ORDER DESIGNATING CLASS COUNSEL.
WALKER, District Judge.On June 30, 1994, the court issued an order soliciting bids from all law firms interested in being designated class counsel in this securities action brought on behalf of purchasers of Wells Fargo common stock. In response, three firms have submitted proposals: Lieff, Cabraser & Heimann; Milberg Weiss Bershad Hynes & Lerach; and Lowey Dannenberg Bemporad & Selinger.
In selecting amongst the three bids, the court’s fiduciary obligation to the plaintiff class compels it to secure the best representation possible. To do so, the court must examine not only quantitative factors, such as the cost of each film’s services, but also the various qualitative attributes of the three firms and their bids. In performing this analysis, the court must strive to emulate the arrangements and decisions that the class itself would make were it able to negotiate. See In re Continental Illinois Securities Litigation, 962 F.2d 566, 572 (7th Cir.1992) (in determining fee award, court must act as a “surrogate client” and award the fee that would have obtained had negotiations between counsel and the class been feasible).
Because a well-advised class in this case would seek to avoid unnecessary duplication of effort, the court earlier decided that only one firm should be selected to represent the class. Order dated June 30, 1994. Having reviewed the three submitted bids, the court concludes that the proposal submitted by Lieff, Cabraser & Heimann is the most favorable to the class. This order examines the three proposals and then explains the reasons why the Lieff firm’s bid is the superior one.
I '
At the time the bids were solicited, the court requested that each firm specify its charges for class counsel services as a percentage of any recovery attained for the class, with time and event contingencies of the firms’ choosing. Order dated June 30, 1994 at 14. In addition, interested firms were asked to describe their: (1) qualifications; (2) willingness to post a bond or other security toward complete performance of their duties; and (3) malpractice insurance coverage. Id. A summary of each firm’s bid follows.
A
The proposal submitted by Lieff, Cabraser & Heimann bases its charges for attorney *469and paralegal services on both the amount of any recovery, net of expenses, and the time at which it is obtained. The first portion of the proposal (“Lieff # 1”) contemplates that for any recovery obtained on or before July 8, 1995, the base fee would be 24% of the first $3 million of the recovery (net of reimbursable expenses); 22% of any incremental recovery from $3 million to $10 million; and 20% of any incremental recovery above $10 million. Lieff Prop at 7. The second portion of the proposal (“Lieff # 2”) provides that in the event of a settlement after July 8, 1995, but before trial, these percentages would be increased by three percentage points. Id. Finally, the third part of the proposal (“Lieff # 3”) states that if the matter were to proceed to trial, the fee would be increased by an additional five percentage points. Id. The Lieff proposal therefore can be arrayed as follows:
[[Image here]]
Because the above percentages are applied only after expenses have been deducted from the total recovery, the actual fee as a percent of total recovery (i.e., before expenses) would be lower than the percentages shown. For example, if the recovery under Lieff # 1 were $3 million and reimbursable expenses were $600,000, the fee would be calculated as 24% of $2.4 million, which is $576,000, or 19.2% of the gross recovery.
Notably, the Lieff proposal provides that paralegal services are not to be considered a reimbursable expense. Accordingly, the plaintiff class will obtain both counsel and paralegal services for the above percentages.
The Lieff firm offers to post a completion bond in the amount of $5 million, and includes as an attachment an example of the type of bond envisioned. Id at 6; id Exh D. The firm does not, however, have malpractice insurance coverage. Id at 6.
B
Milberg Weiss Bershad Hynes & Lerach proposes a fee award of 25% of any recovery, regardless of how large the recovery or when it is obtained. Milberg Prop at 3. Unlike Lieffs bid, however, Milberg’s proposal is not net of expenses. Instead, the fee is calculated as 25% of the total recovery, and expenses are reimbursed out of the remaining 75%. See id at 3^4. Milberg asserts that not assessing the class for all litigation expenses would be “unwise, if not unethical.” Id at 4.
Although asserting that completion bonds are not available to the legal profession, Mil-berg states that it would be willing to obtain such a bond if one could be found. Milberg Prop at 2. Finally, Milberg has submitted copies of the malpractice insurance policies it currently has in force. Lerach Decl Exh B.
