Chawezi MWANTEMBE, et al. v. TD BANK, N.A., et al., 268 F.R.D. 548

MEMORANDUM OPINION SAVAGE, District Judge. Introduction In moving to certify this putative consumer class action against two banks that allegedly failed to disclose the deduction of dormancy fees from gift cards prior to their expiration date, the plaintiffs seek to represent all Pennsylvania residents who purchased from or received gift cards issued by Commerce Bank and TD Bank between 2004 and the present, and whose cards were assessed a dormancy fee.

As characterized by the plaintiffs, the “core” issue is whether the banks violated Pennsylvania’s consumer protection law by failing to provide any materials with the gift cards designed to inform the holder of the card’s issue date. They also contend that the banks breached their contracts with gift cardholders when they failed to properly disclose that the gift cards were subject to dormancy fees, either by not providing this information at all, or by failing to disclose the information conspicuously and clearly enough.

Opposing certification, the banks argue that the plaintiffs cannot meet their burden of proof to satisfy the typicality and adequacy prerequisites for a class action under Federal Rule of Civil Procedure 23(a), nor can they establish predominance and superiority as required by Rule 23(b)(3). They contend that the plaintiffs now assert a single theory: that class members were duped into holding their cards too long and incurring dormancy fees because they were not informed of the purchase date used to compute the period when the dormancy fees would start.

The banks are correct that a class action is not an appropriate vehicle for litigating this case. The plaintiffs fail to meet the typicality and adequacy prongs of Rule 23(a). Typicality is wanting because none of the three named plaintiffs incurred a dormancy fee because of her failure to know the gift card’s date of purchase or because she was not aware that dormancy fees would be charged at some point. Each named class member faces these unique defenses, which are markedly different from those of the putative class members.

These same differences result in interests so divergent that the named plaintiffs are inadequate representatives of the absent class members. Finally, the predominance and superiority requirements of Rule 23(b)(3) are lacking-predominance because the proposed common issues are overwhelmed by the differences among the factual and legal issues affecting individual causation and damages; and, superiority because the proposed class would be unmanageable in light of proving class membership, which would require individualized fact-finding. Therefore, the motion for certification will be denied.

Background The three named plaintiffs, Chawazi Mwantembe (“Mwantembe”), Margaret Munthali (“Munthali”) and Fern Rutberg (“Rut-berg”), assert causes of action under Pennsylvania law for violations of the Unfair Trade Practices and Consumer Protection Law (“UTPCPL”), 73 P.S. §§ 201-2(3), 201-2(4)(xxi), 201-3 (2008), breach of contract and third party beneficiary on behalf of Pennsylvania residents and Pennsylvania purchasers who held or hold gift cards sold by the banks where dormancy fees were imposed before each card’s “Good Thru” date.

After the banks’ motion to dismiss the amended complaint based on the federal preemption doctrine was denied, the parties conducted both class certification and merits discovery. Depositions of the three plaintiffs and eight witnesses for the banks were taken. More than 1500 pages of documents were produced. Both sides retained experts who issued reports. Factual Allegations and Legal Claims in the First Amended Complaint According to the amended complaint, the gift card at issue is a credit-card sized plastic card with a magnetic stripe on the back.

On the front, there appears a “Good Thru” date, in raised, large letters, and the value amount, which is the card’s value at the time of purchase. There is no issue date anywhere on the card. In very small print on the back of the card, it states, “Cardholder by using or permitting use of this Gift Card, you agree to the terms and conditions that accompanied the Card.” The gift card comes in a prepackaged decorative box which is tied shut. Inside the box, in a hidden pouch within a cardboard folding envelope, a piece of paper containing terms and conditions may be found.

Nowhere on the box or on the cardboard folding envelope is there notice of the material terms and conditions related to the card, or notice of the existence of the hidden pouch where the terms and conditions can be found. There is no procedure, such as an 800 number or a website address, available for a cardholder to ascertain the issue date or fees that have been deducted from the card.

The plaintiffs allege that after the “dormancy period,” a set period of time beginning on the date the card was purchased, the banks automatically deduct a $2.50 monthly “dormancy” fee, silently reducing the value of the card prior to the expiration date. Some gift cards are devoid of any disclosure of the dormancy fee on their face. Others contain a non-bolded notice concerning the dormancy fee in “minuscule font on the back corner” on the reverse side behind the raised-letter impressions from the front of the card, rendering the notice distorted and unreadable.

The plaintiffs contend that even if the cardholder knows that the card is subject to dormancy fees after a set period of time after the purchase date, she cannot calculate the potential diminution in the card’s value at a given time without knowing or being able to ascertain the issue date. They also maintain that deducting dormancy fees renders the “Good Thru” date and value amount displayed on the front of the card materially misleading, deceptive and confusing because the card will have either diminished or no value prior to the “Good Thru” date without the cardholder having made a single purchase.

According to the amended complaint, the banks marketed and sold the gift cards without adequately disclosing the material terms and conditions to purchasers and recipients. Prior to purchase, the banks’ representatives never discussed or otherwise disclosed to purchasers the dormancy and replacement fees, or issue and expiration dates. Additionally, the plaintiffs allege that the banks’ local branches advertising “free” and “no fee” gift cards is deceptive because the advertisements do not disclose the application of dormancy fees.

The amended complaint essentially defines the proposed class as Pennsylvania residents who purchased gift cards from Commerce Bank and/or TD Bank anywhere, and persons (no matter their state of residency) who purchased the banks’ gift cards in Pennsylvania, where dormancy fees and/or replacement card fees were deducted from the card balance prior to the “Good Thru” date as a result of the banks’ deceptive course of conduct and advertising. It also defines a subclass as “all intended third party beneficiaries of the contracts between the banks” and the class-presumably gift card recipients.

No time limit is placed on the class or subclass. Factual Allegations and Legal Claims in the Class Certification Phase Facts Established in Discovery Commerce Bank sold the gift cards until May 31, 2008, when it ceased operating, at which time TD Bank succeeded Commerce Bank. The gift cards are credit-card-sized pieces of plastic embossed with the Visa logo that can be used anywhere that Visa is accepted. Customers were not charged a fee when they bought the cards. In other words, they received a card for the full value of what was paid.

