
The Bouchers revised their class definition so that it was not fail-safe, and they rely on essentially the same definition in their renewed class certification motion:
All persons who, while residing in the State of Washington, and in connection with a mortgage financing transaction where the loan amount did not exceed $1,000,000: (a) paid a premium on or after November 28, 2003 for the purchase of residential title insurance from First American . . . ; (b) had a lender's title insurance policy that insured the lien of a deed of trust on the same property that had been issued at the scheduled rate; and (c) did not receive the reorganization rate specified in the applicable First American rate manual.The Bouchers propose no subclasses.
If the Bouchers hope to manage this case as a class action, they must be able to identify class members. If they cannot, they are unlikely to be able to satisfy their obligation to give notice to the class. Rule 23(c)(2)(B).
There is little point, however, in allowing the Bouchers to give class members notice if they cannot prove their claims without a cumbersome individualized review of evidence pertaining to each class member's title insurance purchase. Or, put in Rule 23(b)(3) terms, if the success or failure of every class member's claim depends on individualized proof, the Bouchers cannot demonstrate that "common questions of law or fact predominate over any questions affecting only individual members."
2. The Bouchers Cannot Reliably or Efficiently Identify Class Members.
Can the Bouchers identify class members?
To answer that question, the court focuses on evidence pertaining to customers who purchased title insurance directly from First American. Most of First American's customers fall into this category, but a significant minority of them purchased their policies from independent agents. Among the many complications in this case is that the evidence pertaining to First American's direct-issue policy sales and its agent-issue policy sales resides on different computer systems. The FAST system covers direct-issue sales, and has done so since at least 1992. According to the Bouchers, FAST data reveals about 250,000 direct-issue sales of First American policies in refinance transactions during the class period. The STARS system, which covers agent-issue sales, has existed only since 2008. Using STARS data, Plaintiffs have identified about 30,000 agent-issue refinance customers. Murray Decl. (Dkt. # 157) ¶ 10. To get data on agent-issue sales before 2008, the Bouchers must subpoena records directly from each of 34 agencies who have sold First American policies during the class period. The Bouchers' subpoena campaign has so far yielded files from 13 agents on about 12,000 transactions, but no evidence of how many of the 12,000 are potentially part of this class action. Id. ¶ 21. Other than to roughly describe the relative number of agent-issue and direct-issue transactions, the court will not dwell on the complexity of obtaining information about First American's agent-issue sales. Uncovering data and documents on agent-issue sales is more complicated than the process for obtaining information on direct-issue sales, and, as the court will now discuss, the direct-issue process is complicated enough.
The court uses round numbers throughout this order.
The Bouchers began their quest for qualifying direct-issue customers with a spreadsheet culled from FAST data identifying about 250,000 Washington direct-issue sales since late 2003 in which First American issued only a lender's policy. Customers who buy a home typically purchase policies for themselves and for their lenders; customers who refinance typically buy policies only for their lenders. The Bouchers thus requested FAST data on lender's-policy-only transactions because they were most likely to reflect refinancing transactions. There are, however, lender's-policy-only transactions that are not refinances, including construction loans and transactions in which a home seller buys a title insurance policy. Stuczynski Decl. (Dkt. # 173) ¶ 27.
The Bouchers analyzed the 250,000-transaction spreadsheet by creating a series of their own spreadsheets using formulas to compare the prices the customers paid to prices they should have paid based on their counties of residence, dates of the sale, and the amount of their loans. Nordby Decl. (Dkt. # 158) ¶¶ 3-7. The spreadsheets the Bouchers developed attempted to duplicate the pricing formulas of the applicable First American rate manuals. The spreadsheets only compared the price the customer paid to half of the general schedule rate for a policy, thus identifying transactions with a discrepancy between the price paid and the reorganization rate. The spreadsheets did not account for whether the customer had bought Super Eagle or Simultaneous Issue policies, or for other pricing quirks described in the rate manuals. The spreadsheets identified about 9,500 transactions in which the price that the customer paid exceeded the formula price by more than $100. Id. ¶ 8-14. The Bouchers do not explain their choice of a $100 tolerance for overcharging, nor did they attempt to tailor the spreadsheets to identify transactions in which the customer paid approximately double the price she should have paid, which is what one would expect if First American had failed to give the reorganization discount.