C
The bid submitted by Lowey Dannenberg Bemporad & Selinger advances a three-tiered approach, in each case proposing a percentage of the total recovery that will compensate Lowey for attorney services and expenses. If the ease is settled before February 1, 1995, Lowey proposes a total award for attorney services and expenses of 25% of any recovery (“Lowey # 1”). Lowey Prop at 6-7. If the case settles after February 1, 1995, but before trial, Lowey requests 35% of *470the recovery (“Lowey # 2”). Id. Finally, if the matter is resolved after trial or by settlement during trial, Lowey seeks 45% of any recovery (Lowey # 3). Id.
Although expenses incurred in prosecuting the case are included in the above percentages, settlement administration expenses are specifically excluded for any award arising out of a settlement obtained before trial. In the event such a settlement is obtained, administration expenses are to be reimbursed after the fee award is calculated. These excluded settlement administration expenses are capped, however, at $300,000. Id.
Lowey claims that it has a malpractice insurance policy in force, but provides no details. Lowey Prop at 5. Finally, the Low-ey firm states that it is unwilling to post a completion bond. Id.
II
The plaintiff class, were it able to choose among the three bids, would certainly consider qualitative factors in addition to the prices quoted by prospective counsel. Because the objective of the class is to maximize the expected value of any monetary recovery, qualitative factors that can reasonably be expected to affect that recovery are of particular relevance. Given that the court must attempt to choose the firm that would be selected by the class, the court, too, must examine such qualitative factors.
A
Each of the three proposals includes a recitation of the submitting firm’s previous experience with securities class action work by way of descriptions of, and testimonials about, past successes. Additionally, the three firms have provided firm and individual resumes, as well as lists of prior class actions litigated by them. This type of information, while impressive, is of limited utility, as none of it translates into concrete evidence of the firms’ ability to obtain a favorable settlement or judgment in this case. See In re Oracle Securities Litigation, 132 F.R.D. 538, 542 (N.D.Cal.1990) (subjective information about bidding firms “particularly unhelpful”). The usefulness of these recitals is further impaired by the fact that no firm has submitted evidence distinguishing itself from the others.
Nevertheless, the court is convinced that all three firms are fully capable of competently conducting this litigation, each having had extensive securities class action experience. Indeed, the undersigned has had an opportunity to observe each of the firms perform, either in this case or in other cases. Lowey, for example, amply demonstrated its competence before the undersigned in the Oracle litigation. See In re Oracle Securities Litigation, 852 F.Supp. 1437 (N.D.Cal. 1994). Lieff and Milberg, on the other hand, successfully appealed this court’s order dismissing the complaint in the present action. In sum, plaintiffs would be well-represented by any of the three firms.
Lieff and Milberg may, however, have one advantage over Lowey. Because Lieff and Milberg have represented plaintiffs up until this point, they have considerable experience in this case. To the extent this experience could be translated into a benefit to the class, the court would naturally prefer Lieff or ■ Milberg over Lowey. Yet neither Lieff nor Milberg has even attempted to quantify any advantage they may have over Lowey. Without such information, the court can only conclude that all three firms are identical in terms of their professional credentials.
B
The court now turns to one of the qualitative attributes of the submitted bids—the treatment of litigation expenses. Specifically, there are two general principles that are worth noting, both of which start with the premise that litigation expenses are ordinarily paid for by the client. First, an attorney generally has no incentive to minimize litigation expenses unless his fee award is inversely related to such expenses. Second, when an attorney treats a resource devoted to litigation as a reimbursable expense, the attorney has a clear incentive to substitute that resource for those paid for out of the attorney fee, even if it increases the overall cost of the litigation to the client. Because treatment of expenses in the three bids differs, *471each creates different incentives regarding expenses.
The first of the two principles is quite simple. If an attorney risks losing some portion of his fee award for each additional dollar in expenses he incurs, the attorney is sure to minimize expenses. Under both the Lowey and Lieff proposals, the fee award is in fact reduced with each additional dollar spent on litigation expenses. Under the Mil-berg proposal, on the other hand, the fee award is wholly unaffected by the amount of such expenses.
