Paul M. PRUSKY, et al.
v.
RELIASTAR LIFE INSURANCE COMPANY
No. CIV.A.03 6196.
United States District Court, E.D. Pennsylvania.
February 22, 2007.*704 Kellie A. Allen, Salmon, Ricchezza, Singer & Turchi LLP, Philadelphia, PA, Bruce P. Merenstein, Dennis R. Suplee, Schnader Harrison Segal and Lewis, L.L.P., Philadelphia, PA, David H. Weinstein, Andrea L. Wilson, Weinstein Kitchenoff & Asher LLC, Philadelphia, PA, for Paul M. Prusky, Steven G. Prusky, Plaintiffs.
Ann H. Mathews, Troy A. McKenzie, Edwin G. Moodhe, Joseph P. Moodhe, Lorna G. Schofield, Debevoise & Plimpton, New York, NY, Mathieu Shapiro, Obermayer, Rebmann, Maxwell & Hippel, LLP, Philadelphia, PA, for Reliastar Life Insurance Company, Defendant.
MEMORANDUM
DALZELL, District Judge.
This case addresses the liability of an investment intermediary for its refusal to execute trade requests conformable with the parties' contract. Having recently entered partial summary judgment for plaintiffs as to liability, we convened a hearing to address the uncommon and largely unexplored damages issues this case presents. This Memorandum constitutes our Rule 52(a) findings of fact and conclusions of law.
Findings of Fact[1]
Paul and Steven Prusky, father and son, are investment advisors. Over, the past three decades, they have developed proprietary analysis techniques that allow them to profit from short-term anomalies in mutual fund pricing caused by market psychology and other factors. Based on their daily analysis, they have invested very successfully for their clients and for themselves. Because their strategy focuses on short-term discrepancies between a fund's price[2] as the fund calculates it daily and its value, their approach requires them to make frequent, often daily, exchanges of some or all of their investment capital. It is rare for a significant portion of their money to remain in a single mutual fund for more than a week. Investment strategies, of this sort are known as "market timing." Although mutual fund companies often frown upon market timing, see, e.g., Windsor Sec., Inc. v. Hartford. Life Ins. Co., 986 F.2d 655, 666 & n. 15 (3d Cir.1993), it is a perfectly legal investment strategy.
In addition to managing funds for their investment clients, the Pruskys manage significant funds of their own through the MFI Associates, Ltd. Profit Sharing Plan (the "Plan").[3] Paul and Steven Prusky are *705 the sole trustees of the Plan. In 1998, the Plan bought seven variable life insurance policies from ReliaStar with face values of between $2 million and $10 million each.[4] The policies jointly insured the lives of Paul Prusky and his wife, Susan. They are so-called "second to die" policies, and so provide for payment of a death benefit upon the death of both insureds. These policies permitted the Plan to invest their cash values in the Select*Life Variable Account, a unit investment trust created under the Investment Company Act of 1940. See 15 U.S.C. § 80a-4. The Variable Account was divided into a series of mutual fund sub-accounts, allowing the trustees to select from a portfolio of mutual funds for investment.
When it issued the policies, ReliaStar knew that the Pruskys intended to engage in market timing and would need to make frequent trades in order to execute their strategy. The prospectus for the Select*Life Variable Account allowed only four sub-account transfers a year, so the Plan negotiated an amendment to all of the policies, which was signed in each instance by ReliaStar Vice-President M.C. Peg Sierk. Jt. Ex. 10-H. These amendments have been referred to throughout the litigation as the "Sierk Memos." The Sierk Memos allowed the Plan to make unlimited transfers between the sub-accounts within the Variable Account by phone or fax without any fee and waived any restriction as to the dollar amount of those transfers.
The Plan began making sub-account transfer requests in March, 1998. Often, these requests were made daily. For more than five years, ReliaStar executed the Plan's trades in conformity with the prospectus as the Sierk Memos amended it.[5]
On October 6, 2003, ReliaStar received an inquiry from Pioneer Investment Management, one of the fund companies the Variable Account was invested with, about a series of trades that Pioneer thought might be linked to market timing. When ReliaStar investigated, it found that the Plan had indeed made the trades. On October 8, 2003, Christie Gutknecht, a director at ING, ReliaStar's parent company, sent a letter to Paul Prusky noting that Pioneer was concerned about the transactions and had a no-market-timing policy. Jt. Ex. 10-V. The letter informed Prusky that, effective immediately, the Plan would only be permitted to make trades in Pioneer funds by U.S. mail. The October 8 letter went on to warn that any further market timing transactions would result in the placement of a similar restriction on all trading under the policies.