First American asks the court to strike the declaration of the Bouchers' paralegal, who prepared the spreadsheets that break down First American's data. The court finds no reason to strike the declaration, which carefully describes the paralegal's methodology. Indeed, it was only because of the paralegal's care in describing her methodology that First American was able to so meticulously attack it.
The Bouchers then attempted to identify which of these 9,500 customers had a prior lender's policy. First American provided some data from the electronic document repository that is part of its FAST system. Despite spending more than $10,000, the Bouchers had no success in extracting useful information from these data. Murray Decl. (Dkt. # 157) ¶¶ 13-14. They turned to an undisclosed third party to purchase property finance histories for an undisclosed sample of these 9,500 customers. Id. ¶ 15. Those finance histories apparently identified, in some cases, the name of the insurance company who recorded the deed in the transaction preceding the refinance. The Bouchers assumed that the insurance company who recorded the deed also issued the lender's policy for the prior transaction, and discovered that First American itself had likely issued a prior lender's policy for an unknown number of these customers. Id. They asked First American to produce the policies for 275 of those customers. Id., Ex. 12. First American produced 68 policies. Id. ¶ 15. Plaintiff claim that 66 of these policies met condition (b) of their class definition. They also claim to have identified eight other people who meet the class definition, although it is not clear how they did so. Thus, at the end of this process, the Bouchers claim to have identified 74 direct-issue customers (out of about 250,000 transactions) who would be members of the class they hope to certify. The Bouchers concede that in cases where First American did not record the deed in the preceding transaction, they can determine the prior insurer only by obtaining deeds of trust for the underlying transaction.
The Bouchers' spreadsheet analysis proves two points. First, the Bouchers' prior suggestion of an "extremely conservative overcharge rate of 20%" was fanciful. The Bouchers' analysis identified about 9,500 customers out of 250,000 refinance transactions in which there was even potentially a denial of the reorganization discount. In even the most charitable view of the Bouchers' evidence (a view that the evidence does not support), First American denied the discount to less than 4% of its direct-issue customers. Taking a more realistic view of the evidence, First American overcharged only a miniscule fraction of its customers. (As the court will later discuss, the evidence suggests that First American undercharged a similar fraction of them.)
Second, the Bouchers have demonstrated no reasonable method to narrow their list of 9,500 potential class members to a list of actual class members. The Bouchers worked hard to sift through the 9,500-customer subset to come up with 74 people that they believe belong to the class. But the Bouchers were able to identify them only by the happenstance that First American issued both their (alleged) prior lenders' policy and their most recent one. Even in those circumstances, the Bouchers were not able to obtain the prior lender's policy for more than 200 of the 275 transactions. The record is silent as to why First American was unable to produce those policies. The Bouchers propose to obtain the prior lender's policies from insurers other than First American only by obtaining deeds of trust for each transaction to make an educated guess about who issued a prior lender's policy, then issuing subpoenas to as many as 14 active title insurance underwriters in Washington and, where necessary, their agents. Murray Decl. (Dkt. # 157) ¶ 18; Stuczynski Decl. (Dkt. # 173) ¶¶ 13-16 (noting that Washington has 14 current active underwriters and almost 80 active title agents). The Bouchers give the court no reason to think that their efforts to obtain prior lenders' policies from other title insurers would be more successful than their efforts to obtain them from First American.
In short, the Bouchers have not shown that they can identify potential class members. The court finds that they have not shown that they could satisfy their obligation to give notice to class members.
