John ZERVAS, et al., Plaintiffs-Appellees, v. D.L. FAULKNER, et al., Defendants, Spencer Blain and Jane Nix, Defendants-Appellants, 861 F.2d 823

GARWOOD, Circuit Judge: In this civil suit based on the Racketeer Influenced and Corrupt Organizations Act (RICO) and related pendant state law fraud theories, defendants-appellants Spencer Blain (Blain) and Jane Nix (Nix) appeal the judgment on the jury’s verdict against them and in favor of plaintiffs-appellees John Zervas (Zervas) and certain entities controlled by him. We reverse.

Facts and Proceedings Below The crux of Zervas’ claim is that in connection with his September 1982 purchase and subsequent development of a 14.55-acre tract of land near Dallas, Texas, he was a victim of a complex, wide-ranging, and ongoing conspiracy centered around Empire Savings and Loan Association (Empire) of Mesquite, Texas, just outside of Dallas, and involving Blain, Nix, and many others.

To place the contentions of the parties in an understandable context, we outline the background facts and circumstances surrounding Zervas’ purchase and development, the alleged conspiracy, and the roles therein of Blain and Nix, as shown by the evidence at trial. In March 1982, Blain left his job as president of First Federal Savings and Loan Association (First Federal) in Austin, Texas to become president of Empire.

As Empire’s president, Blain was to receive an annual salary of $30,000 plus twenty-five percent of the profits generated by Statewide Service Corporation (Statewide), a wholly-owned subsidiary of Empire that engaged in real estate transactions. Shortly after his arrival in the Dallas area, Blain was given a Rolex watch by D.L. Faulkner (Faulkner), a local developer whom Blain had come to know while working at First Federal and who was instrumental in Blain’s decision to move from Austin to the Dallas area.

Two months after his arrival at Empire, Blain learned that a controlling block of Empire’s stock was for sale. After obtaining a loan of some $800,000 from Faulkner and Faulkner’s business partner James L. Toler (Toler), Blain purchased this block of stock and became Empire’s majority shareholder. Nix owned and ran nearby Dallas Title Company of Rowlette (Dallas Title), which frequently did large volumes of business with Empire, Statewide, and Faulkner, among others. Nix served on the Empire board (though not on its loan committee) for approximately six months until she resigned in October 1982.

In the summer of 1982, Toler informed Blain that he had an option to buy some land adjacent to other land that he owned in the Faulkner Creek area of the so-called 1-30 corridor located in eastern Dallas County, Texas, and that he believed that this land could be sold at a substantial gain. Based on this information, Blain caused Statewide to borrow money from a local bank in order to purchase this land. This land was subsequently sold to certain builders and developers, netting Statewide an $800,000 gain.

The Faulkner Creek transaction was not the only land transaction in which Blain was involved with Faulkner and Toler. In August 1982, Faulkner and Toler sold to Blain personally a sixty-five-acre tract of land in Rowlette, Texas for between $800,-000 and $1,100,000. This property was subsequently appraised by appraiser Paul Tannehill (Tannehill) at $5,800,000. Based on this appraisal, Blain was able to obtain a loan of $2,250,000, which he used both to pay for the land and to pay off the $800,-000 loan that Faulkner and Toler had previously made to him. The closing for this transaction took place at Dallas Title.

In December 1982, Blain agreed to sell this tract to developer Cliff Sinclair (Sinclair) for some $16,000,000 in cash and real estate. The closing for this transaction, which occurred on February 17, 1983, also took place at Dallas Title. Nix personally handled that closing. Shortly before that closing took place, Empire loaned some $32,000,000 to certain investors buying land in which Sinclair had an interest. From March to October 1982, Empire, under Blain’s direction, initiated an aggressive campaign to increase its assets by soliciting “brokered” deposits from out of state.

These deposits vastly multiplied Empire’s size and enabled it to expand its construction loan business substantially. In soliciting these deposits, it is not disputed that interstate wire transmissions were made and that the mails were used. In July 1982, Zervas, a real estate developer with a somewhat checkered past, who was living in Dallas and was also involved in a condominium project in Galveston, learned from his friend Mike Kammerling (Kammerling), a real estate broker with Merrill Lynch in Dallas, that some interesting real estate activity was occurring in the 1-30 corridor.

At Kammerling’s suggestion, Zervas decided to attend an informal Saturday morning breakfast gathering at a local restaurant that was frequently attended by many of those involved in developing the 1-30 corridor. On this occasion, Paul O’Dell (O’Dell), a Merrill Lynch colleague of Kammerling who worked with many of the developers in this area, introduced Zervas to Faulkner and Toler. Although Zervas indicated- that he might be interested in building some condominium units in that area, Toler informed him that there was no land available for purchase at that time.