The Lowey proposal provides that counsel are to be awarded a temporally-dependent percentage of the recovery that will compensate them for fees and expenses. Under this approach, each additional dollar spent on expenses results in one less dollar in attorney fees. The Lieff bid provides that counsel are to receive a certain percentage of the total recovery after expenses have been deducted. Under Lieffs proposal, each incremental dollar of expenses simultaneously results, on average, in a twenty-five cent reduction in attorney fees. Because the Milberg proposal provides for a flat 25% of the recovery as the attorney fees in addition to reimbursement for all litigation expenses, fees are divorced from expenses.
Because fees under the Milberg bid are unchecked by the amount of litigation expenses, and because Lieffs approach does provide such a check, Milberg’s expenses would tend to be an indeterminate amount higher than Lieffs. Although the magnitude of the difference is impossible to predict, this difference can nevertheless be considered a qualitative factor differentiating the bids. Accordingly, Lieffs proposal is more advantageous to the interests of the class than is Milberg’s.
The Lowey proposal, however, is even more advantageous to the class than Lieffs, as Lowey would absorb 100 percent of its expenses. But, as discussed infra part III, the fee terms of the Lowey proposal are so much higher than the Lieff bid that they overpower this advantage of the Lowey bid.
The second of the two principles mentioned above concerns the “substitution” of reimbursable work for work which is otherwise not reimbursable. For example, a law firm which charges for paralegal assistance as a reimbursable expense might substitute excessively toward paralegal services and away from secretarial services, as the latter is usually an item of law office overhead. By doing so, the attorney is able to transfer certain costs associated with overhead to the client, even if this substitution would be irrational from the client’s viewpoint.
Lieffs exclusion of paralegal costs from reimbursable expenses gives that firm a substantial qualitative advantage over Milberg. As Lieff notes in its proposal, paralegal costs are often considered expense items even though they should probably be included in overhead. Lieff Prop at 9. Lieff offers to absorb paralegal costs as part of its fee, while Milberg would presumably seek reimbursement for such costs. In this sense, Lieffs expenses clearly will be lower than Milberg’s. As with the effect of expense-minimization incentives, the Lowey bid is unaffected because Lowey proposes to absorb all of its expenses.
It is true that, notwithstanding the above concerns regarding inflated expenses, the court has some supervision over which expenses are ultimately charged. After all, the court can determine after the fact that some expenses for which a firm seeks reimbursement were unreasonably incurred. Yet because judicial scrutiny is highly imperfect in this regard, it is also apparent that the court will generally be less able to determine the cost-minimizing allocation of expenses than will be the law firm incurring those expenses. Put another way, incentives to minimize expenses and to allocate resources properly go much farther toward cost efficiency than can post hoc judicial review.
The court has attempted to make clear that it prefers bids that include both fees and costs in the awards proposed. See Order dated June 30, 1994, at 14; see also Oracle, 132 F.R.D. 538, 539 n. 4 (“The court believes that the risk of misestimating * * * expenses should rest with class counsel.”). When counsel absorb all the costs of litigation, they have an incentive to minimize those costs; and the expected value of that mini*472mization can be shared with the class in the form of a lower-cost bid. Despite this advantage and the court’s expressed preference, several firms have continued to submit bids that include fees only. Compare Oracle, 132 F.R.D. at 539 (only two of four competing firms submitted bids that included expenses, despite court’s direction to do so). In future class counsel selections, courts should continue to press firms seeking authority to represent a class to submit proposals in which expenses are included in the proposed fee award calculations.
C
Another qualitative factor meriting consideration is the willingness of the bidding firms to post a bond or other security toward completion of performance as class counsel. Lieff expressed its willingness to post such a bond and appears to have located a surety willing to write one in the amount of $5 million. See Lieff Prop at 6; id Exh D. The Lowey firm, on the other hand, is unwilling to post a bond; its proposal recites the firm’s belief that such security is “totally unnecessary.” Lowey Prop at 5. Milberg’s response to the court’s interest in completion bonds was that such bonds “do[ ] not appear to be available to the legal profession at this time.” Milberg Prop at 2. Milberg then allows that it would be willing to purchase a completion bond “[i]f such a bond can be found.” Id.