The next day, Steven Prusky sent a response to Gutknecht, with a copy to the Pruskys' attorney, asserting that ReliaStar had violated the terms of the insurance contracts and threatening to hold ReliaStar liable for any losses as a result of refused transaction requests. Jt. Ex. 10-W. Notwithstanding these threats, on November 5, 2003 after another inquiry from the Fidelity group of funds Gutknecht informed Prusky that ReliaStar *706 would no longer accept any trades from the Plan by phone or fax but would require that all trades be made by U.S. mail. Jt. Ex. 10-X.
The following day, Steven Prusky again wrote to Gutknecht, with an indicated copy to the Pruskys' lawyer:
In response to your letter of this week regarding restrictions on transfers concerning the Fidelity Advisor High Income fund, I hereby strenuously protest. Your actions are in violation of your contracts with us.
In light of this breach, we will follow these procedures: we will continue to send you two faxes, one noting our "desired" exchanges, representing what our transfers would be if not restricted by you, the other fax noting our "actual" exchanges, representing exchanges that meet your restrictions.
By this method we will track our damages and hold ING Reliastar responsible for them. The actual exchanges are our best attempt at mitigating those damages. If you believe there is a better course of mitigation, please inform us immediately so we can consider it.
Jt. Ex. 10-Y.
Within a week, the Pruskys sued. The Plan has continued to send ReliaStar daily sub-account transfer requests, which ReliaStar has not executed. Jt. Ex. 8. The Plan has not, however, sent ReliaStar requests for the "actual" exchanges mentioned in Steven Prusky's letter. Instead, the Plan transferred the balance of all seven policies well over $7 million into ReliaStar's money market sub-account "in order to mitigate any damages and minimize risk." Jt. Ex. 12, ¶ 5.
On January 5, 2007, we held that ReliaStar's refusal to execute the Plan's trades was a breach of their contract. We reserved the question of the proper measure of damages, and on February 15, 2007 held an evidentiary hearing to determine what the Plan suffered as a result of ReliaStar's breach.
On January 12, 2007, we clarified that ReliaStar should begin performing under the terms of the contract immediately. Although the trades of January 12, 2007 through January 18, 2007 were rejected because they requested trades into or out of funds that were no longer available to the Plan,[6] since January 19, 2007 ReliaStar has been executing the Plan's requested trades.
Possible Measures of Damages
The parties have offered up a number of possible measures of the damages the Plan suffered. On October 9, 2003, the day ReliaStar imposed the first restrictions on the Plan's trading, the combined value of the seven policies was $7,240,882.41. It is undisputed that, as of January 31, 2007, the actual cash value of the policies was $7,045,349.01.[7] The parties have stipulated that, had all of the proposed trades been executed, the cash value of the policies on January 31, 2007 would have been $8,064,642.29, a difference of $1,019,293.28. *707 Pl.Ex. 69. If the trades had all been processed but, in keeping with ReliaStar's policies, trades into or out of closed funds had been rejected, the total cash value would have been $8,705,048.47.[8] Pl.Ex. 68.
ReliaStar, which among other claims asserts that the Pruskys failed to mitigate damages, provides some additional benchmarks against which to measure the return on the Plan's investments.[9] These benchmarks are all predicated on the assumption that ReliaStar's breach forced the Plan to adopt some sort of buy-and-hold strategy rather than the daily trading strategy the Plan had used. Def. Ex. 14.
The first benchmark ReliaStar provides is the result of simply keeping the funds distributed as they were on November 5, 2003 when ReliaStar halted all of the Plan's telecommunicated trading. That approach results in a balance of $9,404,000, over $1.3 million more than the Plan's proposed trades.
ReliaStar's next proposal accounts for the fact that, on average during the breach period, 44% of the value of the policies remained invested in the money market account.[10] As ReliaStar's expert, Dr. Vincent Warther, testified, an investment strategy's risk is primarily driven by its degree of exposure to equities and other securities whose values fluctuate with the markets. In order more closely to replicate the risk profile of the Pruskys' plan, Dr. Warther calculated the results of leaving 44% of the value in the money market fund and placing the other 56% in securities the portfolio held on November 5, 2003. This resulted in a cash value of $8,364,000 or $264,000 more than the Pruskys' desired trades would have yielded.