For most of the time period at issue, a monthly “dormancy” or maintenance fee of $2.50 was assessed beginning in the thirteenth month after the date of the card purchase. The fee was assessed until the value on the card was depleted, either through purchases, fees or a combination of both. The monthly fee never exceeded the residual value of the card. A statement of terms and conditions that disclosed the dormancy fee accompanied all gift cards at the time of sale. Prior to the fall of 2007, the dormancy fee was not disclosed on the back of the cards.

Since the fall of 2007, in addition to being explained in the terms and conditions accompanying each card, the dormancy fee was disclosed on the back of all cards as follows: A monthly service fee of $2.50 will be deducted from your balance starting on the first day of the month immediately following the 365th day after the date your card was purchased. There has always been an 800 number on the cards that could be called to ascertain the card’s date of purchase and to register the card.

Since 2005, both banks had a website that could be accessed to obtain the card’s purchase date and statement of terms and conditions, and to register the card. Each card has a “Good Thru” or “Active Thru” date printed on the front as required by Visa. Like an expiration date on a credit card, it is the date when the card no longer works. The cardholder can request a replacement card and then continue to use the card until the balance is depleted.

The banks can trace information about the card’s use and the fees that were deducted if the cardholder is the purchaser and had an account at the bank or if the card recipient registered the card. If the recipient did not register the card, it cannot be traced with the recipient’s name, unless he or she is still in possession of the card. Class Definition In their class certification motion, the plaintiffs narrowed the proposed class definition.

The new proposed class definition is as follows: All Pennsylvania persons and entities who purchased or received Commerce Bank, N.A. or TD Bank N.A. gift cards on which dormancy fees were assessed from 2004 to the present. Then, in their reply brief, plaintiffs propose to divide the class into two subclasses: one comprised of gift card recipients whose cards were assessed dormancy fees; and the other, gift card purchasers who never gifted the cards or whose recipients assigned their rights regarding the card back to them.

UTPCPL Claim Although the plaintiffs continue to assert claims based on a violation of the UTPCPL, breach of contract and third party beneficiary, they have trimmed the factual allegations for these claims. The factual basis for their UTPCPL claim, which they initially describe as forming the crux of all of their claims, is that the banks engaged in a deceptive practice in which they intended to “trick and confuse consumers” into not retaining or claiming the full value of the card by failing to provide any materials with the card designed to inform the cardholder of the card’s issue date.

Because the dormancy fee is tied to the card’s issue date, even if the cardholder knows the card is subject to a dormancy fee after a set period of time starting on the card’s purchase date, she cannot determine the value of the card or the potential loss in value unless she also knows the issue date.

The plaintiffs also contend that deducting dormancy fees renders the “Good Thru” date and value amount displayed on the front of the card materially misleading, deceptive and confusing because the card can have either diminished or no value prior to the “Good Thru” date without the cardholder having made a single purchase.

Breach of Contract/Third Party Beneficiary Claims The foundation for the breach of contract and third party beneficiary claims rests upon the banks’ alleged failure to properly disclose that the gift cards were subject to dormancy fees by failing either to provide this information at all, or to disclose the information conspicuously and clearly. Specifically, there was nothing regarding a dormancy fee on the actual gift cards from 2004 through 2007. Since 2008, a dormancy fee was referenced on the back of the card, but it was much smaller than all other information on the card.

In inserts accompanying all of the cards, information regarding the dormancy fee was not in bold or set off from the other fine print, or the insert itself was hidden in the gift card box. Because they are not aware of the dormancy fees, purchasers and recipients of the gift cards cannot get the full benefit of the bargain entered into by the purchaser and the bank.

The plaintiffs also assert that because these contract terms regarding the dormancy fees unreasonably favor the bank, and the gift card purchasers did not have an opportunity to negotiate the contract terms, it is a contract of adhesion, rendering the dormancy fees unenforceable. In that case, the collection of the dormancy fees is a breach of contract, and the fees deducted constitute the damages resulting from the breach. Facts and Claims Post-Discovery Based on the new proposed class definition, the plaintiffs no longer assert any claims based on fees other than dormancy fees, e.g., replacement card fees.

In addition, they have abandoned their false advertising claim based on the banks’ advertising “free” and “no fee” gift cards, because, according to plaintiffs, “Pennsylvania law precludes certification of such actions.” It is undisputed that the issue date has never been imprinted or labeled on the card itself, and that no insert stated the issue date. However, the plaintiffs’ allegation that there is no procedure, such as an 800 number or a website address a cardholder can use to ascertain the issue date or fees that have been deducted from the card, is incorrect.

The banks have shown, by a preponderance of the evidence, that a cardholder could always obtain the issue date, balance on the card and actual fees deducted using an 800 number; and since 2005, a website could be used to obtain the same information. Facts Pertaining to the Three Named Plaintiffs Chawezi Mwantembe Mwantembe purchased five to seven gift cards from Commerce Bank in December of 2005 through 2007. She gave them to relatives who live abroad, and to her children and niece who live in Pennsylvania. She has never received any gift cards.

She never read the terms and conditions written on or that accompanied the cards. At the time she purchased the gift cards, she assumed no dormancy fees would be assessed. Although she does not know if any specific fees were charged against any of the cards, she suspects there was some kind of fee deducted in 2005 when her son’s card was declined for lack of value when there should have been money left on it. Of the three cards the banks have records of Mwantembe purchasing, only one of them incurred any fees.

That card, purchased on December 24, 2005 for $100, had a balance of only 29 cents after $99.71 had been spent between December 27, 2005 and January 10, 2006. Almost one year after the card was last used, a 29-cent dormancy fee was assessed on January 3, 2007. Mwantembe wants the fee refunded to her, reserving the right to choose whether to give the fee back to the recipient. Margaret Munthali Munthali received five gift cards between 2006 and 2008. She never purchased any.

Two cards were given to her by co-plaintiff Mwantembe: one from Commerce Bank in December of 2006, and the other from TD Bank in December of 2007. In December of 2007, her employer gave her an additional gift card. Customers at a restaurant where she worked gave her two cards, in January of 2008 and December of 2008. She believes she spent the entire value of four of the five cards, and all but 93 cents of one. She does not know if any specific fees were charged against any of the cards.