3. The Bouchers Cannot Prove Any Class Member's Claim Without an Individualized Review of Evidence.
For the sake of argument, the court assumes that the Bouchers could succeed in identifying direct-issue class members. That is, the court assumes that despite the problems it has just identified, the Bouchers could obtain discovery from First American and third-party discovery from dozens of other title insurers and their agents to identify class members with reasonable accuracy. Even under that unwarranted assumption, the evidence shows that the proof (or disproof) of class members' claims would depend on a cumbersome file-by-file review of each class member's transaction. The explanation for that conclusion lies in First American's response to the Bouchers' identification of 74 class members.
First American's underwriting counsel conducted a review of each of the 74 direct-issue customers the Bouchers identified. Stuczynski Decl. (Dkt. # 173) ¶ 24. The review, which incorporated FAST data plus documents from the electronic repository, took about 20 minutes per customer. Id. ¶¶ 25-26. According to counsel, 5 of the 74 customers were either purchasing homes or taking out construction loans, and thus not qualified for the reorganization discount. Id. ¶ 27. Of the remaining 69, 64 paid premiums below the general schedule rate, meaning that it is not possible that First American simply denied them the reorganization discount. Id. ¶ 28. 57 of those 69 customers took out two loans to refinance, and thus paid the Simultaneous Issue rate instead of the reorganization rate. Id. ¶¶ 29-30. In the view of underwriting counsel, 51 of the 57 paid no more than the proper price, and 17 of those 51 paid less. Id. ¶ 31. He concedes that 6 of the 57 paid too much, but contends that this was the result of miscalculation, not the improper denial of the reorganization discount. Id. ¶ 32. Many of the 69 customers were not eligible for the reorganization rate for other reasons, including a few who had uninsured prior loans, 5 who bought Super Eagle policies, a few who had paid off their prior loans, and a few in which the FAST data itself was inaccurate. Id. ¶¶ 34-36. Underwriting counsel identified 3 transactions of the 69 where the customer was potentially entitled to the reorganization rate but paid the general schedule rate. Id. ¶ 37. To determine whether the customer was actually entitled to the reorganization rate would require proof of a prior lender's policy. Id. The Bouchers obtained a prior lender's policy for only 1 of the 3 customers. Id. For 1 of the remaining 2, the Bouchers relied on a title commitment instead of a prior lender's policy, and in the other, the Bouchers relied on a prior lender's policy for a loan predating the one in existence at the time of the refinancing. Id.
A title commitment is a document summarizing the cost of title insurance for a particular transaction. Ordinarily, a customer ultimately purchases title insurance on the terms stated in the title commitment. In some cases, however, such as when a customer backs out of a transaction or is unable to secure financing, a title commitment does not reflect the terms of a later-issued policy.
Underwriting counsel also conducted the same file-by-file review of a random sample of 100 transactions out of the 9,500 that the Bouchers identified as candidates for class membership. Stuczynski Decl. (Dkt. # 173) ¶¶ 39-40; see also Simon Decl. (Dkt. # 172) (establishing that First American used outside consultant to ensure that the 100-transaction sample was random). In counsel's view, 13 of them were not refinance transactions, 26 of them did not extinguish an existing prior lien, 10 of them were "Super Eagle" policies, 33 of them were customers who took out two loans, and 9 of them were the result of errors in FAST data. Id. ¶¶ 42-46. Eight of the 100 were customers who may have been overcharged, although there is no way to prove as much without examining the prior lenders' policies associated with each transaction. Id. ¶ 48.
The court emphasizes that it does not draw substantive conclusions about the merits of class members' claims from the parties' analysis of the data. The Bouchers dispute most of underwriting counsel's conclusions, and the court will not resolve those disputes. Indeed, First American appears to concede that it overcharged some of its customers (and undercharged others). Overcharging is unlawful in any event, and the court does not suggest otherwise. First American's analysis suggests, however, that instances of overcharging and undercharging were not systematic (as the Bouchers claimed), but rather the result of errors that are apt to occur in any set of hundreds of thousands of customer transactions.