Finally, in an attempt to replicate the Pruskys' risk profile even more closely, Dr. Warther presented a portfolio that held the funds in which the Plan invested in the same proportions and during the same time periods as the Pruskys' desired trades.[11] This approach results in a cash value of $7,952,000, significantly better than the actual money market return, but $148,000 less than the Pruskys' desired trades would have netted.
Conclusions of Law
Amount of Damages
We begin our analysis of the proper measure of damages with the guidance of our Court of Appeals in the Pruskys' nearly identical lawsuit against Aetna. "Damages can be calculated based on the never-executed faxed instructions for transfers between sub-accounts, which instructions the Plaintiffs (and presumably the Insurance Companies) have retained." Prusky v. Aetna Life Ins. & Annuity Co., *708 2006 WL 952320 at *1 (3d Cir. April 13, 2006) (non-precedential).[12] Even had our Court of Appeals not specifically endorsed this approach, it would be the obvious place to begin our analysis. We will refer to this amount as the "Desired Trade" value.
While the Desired Trade value is a good polestar to guide our analysis, we cannot simply accept that value without further inquiry. In addition to ReliaStar's claims about mitigation, which we address shortly, the Desired Trade value potentially fails to account for two important factors in assessing the Plan's actual damages.
First, it is by no means certain that, even had ReliaStar accepted each of these trades, the fund companies would have honored them. The fund companies had already raised concerns about some of the Plan's earlier trades. See, e.g., Def. Ex. 1 (e-mail exchange between Fidelity and ING expressing concern about the Plan's trading activity). Had the trading continued, some or all of the funds could have refused to honor trades from the Plan. As we found in our January 5, 2007, ruling, ReliaStar was entitled to enforce restrictions on the Plan's trading that the funds themselves imposed.
ReliaStar, however, has introduced no evidence that would allow us to find that some or all of the Plan's desired trades would have been rejected. Because of this lack of evidence, any decision to exclude certain trades on this basis would be pure speculation. While we expect that, had all of the desired trades been carried out, the Pruskys' would have been subject to some restriction by some of the funds, the record before us simply does not permit modification of the damage award on that basis.
Second, because the Pruskys knew that none of the desired trades would be executed, they were protected from any serious downside risk associated with their investment strategy. As an illustration of this point, imagine that the result of executing the Pruskys' desired trades was $6 million instead of slightly more that $8 million. Under that scenario, although the Pruskys would not be entitled to compensatory damages, they would still be entitled to keep the $7,045,349.01 that was the result of their actual investment in the money market fund. Thus, unlike where they were investing real money, the Pruskys could do no worse than realize the result of their safe money market investment. If their strategy worked, they could sue for damages; if it did not, they could keep their money market return.
Given this insulation from most risk of loss, a sophisticated investor would likely undertake a more aggressive read, riskier investment approach than would be warranted without any such protection. In order to prove their damages, therefore, it was incumbent on the Pruskys to demonstrate that their so-called desired trades were the same trades they would have made if they had really thought that the trades would be executed.
Steven Prusky testified, however, that their daily analysis was unchanged during the breach period. He further reported that, based on that analysis, he and his father made trades closely analogous to the desired trades in other accounts they managed. Defendants did not challenge this testimony. We are satisfied, therefore, that the Pruskys would have made the trades they faxed to ReliaStar during the breach period even had they expected *709 that ReliaStar would actually execute them.
Because the Pruskys have addressed the two concerns that would make us hesitant to adopt the Desired Trade value as our starting point, we find that the Pruskys suffered damages in the amount of $1,019,293.28 as a proximate result of ReliaStar's breach.[13]
Mitigation
That finding, however, does not end our inquiry. The Pruskys are not entitled to recover damages that they "could have avoided without undue risk, burden or humiliation." Rest. (Second) of Contracts, § 350(1); see also Bafile v. Borough of Muncy, 527 Pa. 25, 588 A.2d 462, 464 (1991).