Regarding the two cards Mwantembe gave her, Munthali stated there was no notice on the Commerce Bank card or the box about a dormancy fee. Nevertheless, she expected that the card would incur a dormancy fee after a year. Her expectation was based on her experience with other gift cards and the notice on the TD Bank card stating that the card was subject to a monthly fee after twelve months. Because she spent the full value of each card within one year, no fees were assessed on the two cards. She “knows” the card given to her by one of her customers in January of 2008 had no disclosure about dormancy fees.

But, because she spent the entire value of that card and the card she received from her employer within one year, no fees were assessed. Munthali registered only one of the cards she received, the one given to her by one of her customers in December of 2008. According to the banks’ records, a dormancy fee of 93 cents was assessed on that card, which was purchased on December 20, 2008 for $100. Munthali spent all but 93 cents of the card by April 7, 2009, less than four months after it was purchased. On January 20, 2010, the card was assessed a 93-cent dormancy fee.

The registered card came with an insert containing terms and conditions, which she did not read. “[T]o the best of [her] knowledge,” the card itself did not have a notice about the dormancy fees. Yet, from reading notices on cards she had received previously, she assumed this card would be subject to similar fees. She lost the card when she moved in August of 2009. She had not used it before losing it because it had only 93 cents remaining and the least expensive item she wanted to buy was a media file from iTunes for 99 cents.

She admitted that the 93-cent fee did not affect her because she had lost the card four months before the fee was assessed. Munthali “would like” the 93 cents in fees to be refunded both to her, as card recipient, and to the card purchaser. Fern Rutberg Rutberg purchased four gift cards from Commerce Bank: two in December of 2007 for $340 each; and two in June of 2008 for $280 each. The cards were purchased with funds pooled from a group of parents and given to two teachers. Rutberg contributed $25 toward each gift card. Only one card incurred a fee.

A charge of 2 cents was imposed on one of the $280 cards purchased in June of 2008. Between December 5 and 23, 2008, all but 2 cents had been spent. On July 1, 2009, a 2-cent dormancy fee was charged. Rutberg testified that when some parents objected to giving the teachers gift cards because there could be fees, she reassured them that Commerce Bank gift cards had no fees.

She gave this assurance based on signs she had seen in the bank branches and notices in her bank statement describing the gift cards as “Free, no fees.” Additionally, she claims that the bank customer service representative told her there were no gift card fees. Rutberg wants the 2 cents refunded to the gift card recipient. Elements of the Cause of Action In evaluating whether the requirements of a class action have been met, we must examine the facts that each plaintiff would have to prove at trial.

To establish a claim under the UTPCPL, a plaintiff must prove: (1) she purchased or leased goods or services; (2) the goods or services were primarily for personal, family or household purposes; and (3) she suffered an ascertainable loss as a result of the defendant’s unlawful, deceptive act. 73 P.S. § 201-9.2(a). The plaintiff must show that the loss was caused by her justifiable reliance on the deceptive conduct. Hunt v. U.S. Tobacco Co., 538 F.3d 217, 221-22 (3d Cir.2008); Weinberg v. Sun Co., Inc., 565 Pa. 612, 777 A.2d 442, 446 (2001).

The elements of a breach of contract claim are that there was a contract, the defendant breached it, and the plaintiffs suffered damages from the breach. McShea v. City of Philadelphia, 995 A.2d 334, 340 (Pa. 2010). To be deemed a third party beneficiary, both parties to the contract must have intended that the non-party be a beneficiary of the contract at the time the contract was formed. Burks v. Fed. Ins. Co., 883 A.2d 1086, 1088 (Pa.Super.2005) (citing Scarpitti v. Weborg, 530 Pa. 366, 609 A.2d 147, 149 (1992)).

Legal Standards for Class Certification To be certified, a class must satisfy all four requirements of Rule 23(

a) and must fit one of the types of class actions described in Rule 23(b). The plaintiffs must demonstrate that: (1) the size of the class is so numerous that joinder of all members is impracticable; (2) there are questions of law and fact common to the class; (3) the claims or defenses are typical of the class; and (4) the representatives will fairly and adequately protect the interests of the class. Fed.R.Civ.P. 23(a). Additionally, the proposed class action must be one of the types recognized by Rule 23(b). Danvers Motor Co., Inc. v. Ford Motor Co., 543 F.3d 141, 147 (3d Cir.2008).

Here, plaintiffs have moved for certification under subsection (b)(3), which requires a finding that common questions of law or fact predominate over questions affecting only individual class 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). Rule 23(

a) is more than a pleading rule. In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 316 (3d Cir.2008). Instead, it sets forth requirements that a plaintiff must satisfy by a preponderance of the evidence. Id. at 320. In a motion to certify a class, the plaintiff must do more than make a threshold showing, more than a promise that she can meet the requirements. Id. at 321. Thus, a court must determine, after a “rigorous analysis,” whether the plaintiff has established, by a preponderance of the evidence, facts necessary to satisfy the Rule 23(

a) prerequisites for class certification. In re Sobering Plough Corp. ERISA Litig., 589 F.3d 585, 595-96 (3d Cir.2009) (quoting Beck v. Maximus, 457 F.3d 291, 297 (3d Cir.2006)). The analysis necessarily looks beyond the complaint and the certification motion. It entails a critical review of the elements of the cause of action through the “prism of Rule 23.” In re Hydrogen Peroxide, 552 F.3d at 311 (quoting Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 181 (3d Cir.2001)). Consequently, it implicates the merits.

Resolving factual and legal disputes necessarily requires some inquiry into the merits. Hohider v. United Parcel Serv., Inc., 574 F.3d 169, 197 (3d Cir.2009) (quoting In re Hydrogen Peroxide, 552 F.3d at 317). Credibility determinations, though not binding on the fact finder, must be made. In re Hydrogen Peroxide, 552 F.3d at 318. Often times, resolving conflicts among experts will be necessary. Id. at 323-24. The Four Requirements of Rule 23(

a) Numerosity The banks do not challenge numerosity. Nor could they. Between 2004 and 2009, the banks assessed at least one monthly dormancy fee on 297,863 gift cards purchased in Pennsylvania. Thus, numerosity is not an issue. Commonality Commonality requires that the plaintiffs share a question of law or fact with the prospective class members. The commonality threshold is low. So long as the named plaintiffs share at least one question of fact or law with the grievances of the prospective class, the existence of individual facts and circumstances will not defeat commonality.