The following week, however, Zervas received a phone call from O’Dell, who informed him that a tract owned by Faulkner, Toler, and others had recently become available for purchase. The next Saturday — August 7, 1982 — Zervas met with Faulkner and Toler and was shown a four- to five-acre tract in an area called Faulkner Point. After seeing this land, Zervas signed a contract to purchase it at six dollars a square foot and gave a $1,000 earnest money check to Faulkner and To-ler.

On August 9, 1982, Zervas went to Empire, which he had been told was heavily involved in lending in the 1-30 corridor, and submitted certain financial information to one of its employees. Subsequently, Zer-vas received another phone call from O’Dell, who informed Zervas that ten acres of land adjacent to the land that Zervas had under contract had just become available due to a previous potential buyer’s having backed out of a contemplated purchase.

Zervas testified that O’Dell told him the value of this land, when added to that Zervas already had under contract, “was somewhere around seven million dollars.” Zervas then called a friend of his named Ralph Falgiamo (Falgiamo), who owned a trucking company in New Jersey, and asked whether he was interested in becoming Zervas’s partner in building some condominium units on this land.

Shortly thereafter, Falgiamo flew to the Dallas area and, on August 25, 1982, after he and Zervas viewed the property that was available for sale, they signed four contracts, each covering a separate parcel and providing for $1,000 earnest money, in Toler’s office to purchase this additional land. When this additional land was added to the land that Zervas had previously contracted to buy, the total acreage amounted to 14.55 acres, for which the sale price was approximately $5,600,000. That same day, Falgia-mo and Zervas went to a lawyer and had a partnership agreement drawn up.

After returning to New Jersey, however, Falgiamo changed his mind about the partnership and informed Zervas that he did not intend to go through with the purchase. Zervas then called O’Dell to inform him that because Falgiamo was withdrawing, Zervas was only going to be able to purchase the tract that he had originally planned to purchase. At O’Dell’s urging, however, Zervas agreed to talk with Faulkner and Toler about going through with the agreement to purchase all 14.55 acres himself, O’Dell again stating it was worth seven million dollars.

At the meeting with Faulkner and Toler, Zervas stated that he would only be able to purchase the 14.55 acres if the purchase price was lowered from the previously agreed upon price of some $5,600,000 to about $4,000,000, which would be approximately $6.31 a square foot. Although no agreement was reached, that evening, Toler subsequently called Zervas and informed Zervas that he and Faulkner would accept the lower amount offered by Zervas. Although Zervas had now committed himself, at least informally, to purchase the 14.55 acres alone, he continued to have doubts about certain aspects of the transaction.

In particular, Zervas informed Faulkner and Toler that he doubted that he would be able to arrange the necessary financing by himself. Faulkner informed Zervas, however, that financing the transaction would not be a problem and that he would take care of it for Zervas. Zervas agreed to let Faulkner act as his “mortgage broker”; however, Zervas never discussed with Faulkner the terms or conditions that Zervas would accept on a loan to buy the property. About two weeks later, Zervas was informed by Merrill Lynch that the transaction to purchase the land was to close at Dallas Title on September 14,1982.

On the closing date, still without any knowledge of the terms of any loan, Zervas went to Dallas Title approximately two hours early. He asked to speak with Nix, whom he had not previously met, and was introduced to her. Nix, who was then preparing to conduct the closing, informed Zervas that Paul Jensen (Jensen), the president of Lancaster Savings and Loan Association (Lancaster), had attempted to reach Zervas at Dallas Title and that Zervas should return his call. Zervas did return the call from a telephone located in Nix’s office.

During the phone call, Jensen informed Zervas that Lancaster would be providing the loan for the transaction and described the terms and conditions under which Lancaster would furnish the loan. The loan would be in the amount of $5,200,000, interest at an annual rate of eighteen percent payable quarterly, ten points payable at closing, and a term of nine months. After speaking with Jensen, Zervas informed Nix of the loan terms and land price and asked her opinion.

According to Zervas, Nix responded, “I feel you will do well” because he was buying the land worth over seven million dollars, the condominium units being constructed in the area were selling well, and Faulkner was going to be assisting him. After hearing Nix’s comments, Zervas left her office. Prior to the actual closing, Zervas again discussed the project with Kammerling and O’Dell. He also had a face-to-face discussion with Jensen when Jensen arrived at Dallas Title. In the late afternoon, the transaction finally closed.