The Pruskys claim that their decision to place the entire $7 million cash value of the policies in a money market fund made mitigation sense because it guarded against the possible loss of principal if the money was invested elsewhere. In determining the appropriateness of the Pruskys' mitigation, we examine whether their conduct was reasonable given all the facts and circumstances at the time they made the mitigation decision. Toyota Indus. Trucks U.S.A., Inc. v. Citizens Nat. Bank of Evans City, 611 F.2d 465, 471 (3d Cir.1979). If the Pruskys' action was reasonable, we may not penalize them because we now know with 20/20 hindsight that another reasonable choice would have reduced damages further. In re Kellett Aircraft Corp., 186 F.2d 197, 198 (3d Cir.1951).
We find that the decision to place more than $7 million in a money market fund for over three years was not reasonable. The Pruskys knew, at the time they made their mitigation decision, that this litigation would extend for months or even years.[14] They also had the opportunity to adjust their mitigation strategy based on the actual progress of the litigation as it ground on.[15] The Pruskys had previously invested those funds in vehicles with substantial exposure to the equity markets and had produced double-digit returns. Placing such a large sum of money in an investment vehicle that has often under-performed inflation (much less the general markets) is not a reasonable mitigation strategy under these, circumstances.[16]
Where a breach of contract prevents an investor from pursuing his desired investment strategy, a reasonable mitigation investment must, to the degree possible, reflect a similar risk/reward profile to that of the desired strategy. See Teachers Ins. & Annuity Ass'n of Am. v. *710 Ormesa Geothermal, 791 F. Supp. 401, 416 (S.D.N.Y.1991) (holding that reasonable mitigation required an investment that "should have investment characteristics as close as possible to the original investment"). The Restatement does not limit required mitigation to alternate plans without any risk; rather, it says "undue risk." We can think of no better measure of what risk is "due" than the very trades the Pruskys wanted to make.
The Pruskys also place great weight on the final sentence of Steven Prusky's November 6, 2003 letter: "If you believe there is a better course of mitigation, please inform us immediately so we can consider it." Jt. Ex. 10-Y. This was surely a ploy[17] that lawyers (and judges) must admire for its cleverness. But the question for us is whether it really works.
We first note that the actual mitigation the Pruskys undertook was very different from what they proposed in the November 6 letter, which contemplated "`actual' exchanges". Thus, even if ReliaStar had deemed the Pruskys' proposed course acceptable, it could still raise the issue of failure to mitigate because the Pruskys, did not in fact execute the mitigation strategy they themselves proposed.
More importantly, however, we are aware of no legal authority that allows a plaintiff to shift his duty of mitigation[18] to the defendant through this stratagem. Honoring such a facile delegation would turn the duty of mitigation on its head, reducing it to merely a duty to comply with the reasonable mitigation requests of the defendant. We are unwilling to so easily pull down a pillar of contract law. The last sentence of Steven Prusky's November 6, 2003 letter, clever a ploy though it was, has no legal effect.
The Pruskys also argue that, because their investment expertise is entirely focused on short-time-horizon investing, they should not be required to mitigate by adopting a buy-and-hold investment strategy. There is no question, however, that any mitigation the Pruskys undertook would have to take some form of a buy-and-hold strategy. ReliaStar's breach required that. Even the decision to place the policy value in the money market fund represented a buy-and-hold strategy. While we understand that the Pruskys have no particular skills in buy-and-hold investing, they are certainly capable of identifying and adopting some reasonable allocation of funds for a longer time horizon. Thus, while we will not impute to them the skills of an expert buy-and-hold investor like Warren Buffett, we do require them to find some reasonable means of investing their money, in spite of ReliaStar's breach, during the pendency of this litigation.
Having now determined that the Pruskys did not fulfill their duty of mitigation, we must take one more step. A failure to mitigate does not act as a complete bar to recovery, but instead reduces recovery by the amount of loss that could have been avoided by reasonable mitigation. State Pub. Sch. Bldg. Auth. v. WM. Anderson Co., 49 Pa.Cmwlth. 420, 410 A.2d 1329, 1331 (1980). Our holding that the Pruskys' chosen strategy was unreasonable *711 allows us to consider other possible mitigation strategies, but the burden of showing what damages could have been avoided lies with the breaching party. Id. In attempting to carry that burden, Dr. Warther proposed and analyzed the three schemes we cataloged above. We will address each of them in turn.