In re Sobering Plough, 589 F.3d at 596-97 (quoting Baby Neal v. Casey, 43 F.3d 48, 56 (3d Cir.1994)). The plaintiffs propose several common questions, but they are essentially the same common question: whether the defendants breached their contract and violated the UTPCPL by failing to provide materials with their gift cards designed to reach the recipient of the gift cards that state the issue date of the card. The banks nominally challenge commonality. They argue that commonality is lacking because none of the three plaintiffs was injured as a result of not knowing the card’s issue date.

Essentially, they contend that the plaintiffs cannot establish reliance, a necessary element of their UTPCPL cause of action. This reliance element is relevant to the analysis of typicality and adequacy of representation. It is not critical to commonality. The plaintiffs have identified at least one common question. It is whether the banks’ failure to disclose information regarding the card’s issue date violated the UTPCPL. Because they are required to share only one question of fact or law with the grievances of the prospective class, they overcome the commonality hurdle of 23(a)(2).

Typicality The typicality prong of Rule 23(

a) requires that the claims or the defenses of the plaintiffs are typical of the class. Fed. R.Civ.P. 23(a)(3).

There are three “distinct, though related,” parts to the typicality assessment: (1) the class representative and the members of the class must share both the legal theory and the factual circumstances supporting that theory; (2) the class representative must not be subject to a defense that is both inapplicable to many members of the class and likely to become a major focus of the litigation; and (3) the class representative’s interests and incentives must be sufficiently aligned with those of the class. In re Schering Plough, 589 F.3d at 598-99.

Typicality requires a strong similarity of legal theories to ensure that the class representatives’ pursuit of their own goals will work to benefit the entire class. Barnes v. Am. Tobacco Co., 161 F.3d 127, 141 (3d Cir.1998) (citing Baby Neal, 43 F.3d at 57); Jones v. GPU, Inc., 234 F.R.D. 82, 97 (E.D.Pa.2005). Typicality is challenged where the plaintiffs claims are subject to unique defenses that may become the focus of the litigation. In re Schering Plough, 589 F.3d at 598 (quoting Beck, 457 F.3d at 301); Jones, 234 F.R.D. at 98. Nevertheless, factual differences will not necessarily defeat typicality.

As long as the claim arises from the same event, practice or course of conduct that affects the class members, and is based on the same legal theory, typicality is satisfied. Baby Neal, 43 F.3d at 58 (quoting Hoxworth v. Blinder, Robinson & Co., 980 F.2d 912, 923 (3d Cir.1992)). Unlike the commonality requirement, however, typicality requires more than just “one unifying factual or legal question.” In re Paxil Litig., 212 F.R.D. 539, 550 (C.D.Cal.2003).

Because we cannot assess whether an individual is sufficiently similar to the class as a whole without knowing something about both the individual and the class, we must consider the attributes of the proposed representatives and the class as a whole, and the similarity between the proposed representatives and the class. In re Schering Plough, 589 F.3d at 597-98. Here, an examination of the numerous factual and legal differences between the three representatives and the class members reveals how marked the differences are. These differences overwhelm any similarities, defying typicality.

Rutberg Because Rutberg purchased, as opposed to received, gift cards, she cannot assert that she did not know the purchase date of the cards. Her factual and legal circumstances do not fit within the “central” claim asserted by the putative class that the banks violated the UTPCPL in failing to provide any materials with the gift cards designed to reach the recipient that stated the issue date of the card in conjunction with charging an automatic dormancy fee tied to the card’s issue date. Consequently, she is not typical.

Rutberg’s claim, as she testified, is for misrepresentations that the gift cards had no fees and were free. She contends that she was falsely informed by a bank employee, posters in bank branches and notices included with her bank statement that there were no fees. Her claim differs from the “purchase-date” theory asserted by the proposed class.

The former assumes that she was misled into believing that no fees would ever be assessed against the card, and the purchase-date theory assumes that the cardholder knew fees could be charged but did not know the purchase date and, therefore, was unable to determine when the twelve-month grace period would expire. The claim of Rutberg’s gift card recipients fares no better. Only one of the four cards incurred a dormancy fee.

The plaintiffs have proffered no evidence that the recipient of this card would have spent the 2 cents had she known the purchase date and/or the written terms and conditions describing the dormancy fee. Six months after Rutberg purchased the card, all but 2 cents of the $280 value was spent in a period of less than three weeks. A 2-cent fee was assessed more than six months after the card’s last use.

These facts, coupled with the de minimus amount, show that it is more likely than not that the recipient would not have spent the remaining 2 cents, whether or not she knew about the purchase date and understood fully the terms of the dormancy fee. These circumstances are atypical of those of the class. Mwantembe Like Rutberg, Mwantembe was a purchaser. Consequently, she too cannot assert that she did not know the purchase date of the cards and that the banks concealed the issue date from her, which is the “central” theory asserted by the putative class. Thus, her situation does not fit that of the class members.

Also similar to Rutberg, none of Mwantembe’s gift card recipients can assert a claim that is factually or legally similar to those of the putative class. None of them can claim that he or she was misled into incurring fees due to ignorance of the purchase date or of the dormancy fee or its terms. Only one of the three cards that Mwantembe purchased incurred a dormancy fee. Within seventeen days after Mwantembe purchased the card, $99.71 of the $100 was spent. A 29-cent dormancy fee was assessed almost a year later.

There is no evidence that the recipient was misled into not using the 29 cents because he or she did not know the purchase date, the terms and conditions governing the card, or that the card could be subject to dormancy fees. Instead, the fact that the card’s value was almost totally depleted in only two weeks immediately after the card was received and there was a de minimus balance of 29 cents for almost a year show that it is more likely than not that the recipient would never have spent the remaining 29 cents, whether he or she knew about the purchase date and fully understood the terms of the dormancy fee.

Mwantembe’s position does not support the legal theory of the action. Munthali Unlike Rutberg and Mwantembe, Munthali was not a purchaser of gift cards. She was a recipient of five gift cards, including two from Mwantembe. Only one card was charged a dormancy fee. The card was purchased on December 20, 2008. Less than four months later, Munthali spent all but 93 cents of the card's value. She lost the card in August of 2009. Five months later, on January 20, 2010, the card was assessed a 93-cent dormancy fee. Munthali did not read the insert containing the terms and conditions governing the card.