Among the documents at the closing was an appraisal of the property, dated September 7, 1982, showing a value of $7,162,000, that was purportedly signed by Tannehill. Zervas had not ordered or previously seen any appraisal on any of the property. With the closing papers, there was also a so-called builder’s commitment letter from Empire (not signed by Blain) to Lancaster, which appeared to commit Empire to make $1,800,000 in permanent loans once the condominium units were built, provided certain broad conditions were met.

After the appropriate documents had been signed, Nix, who was handling the closing for Dallas Title, disbursed the proceeds of the $5,200,-000 loan. The purchase price for the 14.55 acres was $3,990,000, which was paid to Faulkner and Toler out of the $5,200,000 loan proceeds funded by Lancaster and disbursed at the closing.

The balance of the loan proceeds were disbursed as follows: $321,-750 cash to Zervas; $10,000 to Zervas’ Galveston entity “Campeche Development”; $234,000 to a certificate of deposit in Zer-vas’ name at Lancaster to serve as security for the first quarterly interest payment due on the loan three months hence; $520,000 in loan fees, of which $104,000 was paid to Lancaster and $416,000 to First Financial Mortgage Company; $100,000 brokerage fee to Merrill Lynch; and $24,250 in title policy fees and other miscellaneous items.

Zervas paid no cash for the purchase, and at the closing was in fact advanced more than $330,000 in cash, and an additional $234,000 to pay future loan interest, over and above the total of the purchase price, loan fees, and other acquisition and closing costs. Zervas was, however, personally liable on the full $5,200,000 loan. On October 14, 1982, Zervas received the first of several construction loans from Empire during a closing at Dallas Title. These loans eventually added up to approximately $15,000,000.

Zervas created a limited partnership to own and develop the land, and gave Empire's subsidiary, Stateside, a twenty-five percent interest therein as a limited partner; the construction loans were made to the limited partnership but were guaranteed by Zervas. Zervas personally was paid some $375,000 out of the proceeds of all the construction loans, and his stepfather was similarly paid some $75,-000, mostly for development-related services.

While work proceeded on the first of Zervas’s projects, which was named Canterbury Landing, Zervas received a second construction loan from Empire for a project named Pentridge. This second loan closed in January 1983. Between the time that the first and the second of these construction loans closed, a report that severely criticized numerous aspects of Empire’s operations was prepared by an examiner from the Federal Home Loan Bank Board.

Shortly after the September 14 land purchase, Zervas had commissioned Merrill Lynch to do a feasibility study respecting development of the 14.55 acres, and it was completed on November 3, 1982. This study stated, among other things, that since 1980, when Faulkner commenced condominium construction in the area, many individual investors or developers had begun purchasing land there “to begin condominium development. At the present time literally thousands of condominiums are scheduled for construction in the area.” Zervas generally based his plans for the remaining development of the 14.55 acres on this report.

Zervas also testified that in August and September 1982, there was “a very limited amount of construction” in the area; however, by January 1983, he noted “there was about twenty to twenty-five percent more construction going on,” and he then “had some concern” about “overbuilding” but “not to the point that there would be a problem.” In January 1983, Zervas sold a one-half-acre part of the Pen-tridge project to his stepfather; the purchase was financed by Empire, with a guarantee by Zervas, and Zervas received approximately $287,000 cash and Stateside about $100,000.

Previously, in October 1982, Zervas had sold a half acre from the 14.55-acre tract to a third party for $250,-000 cash, or about $11.48 a square foot. During late 1982 and early 1983, Blain and Nix participated in certain unrelated transactions that generated for each of them substantial sums of money. In a transaction that took place in November 1982, Blain received $1,400,000 from Sinclair for performing consulting services that he did not in fact perform.

In another transaction that occurred in January 1983, Nix managed to gain $600,000 for herself and $200,000 for her children by buying and selling a certain piece of land on the same day. In another January 1983 transaction which closed at Dallas Title with the purchase money being at least indirectly financed by Empire, Nix disbursed the funds, some of which were ultimately received by Blain. In yet another transaction that took place in April 1983, Blain, together with certain friends and family members, made approximately $3,800,000.

Faulkner gave Nix a “logo pin,” her daughter a gold Rolex watch, and her sister several thousand dollars. On May 11, 1983, Blain received a report that Empire had commissioned that discussed the absorption of condominium units in the 1-30 corridor. In this report, there was said to be a fifty-five month oversupply of condominium units in that area. After reading the report, Blain had one copy of it sent to Lynn Bowman, the Texas Savings and Loan Commissioner, and another copy sent to the Federal National Mortgage Association (FNMA), which up to this time had approved condominium mortgages in the 1-30 area.