ReliaStar's first proposed mitigation scheme was simply to leave the money allocated as it was after November 5, 2003, the last day on which the Plan's trades were executed. This produces investment results that are more than $1.3 million better than the Pruskys' proposed trades. While this has some superficial appeal, the allocation of funds on that date was designed for a very short time horizon, a few days at most. To adopt that allocation for more than three years, therefore, would be arbitrary at best. While it happens that this allocation significantly outperformed the Plan's desired trades, the results of leaving the funds in the November 5 allocation could just as easily have been disastrous. Furthermore, because the money was, on that date, allocated 90% to equities and 10% to high yield bonds, see Jt. Ex. 8 (trades for November 5, 2003), the November 5 investment portfolio was much riskier than the Plan's desired trades. Since we have held that the mitigation strategy should match, as closely as possible, the risk profile of the strategy the breach prevented, we cannot adopt the November 5, 2003 allocation as a reasonable measure of the damages that could have been avoided through mitigation.
ReliaStar's second proposal reflects the fact that about 44% of the time the Plan had its money in the money market account. Thus, in an attempt to more accurately match the risk profile of the Prusky's desired trades, Dr. Warther left 56% of the money allocated as it was on November 5, 2003 and moved the rest to the money market fund. This allocation still outperformed the desired trades, but not as handsomely. While this allocation more closely mirrors the desired trades' risk, it still cannot escape the fact that the allocation of funds on November 5 is, from a long-term investing standpoint, arbitrary. Because the Plan only intended to keep its money in those funds for a matter of days, there is without the benefit of hindsight no reason to believe that those particular funds would represent a sensible allocation for a buy-and-hold strategy when the music stopped on November 5, 2003. In fact, all of the equity investments on that date are held in small cap and international funds, two of the most volatile sectors of the equities market. This allocation happened to perform remarkably well over the breach period, largely due to the extraordinary performance of some international funds, see Def. Ex. 12, but we cannot use such hindsight in evaluating the reasonableness of a particular mitigation strategy.
ReliaStar's final proposal attempts to match the desired trades' risk profile even more closely. Without trading daily, it allocates the money across all of the available funds in the same proportion, and during about the same times, as the desired trades. This proposal underperforms the desired trades themselves by $148,000, demonstrating that the Pruskys' short-time-horizon analysis does indeed add value. But it is more reasonable than the other two proposals as a measure of mitigation because it does not rely on the arbitrary allocation of funds as they happened to be on November 5, 2003. Instead, it mimics the Plan's exposure to particular funds over time while still complying with ReliaStar's restrictions. Because it closely approximates the risk profile of the desired trades and because it relies on information that Pruskys had at their disposal namely, the desired trades themselves it represents a reasonable *712 mitigation strategy that was readily available to the Pruskys. As this strategy outperformed the Pruskys' unreasonable money market mitigation strategy by $912,000, see Def. Ex. 14, we will reduce the Pruskys' recovery by that amount.
In sum, we find that ReliaStar's breach was the legal cause of $1,019,293.28 in harm to the Pruskys, but that their failure adequately to mitigate their damages reduces their recovery by $912,000. We will therefore enter judgment for the Pruskys in the amount of $107,293.28.
JUDGMENT
AND NOW, this 22nd day of February, 2007, in accordance with the accompanying findings of fact and conclusions of law, JUDGMENT IS ENTERED in favor of plaintiffs Paul M. Prusky & Steven G. Prusky and against defendant ReliaStar Life Insurance Company in the amount of $107,293.28.
ORDER
AND NOW, this 22nd day of February, 2007, the Court having this day entered judgment in this matter, it is hereby ORDERED that the Clerk of Court shall CLOSE this matter statistically.
NOTES
[1] Although the facts presented here are much the same as those in our earlier opinion, see Prusky v. Reliastar Life Ins. Co., 474 F. Supp. 2d 695, 2007 WL 43641, (E.D.Pa. Jan.05, 2007) ("Prusky I") because we are now adopting them as our findings of fact, we rehearse them again in full. We heard testimony and admitted voluminous exhibits at the February 15, 2007 damages hearing.
[2] This is generally referred to as a fund's Net Asset Value (NAV) or Accumulation Unit Value (AUV).
[3] At the time the policies in this case were purchased, the Plan was known as the Windsor Securities, Inc. Profit Sharing Plan.
[4] According to our Court of Appeals, the total face value of the policies is about $42 million. Prusky v. Reliastar Life Ins. Co., 445 F.3d 695, 696 (3d Cir.2006).