Although the card did not contain a notice about the dormancy fees, she assumed it would be subject to such fees because she had read such a notice on cards she had received previously. Although Munthali would not expressly admit that she never intended to use the remaining 93 cents on the card, we find that she did not. She testified that she did not use her card before losing it because the least expensive item she wanted to buy was a song from iTunes for 99 cents and the card had only 93 cents remaining.

She admitted that the 93-cent fee did not affect her because she had lost the card four months before the fee was assessed. Like her co-plaintiffs, Munthali is not a typical class representative. First, she admitted that she was not harmed by the 93-cent dormancy fee because she had lost the card. Second, Munthali knew that the card was subject to dormancy fees. Based upon her prior experience with other bank gift cards, she assumed that the card was subject to a dormancy fee within a year of the card’s issue. Third, she received the card three weeks after she filed this lawsuit.

When she filed this lawsuit, she knew there were dormancy fees triggered by the purchase date. This actual knowledge that dormancy fees were to be charged after a year provides a unique defense to her individual claim for breach of contract, rendering her an atypical class representative.

Any similarity in legal theories among the named plaintiffs and the proposed class of plaintiffs is eclipsed by the individualistic defenses the banks can raise to each plaintiffs claim. “Where the defendant can raise unique defenses to each plaintiffs claim, typicality may not exist if the defenses could threaten to become the focus of the litigation.” Jones, 234 F.R.D. at 98. The danger is that the class representatives will be preoccupied with meeting and defeating those defenses unique to them at the expense of those issues that they share with the class members, a problem also implicating adequacy.

The factual circumstances of the plaintiffs do not support the legal theory. Consequently, they differ from the class members. The defenses the banks may raise in light of each class representative’s unique situation will not be available as to most class members and will dominate the litigation. Faced with their own factual position vis-a-vis the banks, the plaintiffs will be forced to defend their own cases.

Therefore, because the individual circumstances of each of the named plaintiffs are so markedly different from those of the absent class members, the plaintiffs have failed to meet the typicality requirement. Adequacy of Representation Rule 23(a)(4) aims to protect the interests of the class. There are two parts to this test. One assesses the class representatives’ motivation and ability to protect the interests of the class members, and the other goes to the competency of counsel. The banks contend that neither prong of the adequacy test is met.

The test of the adequacy of the named class members is to assure that there are no conflicts of interest between the class representatives and the class members. Schering Plough, 589 F.3d at 602 (quoting In re Warfarin Sodium Antitrust Litig., 391 F.3d 516, 532 (3d Cir.2004)). This test tends to merge with the typicality requirements that the interests and incentives of the representatives be sufficiently aligned with those of the class and the class representatives not be subject to unique defenses. Schering Plough, 589 F.3d at 602 (quoting Beck, 457 F.3d at 296).

The banks argue that there is a conflict of interest between the named plaintiffs and those they seek to represent because the purchasers and the recipients of the gift cards are competing for the same pot of money. The named plaintiffs themselves disagree as to who would receive the refunded fees. Munthali wants any refunded fees to go to both her and the card purchaser. Mwantembe wants them paid to her, the purchaser. Rutberg believes they should be paid to the card recipient.

To overcome this conflict, the plaintiffs changed the class definition to accommodate two subclasses: (1) recipients whose cards were assessed dormancy fees; and (2) purchasers who never gifted the cards or whose recipients assigned their rights regarding the card back to them. This breakdown eliminates the risk of two parties trying to recover from the same fund. The new subclasses create additional problems. Mwantembe is not a member of either subclass. Rutberg is only a member based on counsel’s “understanding” that her recipient “assigned” her rights to the card—2 cents—to Rutberg.

Additionally, the same factual and legal differences among the named plaintiffs and the unnamed class members that defeat typicality render plaintiffs inadequate representatives of the putative class. For example, if the named class members are subject to unique defenses, they may have little, if any, incentive to spend the time, energy and money pursuing the claim. Even if their claims are not barred, the maximum amount that each named plaintiff could recover is between 2 and 93 cents, which is hardly a large enough financial stake to motivate them to fairly and adequately represent the class.

Additionally, one of the class members, Mwantembe, testified that she, not her recipients, should be compensated for the fees assessed, but the new class definition precludes her from obtaining any award unless her recipients assign their rights to recovery to her. Finally, both Mwantembe and Rut-berg testified that they have no obligation to protect the interests of the absent class members. These divergent interests and circumstances will impair the plaintiffs’ ability to adequately protect the interests of the class members.

Therefore, the plaintiffs have failed to satisfy this part of the adequacy requirement. Turning to the adequacy of putative class counsel, Rule 23(

g) governs the analysis. It provides that in appointing class counsel, the court must consider four factors: (

i) the work counsel has done in identifying or investigating potential claims in the action; (ii) counsel’s experience in handling class actions, other complex litigation, and the types of claims asserted in the action; (iii) counsel’s knowledge of the applicable law; and (iv) the resources that counsel will commit to representing the class. Fed.R.Civ.P. 23(g)(1)(A). See also Sheinberg v. Sorensen, 606 F.3d 130, 132-33 (3d Cir.2010).

In addition, the court must determine if counsel will fairly and adequately represent the interests of the class, Fed.R.Civ.P. 23(g)(4), and “may consider any other matter pertinent to counsel’s ability” in order to do so. Fed.R.Civ.P. 23(g)(1)(B). Class counsel is not appointed until the action is certified. See Sheinberg, 606 F.3d at 132; In re Gen. Motors Corp. Pick-Up Truck Fuel Tank, 55 F.3d 768, 788 (3d Cir.1995). Because certification will be denied, we need not consider this prong at this time.

Rule 23(b)(3)—Predominance and Superiority In addition to meeting the four requirements of 23(a), the action must also qualify as one of the types of class actions described in Rule 23(b). In this ease, plaintiffs have moved for certification under subsection (b)(3), which requires that common questions of law or fact predominate over questions affecting only individual class members, and that a “class action is superior to other available methods for the fair and efficient adjudication of the controversy.” Fed.R.Civ.P. 23(b)(3).

Thus, the plaintiffs must satisfy both the predominance and the superiority aspects of Rule 23(b)(3). In determining whether the action fits within Rule 23(b)(3), we consider the interest of class members in individually controlling the litigation, the status of ongoing litigation brought by members of the class, the desirability of concentrating the litigation in the particular forum, and likely management difficulties. Fed.R.Civ.P. 23(b)(3)(A)-(D). In the end, it is the interests of the individual members in controlling their own litigation that drives the certification decision on predominance.