The three firms appear to believe that the need for a completion bond or similar security is obviated by the firms’ interest in maintaining their reputations. See, e.g., Lowey Prop at 5. The court does not doubt that this interest is considerable, but it by no means follows that security is unnecessary. Even prominent, well-respected law firms dissolve, declare bankruptcy or otherwise become unable to carry on their practices. Prosecution of a plaintiff class action, such as this one, in which the resource demands are great and for which payment may never be forthcoming is a particularly risky undertaking.
One potential impediment to the ability of class counsel to complete litigation of a complex securities action is the imposition of sanctions. See, e.g., Garr v. U.S. Healthcare, Inc., 22 F.3d 1274 (3rd Cir.1994). In at least one recent instance, a nationally prominent class action law firm dissolved and its senior name partner was suspended from practice. In the Matter of Richard D Greenfield, Misc. No. 93-205 (DNJ Nov. 16, 1993); see also Henry Gottlieb, Class-Action Lawyers Suspended One Year, NJ LJ, Nov. 22, 1993, at 8. Judge Armstrong recently imposed sanctions against Lieff in another case, see In re Omnitrition International, Inc. Securities Litigation, No. C-92-4133-SBA, MDL No. 965, 1994 WL 476694 (N.D.Cal. Aug. 19, 1994), and the circumstances surrounding these sanctions led to the resignation of one of the firm’s partners. Lieff was also sanctioned in yet another securities class action in this court. In re Keegan Management Co., Securities Litigation, 154 F.R.D. 237 (N.D.Cal.1994). Milberg, too, has been sanctioned, although it has successfully challenged the sanctions in the reported cases in which sanctions were imposed. See Goldman v. G.C. Belden, 580 F.Supp. 1373 (W.D.N.Y.1984), vacated 754 F.2d 1059 (2d Cir.1985); Goldman v. Alhadeff, 131 F.R.D. 188 (W.D.Wash.1990), motion for reconsideration granted, Fischer v. Alhadeff, 131 F.R.D. 585 (W.D.Wash.1990).
Another possible obstacle to completing performance as class counsel is the threat of civil judgments against the law firm. In these increasingly vituperative disputes between the securities plaintiffs’ bar and defendants, their counsel and consultants, personal and venomous attacks are regrettably becoming commonplace. Still worse, these melees are now blossoming into civil litigation. Milberg, for instance, is currently defending against a suit brought by Lexecon, a prominent firm of economists who often testify on behalf of defendants in cases such as the present. See Lexecon, Inc. v. Milberg, Weiss, Bershad, Hynes & Lerach, 845 F.Supp. 1377 (D.Ariz.1993). Lexecon alleges that Milberg intentionally sought to disparage Lexecon’s business reputation by falsely charging that Lexecon had engaged in racketeering and securities fraud. See id. These and other types of civil suits may result in *473adverse judgments and, in any event, can cause significant distractions.
None of the above is meant to impugn the integrity of the plaintiffs’ bar or to question the conduct of the firms currently bidding before this court. Rather, the court offers these examples merely as proof that class counsel could be subjected to sanctions, distractions or business reverses serious enough to make continued prosecution of the instant case a practical or legal impossibility.
Because a bond to complete performance helps to secure the interests of the class in the event of an involuntary default by class counsel, the court believes that a prudent representative of the class would require such a bond as part of counsel’s compensation agreement. Lieffs willingness to investigate the matter and to obtain such a bond gives Lieff a significant qualitative edge over the Milberg firm, whose reaction to the court’s interest in security was lukewarm, and a large advantage over the Lowey firm, which refuses even to entertain the idea of securing a bond.
D
A final qualitative factor to be considered is the malpractice insurance coverage maintained by each of the bidding firms. Milberg Weiss presented evidence that it has substantial malpractice insurance. Lerach Deel Exh B. Lowey asserts that it has malpractice coverage in force, although it declines to reveal the details of that coverage except in camera. In contrast, Lieff has no malpractice insurance, and its proposal merely states that “the individual partners at [Lieff, Cabraser] would be responsible for paying any judgment against the firm.” Lieff Prop at 6.
Although Lieff attempts to present its lack of malpractice coverage as an asset rather than a liability, Lieff Prop at 6, the court is unconvinced. Absent an insurance policy, only the firm’s assets and those of the firm’s partners are available to compensate plaintiffs in the event of professional misfeasance. Because Lieff has not provided any details concerning the combined value of such assets, the court is unable to ascertain the degree of protection afforded thereby. Similarly, because Lowey has not disclosed the coverage limits of its malpractice policy, the court cannot discern the value to the class of this coverage. Only Milberg, with its extensive and comprehensive insurance policy, can definitively be said to provide adequate protection against malpractice.