The Bouchers responded to underwriting counsel's evidence by shifting their focus from improper denial of the reorganization discount to other examples of improper rate calculation. They argue, for example, that First American misinterprets its own rate manuals when calculating the Simultaneous Issue rate, improperly prices Super Eagle policies, and erroneously denies the reorganization discount to customers who paid off their prior mortgage. The Bouchers raised these arguments for the first time in the reply brief of their second class certification motion, more than three years after the Bouchers first asserted their claims. It is far too late for the Bouchers to develop new theories of overcharging. Even if it were not too late, the new theories would be no more amenable to class resolution than their initial claims based on the reorganization discount. Because the Bouchers' new arguments do not persuade the court, the court need not grant First American's motion to strike them.
The court further recognizes that First American has an obvious incentive to make review of the evidence seem as difficult as possible. It is thus possible that, with improvements in methodology, it would take less (perhaps much less) than 20 minutes to review each customer file. The court also observes that because First American understands its own records and recordkeeping systems much better than the Bouchers, it has a significant advantage in parsing each of its customers' files. It is possible, for example, that if the Bouchers could remedy their inability to view documents from the FAST repository, they might be better equipped to assess each class member's claim.
But observations about the conclusions to draw from First American's file-by-file review and precisely how time-consuming it would be are largely beside the point; First American has convinced the court that proving or disproving each class member's claim depends on a file-by-file review of all class members' transactions. This individualized inquiry is incompatible with a class action. Whatever the validity of each of First American's objections to the claims of the 74 potential class members that the Bouchers identified, the existence of the objections shows that proving each class member's claim would require a time-consuming individualized process. Although First American has not disproven any of the claims of the 74 customers the Bouchers identified, it has established legitimate disputes over the validity of each of their claims. First American established those disputes by undertaking a file-by-file review of each of the 74 customers' transactions. First American has the right to present similar evidence in response to every class member's claim. Moreover, on this record, the court can only conclude that the Bouchers would have to conduct the same review to have any hope of proving class members' claims. The Bouchers offer no means to do so manageably. On this evidence, common issues do not predominate, and a class action is not superior to other methods of adjudicating the claims of First American's customers.
In the May 2011 order, the court made a cursory finding that the Bouchers had identified common issues of law and fact in accordance with Rule 23(b)(2). Since then, the Supreme Court has held that merely identifying common issues of fact or law is insufficient. Wal-Mart Stores, Inc. v. Dukes, 131 S.Ct. 2541, 2551 (2011). Instead, a plaintiff must identify a common issue on which all class members' claims depend and must show that the common issue is capable of classwide resolution. Id. Were it necessary for the court to revisit its commonality finding, it might reach a different conclusion in light of Dukes.
--------
The inability to evaluate title insurance customers' overcharging claims without a file-by-file review of their transactions has proven fatal to other plaintiffs' efforts to certify similar classes in other states. In its prior order, the court cited several cases in which plaintiffs had successfully certified classes of title insurance customers who allegedly overpaid. See May 2011 ord. at 11 (citing Campbell v. First American, 269 F.R.D. 68 (D. Me. 2010), Lewis v. First American, 265 F.R.D. 536 (D. Idaho 2009), and Hamilton v. First American, 266 F.R.D. 153, 158-59 (N.D. Tex. 2010)). More recent cases, however, trend sharply against class certification in actions like this one, for many of the same reasons the court has identified in this order. The Fifth Circuit vacated the district court's certification order in Hamilton in light of Benavides v. Chicago Title Ins. Co., 636 F.3d 699 (5th Cir. 2011). Hamilton v. First American, 423 Fed. Appx 425 (5th Cir. 2011). In Benavides, the court affirmed a district court's refusal to certify a Texas- wide class of a title insurer's refinance customers. Among other things, the court agreed that the predominant issues to determine were "individualized inquiries as to whether particular persons qualify for the discount and were denied it." Id. at 703. The court in Scott v. First American, 276 F.R.D. 471 (E.D. Ky. 2011) denied class certification for similar reasons in a putative class on behalf on First American's Kentucky customers. See id. at 479 ("The moral of the story: there is no common method for ascertaining those members falling within the class parameters."), 481 ("While the issue of whether class members were overcharged . . . may - at some broad level of generality - be common to the class, determining whether an overage actually occurred is an individualized inquiry, subject to individualized proof."). In Corwin v. Lawyers Title Ins. Co., 276 F.R.D. 484 (E.D. Mich. 2011), the court declined to certify a similar class of Michigan title insurance customers. Id. at 490 (holding that plaintiff could not demonstrate commonality because "the critical inquiry without which liability cannot attach requires individualized determination"). The Sixth Circuit affirmed a district court's refusal to certify a similar class of title insurance customers in Randleman v. Fidelity Nat'l Title Ins. Co., 646 F.3d 347 (6th Cir. 2011). It agreed with the district court's conclusion that "liability could only be determined on an individual basis by examining each individual homeowner's file." Id. at 350, 353. The court noted held that common issues did not predominate because both class membership and liability depended on individualized review. Id. at 353-54.