The superiority analysis focuses on the advantages and disadvantages of using the class-action device in relation to other litigation methods. Predominance In a class action brought under Rule 23(b)(3), common questions of law or fact must predominate over questions affecting only individual members and must be a significant part of the individual cases. Predominance “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation,” a standard “far more demanding” than the commonality requirement of Rule 23(a).

In re Hydrogen Peroxide, 552 F.3d at 311 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)). Thus, we must predict how specific issues will play out at trial in order to determine whether common or individual issues predominate in a given case. Id. (citations omitted). Class certification is not appropriate if proof of the essential elements of the cause of action requires individual treatment. Id. (quoting Newton, 259 F.3d at 172).

Additionally, subdivision 23(b)(3) encompasses those cases in which a class action would achieve economies of time, effort, and expense, and promote uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results.... It is only wh[en] predominance exists that economies can be achieved by means of the class-action device. Fed.R.Civ.P. 23(b)

(3) Advisory Committee’s Note. For the same reasons typicality and adequacy are lacking, and because determining liability in each putative class member’s case will require an individualized factual inquiry, the predominance requirement is not satisfied. The number and complexity of the questions that must be resolved to determine liability in each individual’s case predominate over any common questions. First, establishing that the putative class member knew the purchase date of the card is an essential element of the UTPCPL and the breach of contract claims.

If a class member knew the purchase date, she could not have sustained damages as a result of a failure to disclose the date. All purchasers necessarily knew the purchase date. Some card recipients also knew the date because the purchaser told them. The only way to determine if such a communication occurred is to question the purchaser and recipient in each transaction. Additionally, the recipient may have learned the purchase date by calling the 800 number on the card or going to the website. There is no record of who inquired using either of these methods.

In all of these situations, the determination of the recipient’s knowledge will require individualized inquiry. Second, for a class member to prevail on his UTPCPL claim, he must show that he relied to his detriment on the deceptive omission of the purchase date. In other words, he must prove that because of the failure to disclose, he did not spend the balance before the imposition of the dormancy fees.

Similarly, to prevail on his contract claim, a class member must show that his card incurred the dormancy fee because the banks failed to conspicuously and clearly notify him about the dormancy fees and when they would accrue. Third, there are multiple possibilities why a gift card incurred a dormancy fee. Of course, in some cases, fees may have been charged because the recipients did not know the purchase date and did not know when the fees started. In other cases, recipients, who may not have known the purchase date, are charged fees after they abandoned the cards because they had minimal value left.

There are cardholders who knew the card’s purchase date and would have spent the total amount had they known dormancy fees applied. Then, there are those who knew the cards were subject to dormancy fees, but thought it did not start until the “Good Thru” date. Because not all cards incur dormancy fees for the same reasons, proof of reliance and causation in a contract claim is inherently individualized. The need for individualized inquiry is exemplified by the circumstances of each of the named plaintiffs. Munthali’s card incurred the fee because she lost it, not because she did not know the purchase date.

In addition, based on her experience, she knew the card was subject to a fee after a period of time. The recipients of Mwantembe and Rutberg’s cards and Munthali spent their cards down to such small balances that they were virtually worthless. Finally, to have standing to bring a UTPCPL claim, the plaintiff must show that the card was used for personal or household purposes, not for business. This would exclude any recipient who spent the card on items used for business purposes, such as parts or equipment in a business.

Presumably, the vast majority of cardholders used the cards for personal or household products or services. That does not preclude the probability that some, albeit a much smaller number, used them for business purposes. Thus, individualized inquiry would be required to determine a necessary element of the UTPCPL cause of action. In short, determining membership and liability in each putative class member’s case will require an individual fact intensive inquiry that will overshadow any common questions.

Superiority Not only do the plaintiffs fail to satisfy the predominance prerequisite, they cannot establish that a class action is superi- or to other available methods for the fair and efficient adjudication of the controversy. Fed.R.Civ.P. 23(b)(3). The superiority analysis assesses the advantages and disadvantages of using the class-action device in relation to other methods of litigation. Danvers, 543 F.3d 141.

The rule itself suggests various factors to consider in making this assessment: the interest of class members in individually controlling the litigation; the state of ongoing litigation brought by class members; the desirability of concentrating the litigation in the particular forum; and likely management difficulties. Id. It requires a rigorous analysis as to how potential relief to individual class members would be managed, “specifying, for example, the methodology by which calculations and awards of relief would be made with respect to individual class members.” Hohider, 574 F.3d at 202.

Class members’ interest in individually controlling their lawsuit Whether putative class members have a significant interest in individually prosecuting their own separate lawsuits is affected by the financial stakes involved in each individual’s case. The greater the damages in one’s claim vis-á-vis others’ claims, the greater the interest the individual has in controlling the litigation. The lesser the potential damages, the less the interest is because separate suits may be impracticable. See Fed.R.Civ.P. 23(b)

(3) Advisory Committee’s Note; Blain v. Smithkline Beecham Corp., 240. F.R.D. 179, 192 (E.D.Pa.2007). The potential damages in each class member’s case are small. Class members do not have a strong interest in controlling their own litigation. Separate actions would be impractical and expensive to litigate. Individuals would have little incentive to direct the litigation. In sum, their individual interests are minimal compared to the overall extent of the issues in the litigation. Extent of existing litigation There is no other existing litigation that would overlap with the claims at issue in this case.

The New Jersey putative class action, which is in the same procedural stage as this action, was filed on behalf of New Jersey residents only. The New York action, filed on behalf of New York residents only, is still in the pleading stage. Appropriateness of forum Because the proposed class consists of only Pennsylvania residents, this forum is appropriate. Manageability of proposed class and choice-of-law impediments In examining the manageability of the proposed class, we consider the manageability of the class for trial, and whether there are choice-of-law conflicts. In re Prempro Prods. Liab.

Litig., 230 F.R.D. 555, 562 (E.D.Ark.2005). Because this is not a nationwide class, choice of law issues are not present. The particularized circumstances of each class member as described in the typicality and predominance analysis defy manageability. “[Cjomplicated mini-litigations” on factual issues would be required to determine liability and damages, which would not be an efficient use of the class action device. Danvers Motor Co., 543 F.3d at 149. Additionally, the plaintiffs have failed to devise a method of determining class membership without individualized fact finding.