Shortly thereafter, Blain met with Bowman in Austin and agreed that Empire would make no more land loans in the 1-30 corridor.

By May 1983, Zervas testified that he had become concerned about the number of condominiums being built in the area and “no longer” felt as he had “in the summer of 1982 and early 1983 ... that there would be controlled growth and ... profitable transactions from construction of condominiums.” However, he received information that an investor was interested in buying his Pentridge project but “wouldn’t buy it unless the entire project was completed because of the configuration and the way the project was laid out.” Zervas was then of the opinion, apparently because of his concern about oversupply of condominiums, that selling a whole project to an investor, rather than individual units to users, would be his best option.

He also realized that “it would be almost impossible to” sell Pentridge without completing its fourth and final phase. He relayed all this information to Blain, whom, Zervas testified, “agreed that if this was the situation that we better hurry and do it and go ahead and build it and sell it.” Accordingly, on May 25, 1983, Empire made another construction loan to Zervas, this for phase four of Pentridge. No further loans by Empire were made. When they discussed this loan, Blain did not disclose to Zervas the above-mentioned report Blain had received on May 11 concerning condominium oversupply.

On June 27, 1983, based in part on the mentioned report that Empire had furnished it, the FNMA decided to withdraw approval of mortgage loans for condominium units in the 1-30 area. In July 1983, Zervas discovered that the appraisal of the 14.55 acres that he purchased in September 1982 had not in fact been signed by Tannehill. Although it was established that Tennehill’s signature was not genuine, there is no evidence of who forged it or furnished the appraisal. The appraisal was on the form used by Tanne-hill’s office, and was completed in the same style as his other appraisals.

Tannehill then had ten or twelve other appraisers working for him, and when asked if this appraisal was done by someone in his office, he responded, “I don’t know. I can’t say — I don’t believe it was done by someone in my office.” Zervas testified that he was never sure that Tannehill did not personally approve the appraisal; Tannehill denied doing so.

With respect to the September 1982 market value of the land, Tannehill noted that in March 1982 he had appraised some substantial portion of the same tract “for only about three million dollars.” However, on August 27, 1982, eleven days before the date of the forged appraisal, he had appraised 6.5 acres out of the 14.55-acre tract at a value of $3,149,-000 (about $11.12 per square foot).

Tanne-hill explained that this August 27 value was significantly higher than that of his March 1982 appraisal because land prices were then escalating rapidly and the March appraisal relied on mid-1981 comparable sales, while the August 27 appraisal relied on recent sales in the summer of 1982.

Tannehill further testified that the portion of the 14.55-acre tract not covered by his August 27 appraisal, approximately eight acres, included a substantial amount of land fronting on 1-30 which was “the more valuable property.” He stated that the value shown for the 14.55 acres in the forged September 7 appraisal — $7,162,000 or $11.30 a square foot — was then “within the range of fair market value” for the tract, although he personally would have appraised the entire 14.55 acres “slightly lower,” namely, “somewhere in the ten dollar [per square foot] range,” or somewhat over $6,000,000.

There was no contrary evidence as to the proper appraisal value range for the 14.55 acres in September 1982, although Zervas stated at trial that he now thought that the land was then worth only one or two million dollars. After discovering the appraisal forgery, Zervas attempted to sell the portion of the land that he had purchased which was still encumbered by the Lancaster loan back to Faulkner and Toler at a $500,000 profit to himself. However, because Faulkner and Toler were only willing to accept the property for the amount that Zervas still owed to Lancaster, no agreement was reached.

Zervas continued to work on his condominium projects. Shortly thereafter, Zervas sold Canterbury Landing to an investor that Blain had obtained. This sale was for the debt against the project, plus enough for Zervas personally to realize approximately $120,000 cash. Zervas never completed the Pentridge project, however. On March 14, 1984, the Federal Savings and Loan Insurance Corporation (FSLIC) closed Empire, thus cutting off Zervas’s source of funds. At the time of trial, Zervas owed approximately $4,500,000 to the FSLIC as successor to Empire.