For many of the reasons these other courts have articulated, the court declines to certify a class today. The court makes no finding as to whether First American has overcharged its Washington customers. It concludes only that the Bouchers fall short of convincing the court that a class action (or at least the class action they propose) is an appropriate method of proving that First American overcharged.
4. The Bouchers' Claims Are Not Typical of Other Class Members' Claims.
Before concluding, the court addresses one question it did not resolve in the May 2011 order: whether the Bouchers' claims are typical of other class members as Rule 23(a)(2) requires. Although the court did not make a typicality finding in its prior order, it noted the unusual circumstances underlying the Bouchers' insurance purchase, and suggested that they consider appointing at least one more class representative. May 2011 ord. at 14. The Bouchers have not offered another class representative. For the reasons the court has already stated, the Bouchers' claims are not typical even of other customers who purchased insurance from an agent. In light of the disparity between the evidence necessary to prove the claims of agent-issue customers and direct-issue customers, the Bouchers' claims are even less representative of customers who purchased insurance directly from First American. The Bouchers are subject to unique defenses based on agency law as well as problems of proof that apply only to agent-issue customers. This action demands division into subclasses of agent-issue and direct-issue customers. Direct-issue customers would require their own class representative. Because the Bouchers could presumably cure this defect with relative ease (despite their apparent reluctance to do so), the court does not base its denial of class certification on the Bouchers' failure to show that their claims meet Rule 23(a)(3)'s typicality requirement.
D. Summary
The court concludes with a summary of the reasons it declines to certify the class the Bouchers propose. The court incorporates by reference its discussion of the standards applicable to class certification motions from its May 2011 order.
1) Although the class the Bouchers hope to certify is likely sufficiently numerous, they cannot identify the members of that class with reasonable certainty, and thus cannot give appropriate notice to class members.
2) Although the Bouchers have identified some issues of law and fact that are superficially common to all class members, they cannot resolve most of them without conducting an individualized review of difficult-to-assemble evidence pertaining to each class member.
3) Although Ms. Boucher is an adequate class representative, her claims are not typical of other class members, both because of the unusual circumstances surrounding her title insurance purchase and because she cannot represent direct-issue customers.
4) Whatever common questions the Bouchers have identified, they do not predominate over individualized questions, because each class member's claim depends on individualized proof.
5) Although the relatively small amount of money at issue in each class member's claim mitigates in favor of resolving them in a class action, a class action is not superior in this case because of the individualized issues of proof the court has identified.
III. CONCLUSION
For the reasons stated above, the court DENIES the Bouchers' motion to certify a class. The parties shall meet and confer to discuss a schedule for trying the Bouchers individual claims. They shall file a joint statement presenting their views no later than August 22, 2012.
_____________
The Honorable Richard A. Jones
United States District Court Judge