[5] The Sierk Memos also allowed the Pruskys to use a practice commonly known as late trading, which allowed them to trade after the daily NAV calculus for the fund. This strategy violates the Securities and Exchange Commission's forward pricing rule, 17 C.F.R. § 270.22c-1(a). The SEC and other authorities have recently cracked down on late trading in the mutual fund industry, and in November, 2002, ReliaStar notified the Pruskys that federal law precluded it from honoring their late trading requests. The Pruskys no longer seek to enforce the late trading provisions of their agreement with ReliaStar through this action.
[6] On April 28, 2006, several sub-accounts available in the Variable Account were closed to new investment and several new sub-accounts were added. See Jt. Ex. 7-C. Although the. Plan was notified of this change, it did not amend its orders accordingly. The parties have, however, stipulated that plaintiffs were "generally aware" of those changes. ReliaStar's policy is that if any aspect of a trade request cannot be executed, it will be disregarded in its entirety. Thus, beginning April 29, 2006, any transfer into or out of one of the closed funds would have caused the entire day's order to be disregarded.
[7] The value has declined because the cost of insurance that is, the premium payments on the policies themselves is automatically deducted monthly. The total cost of insurance during the breach period was about $915,000. All of the valuations in this section are net of those insurance costs,
[8] This approach has the effect of cancelling many of the trades in May of 2006 and all of the trades between May 31, 2006 and January 19, 2007 because each of them involved at least one closed fund.
[9] The values ReliaStar presented are as of January 12, 2007 rather than January 31, 2007, so they are not precisely comparable to the plaintiffs' numbers. In addition, ReliaStar's examples are rounded to the nearest $1000. The actual cash value of the accounts on January 12, 2007 was $7,040,000. The cash value of the Pruskys' desired trades on that date was $8,100,000. Def. Ex. 14.
[10] The Pruskys' strategy called for using the money market fund as a repository when there were no advantageous investments to be made on a particular day. At times, therefore, the entire value of the fund would rest briefly in the money market.
[11] Dr. Warther did not provide details on the precise algorithm used to construct this portfolio. Because plaintiffs raised no objection to his summary, however, we will accept his assertion that it accurately reflects the returns of the strategy he described and that such a strategy could have been implemented in keeping with ReliaStar's trading restrictions on the Plan.
[12] As we noted above, it is undisputed that this calculation produces a damage award in this case of $1,019,293.28.
[13] The Pruskys' other proposed valuation, representing the result of ReliaStar's policy of rejecting trades into or out of closed funds, see Pl.Ex. 68, is not worthy of consideration. The plaintiffs are not entitled to a windfall because they failed to adjust their trading form to account for the changed menu of funds after April 28, 2006. Further, had ReliaStar in fact been processing the trades, the Plan would no doubt have corrected its form in short order. There is, therefore, no reason to adopt any valuation other than the stipulated value of the account had the desired trades been executed during the breach period.
[14] Steven Prusky testified that he hoped that this dispute could be resolved within a matter of weeks. Because he had the assistance of able and seasoned counsel, however, we find that his hope was nothing more than a pipe dream. The realities of federal court litigation precluded such a snap resolution, and the Pruskys knew it.
[15] As grind on it most assuredly did, as we rehearsed in our Jan. 5, 2007 decision. See Prusky I, 2007 WL 43641 at *2.
[16] Indeed, Steven Prusky's November 6, 2003 letter acknowledges the necessity of a more robust mitigation strategy. See Jt. Ex. 10-Y. That letter contemplates that the Plan would request "actual" exchanges as part of its mitigation strategy. No such actual trades ever materialized.
[17] We use the word in Stephen Potter's coinage, which is now embedded in our language. See IX The Oxford English Dictionary 1066, col. 1, def. 2 (2d ed.1989) (citing, as first usage, Potter's Lifemanship 15 (1950)). See also Robert Hamer, School for. Scoundrels or How to Win Without Actually Cheating! (1960) (film based on Potter's books)
[18] Technically, mitigation is not really a "duty" since failure to mitigate does not result in liability. See Rest. (Second) of Contracts, § 350, cmt. b. Mitigation is perhaps better viewed as a condition precedent to recovery. Nevertheless, because the phrase "duty of mitigation" is so common in the jurisprudence, we use it here.