Although all card purchasers had either an account or a loan with the bank, identifying purchasers and recipients will not be reliable because the information pertaining to the initial gift card purchase was stored by the bank in a different database system than was the purchaser’s address and the information about the fees and maintenance of the cards. Further, the names of purchasers were entered manually into the bank’s card purchase records, and the names were not linked to the bank account information.

Consequently, there is no practical or efficient way to identify purchasers of gift cards on which fees were assessed. Moreover, without names and addresses, notice by mail cannot be effectuated. The fact that gift cards are freely transferable makes the last recipient-the class member-even harder to identify and locate. All of these issues would have to be resolved on an individual basis. There is no manageable way to do it. Indeed, plaintiffs have not proposed any.

See Danvers, 543 F.3d at 149 (finding manageability difficulties from the “multitude of individualized issues presented in plaintiffs’ claims [that] would entail complicated mini-litigations within the class action itself’). Despite the numerous problems presented with adjudicating, administering and trying these cases, the plaintiffs have not submitted a proposed trial plan addressing these problems. In short, the plaintiffs have not demonstrated that their proposed class action is superior to other available methods.

Conclusion The proposed class does not satisfy the typicality and adequacy requirements of Rule 23(a), nor the predominance and superiority elements of Rule 23(b)(3). Therefore, the motion for class certification will be denied, . Dormancy fees are monthly charges deducted from a card, beginning a period of time after its purchase date. "Dormancy fee” is a misnomer. The fee deducted is not triggered by inactivity, but by the passage of time. The fee, more akin to a maintenance fee, is charged against any unused balance remaining after the passage of the set period. .

The banks refer to this theory as the "purchase-date” theory. They exaggerate the extent to which plaintiffs have abandoned claims other than the purchase-date claim. Plaintiffs have withdrawn only one claim: false advertising based on misleading posters and press releases advertising "free” or "no fee” cards. They ignore the plaintiffs' breach of contract claim based on the failure of the banks to properly or conspicuously disclose that the gift cards were subject to dormancy fees. . The plaintiffs voluntarily dismissed defendants Toronto-Dominion Bank and its predecessor in interest, Commerce Bancorp, Inc.

The remaining defendants are TD Bank, N.A. and Commerce Bank, N.A. They also withdrew causes of action for civil conspiracy and unjust enrichment, as well as all claims alleging avoidance or violation of the escheat law. See Mwantembe v. TD Bank, N.A., 669 F.Supp.2d 545, 547 nn. 2-3 (E.D.Pa. 2009). . See Nov. 17, 2009 Ord. Denying Mot. to Dismiss Am. Compl. (Doc. No. 30). . Am. Compl. ¶¶ 5v, 5vii, 5xiii, 35v, 35xiii, 36, 39, 42, 50-51, Ex. A; Pis.' Br. in Support of Mot. for Class Certification (Doc. No. 44) (hereinafter "Pis.’ Br.”) at 4-5. .

The amended complaint does not provide the text of the statement concerning the dormancy fee. . Am. Compl. ¶¶ 2, 5xvi, 46-48, 58. . Id. ¶¶ 5i, 5ii, 5vi, 5vii, 5viii, 35i, 35vi, 35vii, 35viii, 39, 41. . Id. ¶¶ 5xi, 5xii, 35xi, 35xii, 44, 64-69. . Id. ¶ 18. . The following facts have been demonstrated by a preponderance of the evidence. See In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir.2008) ("Factual determinations necessary to make Rule 23 findings must be made by a preponderance of the evidence.”). .

TD Bank, N.A. is the surviving entity from the May 31, 2008 merger of TD Banknorth, N.A., Commerce Bank, N.A. and Commerce Bank/ North. Prior to the merger, TD Banknorth, N.A. and Commerce Bank, N.A. marketed and sold gift cards in Pennsylvania. Since the merger, all activities related to sales and marketing of gift cards in Pennsylvania have been conducted exclusively by TD Bank, N.A. See Deck of Matthew Chevalier (Ex. 1 to Defs.' Br. in Opp'n to Pis.' Mot. for Class Certification (Doc. No. 46) (hereinafter "Defs.’ Br.”)), ¶¶ 3-6. . The exceptions to this fee

schedule were that no fees were assessed on any cards Commerce Bank sold between January to mid-November 2005; for cards sold in November and December of 2004, the $2.50 monthly fee was not imposed until July 2006; and for a short period of time around June of 2008, the monthly dormancy fee imposed began in the seventh month after purchase of the card and the fee was $3.50. .

The following language appeared on the back of these cards: "Cardholder by using or permitting use of this Gift Card, you agree to the terms and conditions that accompanied the Card.” See Deck of Lisa Moncilovich (Ex. 5 to Defs.’ Br.), ¶ 4, Tabs 1, 3-4. . Defs.’ Br. at 6; Moncilovich Deck (Ex. 5 to Defs.’ Br.), ¶¶ 4-5 and Tabs 5-9, 12. After the merger, slightly modified cards and terms and conditions were issued, reflecting the change from the Commerce Bank to the TD Bank branding and slightly new language on the cards, as well as different formatting and language on the terms and conditions. .

Defs.’ Br. at 6-7; Deck of Debra Colello (Ex. 6), ¶¶ 3-4; Deck of Eric Hansen (Ex. 7), ¶¶ 18-20; Deposition of James Grimmer (Ex. 4) at 80-81. . Pis.’ Br. at 11; Pis.' Reply Br. (Doc. No. 48) (hereinafter “Pis.’ Reply Br.”) at 15. . Pis.’ Br. at 5, 9, 15, 16, 23-24 (describing the banks' alleged failure to properly inform the cardholder of the issue date as "the gravamen at issue," "at the core of this litigation,” "the focus of any trial in this matter,” "the central legal issue” and "the basis of the Plaintiffs' and the Class' allegations”).