E
To summarize, the court concludes that: (1) the competing firms are identical in terms of professional credentials; (2) Lieff has a significant advantage over Milberg as a result of its treatment of expenses; (3) Lieffs willingness and ability to post a completion bond gives that firm a considerable advantage over the others; (4) Milberg offers the surest protection against malpractice. After weighing the various qualitative considerations, the court concludes that, on balance, the Lieff bid must be considered qualitatively superior to those submitted by the other two firms.
Ill
While qualitative factors are important considerations in the selection of class counsel, what ultimately matters to the class is the amount of the recovery, net of attorney fees and costs. Accordingly, the prices offered for representation must be a pivotal concern in analyzing proposals to act as class counsel.
In order to perform its analysis, the court first adds the expected values of fees and costs to arrive at an expected total price function for each of the three proposals. These functions will then produce differing values based upon the various time and recovery amount possibilities. In this fashion, the court can compare the three bids.
A
Because of the different ways in which expenses are treated in the three proposals, an assessment of those proposals’ relative values to the class depends on an understanding of the probable level of expenses that will be incurred in the prosecution of this case, relative to any recovery that may be attained.
*474A 1990 study of 404 successful securities and antitrust class actions found that reimbursable expenses expressed as a percentage of gross recovery tend to decrease as the recovery amount increases:
Gross Recovery Expenses as % of Recovery > $50 million 2.1% > $50 miffion(w/o WPPSS)1 0—$50 million 1.2% 0—$20 million 2.6% $5—$10 million 3.4% $3—$5 million 3.9% $2—$3 million 2.7% 1—$2 million 4.5% < $1 million 4.4%
See Stuart J Logan and Beverly C Moore, Jr, Attorney Fee Awards in Common Fund Securities & Antitrust Class Actions, 13 Class Action Reports 4-5 at 526-545. 2
By using estimates of expenses based on these survey results, the court can compare the three bids. For example, when these cost estimates are used to generate the total fee plus expenses for an early settlement, the Lieff bid is seen to be less expensive for the class than the other two bids:3
[[Image here]]
*475[[Image here]]
For settlements obtained within one year, the Lieff bid provides for a 2-4 percent lower total exaction from the class than does the Milberg bid, depending on the amount of the settlement. When compared to the Lowey bid, the Lieff bid is higher for settlements less than $10 million obtained prior to February 1, 1995, (i.e., within Lowey # 1) and lower for settlements greater than $10 million obtained during the same period. On the other hand, for settlements obtained after February 1, 1995, but before July 8, 1995 (and therefore covered by Lieff # 1 and Low-ey # 2), the Lieff proposal is much more favorable for the class, with the total price being lower by 7-10 percent. Lieff # l’s price for settlements obtained in the first six months is probably lower than Lowey # l’s, because Lieffs advantage is 1-2 percent of very large settlement amounts while Lowey offers an incremental 1-2 percent of small settlements. See also infra Part III.B. Moreover, even if Lowey # 1 were less expensive than Lieff # 1, the advantage of Lieff # 1 over Lowey # 2 is of sufficient magnitude to outweigh what would be at best a small Lowey advantage for settlements obtained during the earlier period.
For recoveries obtained through settlement after one year but before the commencement of trial, the Lieff and Milberg proposals are within about one percentage point of each other, while Lowey’s bid is substantially higher than the other two:
[[Image here]]
The Lowey proposal is less favorable to the class than the other two proposals by 5-9 percent. The Lieff bid is higher (by about 1 percent) than the Milberg bid for recoveries less than $5 million, and lower for recoveries greater than $10 million (1.2 percent lower at $20 million, with the difference increasing as the recovery amount increases). For recoveries between $5 and $10 million, the proposals generate very similar total expected prices.
Finally, for a recovery obtained after trial begins the Milberg bid is least expensive:
*476[[Image here]]
For the trial scenario, the Milberg proposal is the best from the class’ standpoint at all recovery levels, with the Lieff proposal second and the Lowey proposal a distant third.