Nonetheless, in their reply brief, plaintiffs insist that the "foremost claim” for which they seek redress is the breach of contract claim. Pis.' Reply Br. at 5. . Pis.’ Br. at 3, 5, 8-10, 15-16, 21, 23-24; Pis.' Reply Br. at 9-10. . The plaintiffs submitted expert reports opining that these are characteristics of contracts of adhesion. Pis.' Reply Br. at 7-9. . Pis.' Br. at 3, 5, 7-8, 21; Pis.' Reply Br. at 5-9. . Pis.’ Reply Br. at 16. Plaintiffs provide no legal citation or other authority supporting this conclusion. .

Pis.’ Br. at 5-6 (citing deposition testimony of Matthew Chevalier, James Grimmer and Linda McGuigan). . Defs.' Br. at 6-7; Colello Decl. (Ex. 6), ¶¶ 3-4; Hansen. Decl. (Ex. 7), ¶¶ 18-20; Deposition of James Grimmer (Ex. 4) at 80-81. . See Deposition of Mwantembe (Defs.' Br. Ex. 30) at 22, 26-38, 49-51, 54-58. . Deposition of Munthali (Defs.’ Br. Ex. 29) at 24-28, 33-47, 50-57, 68-70, 83-84. . She registered this card when she called the 800 number on the card to complain that it did not work after she tried to use it for the third time. Once she registered the card, which she did online, it worked.

Munthali dep. at 33-34. . The banks can find information about a gift card belonging to a recipient only if the card was registered. . Munthali dep. at 40-42, 50-51. . Id. at 69-70. . Deposition of Rutberg (Defs.’ Br. Ex. 28) at 14-18, 53-57, 105-06. . Defs.' Br. at 32 n. 140. . Pis.' Br. at 13; Defs.' Br. at 54 n. 237. . Pis.’ Br. at 15-16. . Defs.’ Br. at 31-32 n. 140. . Specifically, she said that she did not spend the last 93 cents remaining on the card because she "hadn't found a suitable use for it yet at the time,” but denied that the amount was "too little money to be of use” to her.

Munthali dep. at 39-40. . At oral argument, plaintiffs' counsel described the assignment process as occurring during the class notification process, during which there will be an opportunity for recipients to assign their rights back to the purchaser. He conceded that we could not identify any assignments prior to certification of the class. See Tr. of Jul. 8, 2010 oral argument on Pis.’ Mot. for Class Certification, at 20-23. . Mwantembe dep. at 63-64; Rutberg dep. at 74. .

Although questions concerning the adequacy of class counsel were traditionally analyzed under the adequate representation requirement of Rule 23(a)(4) of the Federal Rules of Civil Procedure, those questions have, since 2003, been governed by Rule 23(g). See Fed.R.Civ.P. 23(g), 2003 Advisory Committee Note. That rule instructs that “a court that certifies a class must appoint class counsel.” Thus, under the plain language of the rule, a district court's decision to certify a class must precede the appointment of class counsel. Sheinberg v. Sorensen, 606 F.3d 130, 132-33 (3d Cir.2010). .

See Tr. of Jul. 8, 2010 oral argument, at 56. . With respect to the cardholders who did not intend to spend the last few cents on their cards, they suffered no damage and the bank gained no windfall when fees were assessed. The value remaining was not going to be spent. . The plaintiffs state that they can prove justifiable reliance class-wide on the UTPCPL claim because the evidence that each named plaintiff will use to prove individual reliance in her case is the same evidence that will show class-wide reliance, namely, that the lack of disclosure of the card issue date caused them to incur the dormancy fees.

Pis.' Br. at 25-26. However, as described in the discussion of typicality and adequacy, the named plaintiffs cannot show that the failure to disclose the purchase dates on the cards themselves caused the imposition of dormancy fees in their own cases, much less on a class-wide basis. . Although the plain language of Rule 23(b)(3) directs the court to consider these factors in evaluating both predominance and superiority, in a majority of cases, courts consider these factors solely with respect to making a determination of superiority. Moore's Federal Practice, §§ 23.44[1], 23.46[2][a] (3d ed.2006). .

The 2003 Comment to Rule 23 states that an "increasing number of courts require a party requesting class certification to present a 'trial plan’ that describes the issues likely to he presented at trial and tests whether they are susceptible of class-wide proof.” See 2003 Advisory Committee Note. Plaintiffs have not submitted a trial plan. . Information about fees charged to the card was managed by several third-party vendors.

Specifically, for cards purchased from Commerce Bank or former Commerce Bank branches from 2004 through September of 2009, the purchaser’s name, account number, and the amount and date of each gift card sold were recorded in Commerce Bank's computer system known as "Card Genie." Hansen Deck (Ex. 7 to Defs.’ Br.), ¶¶ 2-3. Then, Commerce Bank used third-party data processors to manage all of the information about gift card usage, including any fees assessed and any cards that were registered.

Commerce Bank has "look up” access to data maintained by the third-parly vendors, allowing it to look up individual gift cards, but not every vendor allows access to transaction data, such as fees charged, via this "look-up” access, requiring the banks to pay the vendor a fee for such information. Hansen Deck (Ex. 7 to Defs.’ Br.), ¶¶ 2-5. For cards purchased from TD Bank from 2007 through the present, the bank's "old gift card reporting database" contains the purchaser's account number and fees incurred, but not the purchaser’s name or address.

Since November 2009, the gift card database contains the purchaser's name, account number and social security number, but no information about fees incurred. From June 2008 through the present, TD Bank used a third-party data processor called Metavante to manage the information about gift card usage, including any fees assessed and any cards that were registered. TD Bank has “look up” access to data maintained by Metavante, allowing it to look up individual gift cards, but it cannot conduct “bulk searches” without negotiating with Metavante for the right to do so for a fee.

Colello Deck (Ex. 6 to Defs.' Br.), ¶¶ 5-7. . In light of this set-up, the only way to identify Commerce Bank gifts card purchasers whose cards incurred fees is to look them up on Card Genie one at a time and then look on other sources for addresses. Although this process could yield complete information in many cases, it is extremely time consuming and often yields incomplete results. The process involved in ascertaining information about the named plaintiffs' cards exemplifies the problems inherent in this case.

For example, in trying to locate information about cards purchased by Mwantembe, the banks were unable at first to find anything, even after plaintiffs identified her full name and maiden name in the complaint and in response to interrogatories. Only after taking her deposition and learning that she goes by the name "Bertha” were they able to locate the record of her card purchases in Card Genie. Colello Deck, ¶¶ 11-13, 15(a)-(c); Hansen Deck, ¶¶ 15-16.