B
Examination of the foregoing tables shows that the Lieff proposal provides for a lower expected total price for the representation than does the Lowey proposal. Lowey’s expected total price is considerably higher than Lieffs except for settlements obtained before February 1, 1995, and any advantage the Lowey proposal might have in the first few months is far outweighed by its disadvantage thereafter.
Comparison of the price terms of the Lieff and Milberg proposals requires more detailed analysis. The Lieff bid provides a total price that is two to four percentage points lower than the Milberg bid for settlements obtained within one year of the date of bid submittal (i.e., before July 8, 1995, and therefore covered by Lieff # 1). On the other hand, the Milberg bid is three to six percentage points lower for recoveries obtained after trial begins (i.e., covered by Lieff # 3). For settlements obtained after the one year' deadline, but before trial (i.e., under Lieff # 2), the proposals’ prices are similar, with the relative price advantage depending on the size of the settlement.
The court first notes that most securities class action suits in which a recovery is ultimately obtained by the class are settled before trial begins. See J Cooper Alexander, Do the Merits Matter: A Study of Settlements in Securities Class Actions, 43 Stan L Rev 497, 498 (1991). While the possibility that the case may proceed to trial should not be entirely ignored, the likelihood of such an occurrence is significantly lower than the likelihood that a settlement will be reached prior to trial. The Milberg proposal’s price advantage relative to Lieff #3 is therefore considerably less important than the relationship between the two proposals with respect to pre-trial settlements.
Because the selection of class counsel will rest primarily on pre-trial settlement prices, it is appropriate to consider the comparison between Lieff #2 and the Milberg bid in greater detail. When the total price to the class is shown in dollars instead of in percentage terms, it becomes apparent that Lieff has an advantage over Milberg:
*477[[Image here]]
The two proposals generate quite similar total prices for recoveries of $10 million and less. But as the settlement amount climbs above $10 million, the Lieff proposal shows an advantage that soon exceeds $100,000 (at a settlement value of about $14 million). The court does not wish to guess at the likely magnitude of any settlement, but as long as there is some reasonable possibility of obtaining a large settlement, Lieff # 2 is less expensive than the Milberg bid.
Recalling that the Lieff proposal presents a clear advantage over the Milberg proposal for settlement within one year, and noting that the Milberg bid is preferable only in the unlikely event that the case proceeds to trial, the court concludes that the Lieff bid’s price provisions are superior to those in the Mil-berg bid.
IV
The court is charged with selecting as class counsel the firm that offers the class the best combination of price and non-price terms. Lieff, Cabraser’s proposal is superior to the other two proposals received by the court, both qualitatively and quantitatively. Accordingly, Lieff, Cabraser & Heimann is hereby DESIGNATED as class counsel for all classes asserting claims against the defendants that have been named in these actions.
Because the court believes that a prudent representative of the class would require counsel to post a completion bond, Lieff, Cabraser is hereby ORDERED to secure a bond in the amount of $5 million substantially conforming to the structure of the sample appended to its proposal.. See Lieff Prop Exh D. The firm shall file the bond with the clerk of court within sixty days of the date of this order.
The clerk is also directed to file the proposals of each firm seeking to represent the class.
IT IS SO ORDERED.
. This entry omits from consideration In re Washington Public Power Supply System Securities Litigation, a case that is aberrational in several respects. The recovery in that case ($757 million) exceeds that in any other case in the survey; more importantly, the expense total of more than $50 million is greater than the next highest expense figure by a factor of nine. The case is also the only one in which expenses exceeded fees. It appears that the case was managed in a manner that would have been inappropriate for a smaller case. See Stuart J Logan and Beverly Moore, Jr, Attorney Fee Awards in Common Fund Securities & Antitrust Class Actions, 13 Class Action Reports 4-5 at 258-264.
. These cost figures appear to include settlement administration costs, and generally include paralegal costs. See Stuart J Logan and Beverly C Moore, Jr, Attorney Fee Awards in Common Fund Securities & Antitrust Class Actions, 13 Class Action Reports 4-5 at 252-53.
. Estimates for recovery amounts on the border between two survey ranges are computed by averaging the two values. These computations assume that settlement administration costs are zero, and therefore underestimate the total cost of the Lowey bids.