On February 16, 1984, Zervas filed this suit against Blain, Nix, Faulkner, and numerous others. Zervas’ complaint asserted five claims: a claim of substantive RICO violation under 18 U.S.C. § 1962(c) (participating in the conduct of the affairs of an enterprise through a pattern of racketeering activity); a RICO conspiracy claim under 18 U.S.C. § 1962(d) (for conspiracy to violate

section 1962(c)); a pendant Texas common-law' claim for civil conspiracy to defraud; a pendant Texas common-law fraud claim; and finally, a pendant Texas claim for fraud in the sale of real estate under Tex.Bus. & Com.Code Ann. § 27.01. Prior to trial, all defendants other than Blain and Nix were dismissed, all or most having settled or taken bankruptcy. Thereafter the case was tried to a jury, which returned a verdict finding in Zervas’ favor, and against both Nix and Blain, on Zervas’ RICO substantive and conspiracy claims, on his civil conspiracy claim, and on his

section 27.01 claim, fixing damages at $1,000,000 on each of these four claims. The jury found in favor of Nix and Blain on the common-law fraud claim. More specifically, the jury found: (

a) as to the substantive RICO claim (labeled in the charge the “first claim”), that Blain and Nix each committed two or more wholly unspecified acts of mail or wire fraud (contrary to 18 U.S.C. § 1341 or § 1343) or commercial bribery (contrary to Tex.

Penal Code Ann. § 32.43) amounting to a pattern of racketeering activity (Question 1), that through the pattern of racketeering activity defendants each conducted or participated in the conduct of an enterprise (Question 2) which they were employed by or associated with (Question 3) and which affected interstate commerce (Question 4), and that defendants’ operation of the enterprise through a pattern of racketeering activity injured Zervas in his business or property (Question 5), for which $1,000,000 would be reasonable compensation (Question 18); (

b) as to the RICO conspiracy claim (labeled in the charge the “second claim”), that each defendant was a member of a conspiracy of two or more persons to conduct the affairs of an enterprise through unspecified acts constituting a pattern of racketeering activity (Question 6), and that “by reason of such conspiracy” Zervas was injured in his business or property (Question 7), for which $1,000,000 would be reasonable compensation (Question 19); (

c) as to the common-law conspiracy to defaud claim (the “third claim”), that there was “a conspiracy to defraud” Zer-vas “with respect to ... [his] purchase of the 15 acre tract which is the subject of this suit” (Question 8), that Blain and Nix were each members of the conspiracy (Question 9), and that acts done in furtherance of it damaged Zervas (Question 10), for which $1,000,000 would be reasonable compensation (Question 20); (d)as to the common-law fraud claim (the “fourth claim”), that neither defendant, in connection with a transaction with Zervas, misrepresented a material existing fact to him (Question 11).

The jury was instructed in connection with Question 11 that: “Plaintiffs allege that defendants misrepresented to them the following facts: “(1) the market value of the 15 acre tract purchased by Zervas; “(2) that the area would not be overbuilt with residential condominiums; “(3) that the condominiums being constructed were selling and closing quickly upon completion of construction; “(4) that the projects in the 1-30 area would be approved by the Federal National Mortgage Association; “(5) that the difference between the reasonable cost of construction of condominiums in the area and the reasonable anticipated sales price would generate substantial profit to plaintiffs; and “(6) that all condominium units to be built by Zervas would sell promptly at a profit.” The jury also found that neither defendant knew or should have known of the falsity of any misrepresentation nor made a misrepresentation in reckless disregard of its truth or falsity (Question 12), nor intended to induce Zervas to rely on their misrepresentations (Question 13), that Zervas did not justifiably rely on “any misrepresentation(s)” of either defendant (Question 14) and was not damaged by their misrepresentation(s) (Question No. 15), and that zero dollars would be reasonable compensation for Zervas’ damages caused by either defendants’ misrepresentation(s) (Question 21); (

e) as to the

section 27.01 claim (the “fifth claim”), that “in the 1982 real estate transaction with” Zervas both Blain and Nix “committed fraud, as defined in

Section 27.01(a)” (Question 16), and did so willfully (Question 17), for which damages, defined as the difference between the real estate’s value as represented or promised and as delivered, were $1,000,-000 (Question 22); and (

f) the jury further awarded punitive damages of $50,000 against Blain and $25,000 against Nix for acting “willfully, deliberately, maliciously, or with reckless disregard” (Question 23), and also found that Zervas’ reasonable attorneys’ fees were $450,000 (Question 24).

The district court thereafter entered judgment “based on the jury’s findings on the first and second [RICO] claims in the verdict” for Zervas and against Blain and Nix, jointly and severally, for $6,450,000, representing a trebling of each of the jury’s two $1,000,000 RICO damage findings (Questions 18 and 19), and Zervas’ attorneys’ fees (Question 24), as provided by 18 U.S.C. § 1964(c). This appeal followed. Discussion Nix and Blain contend that the evidence is insufficient to sustain the verdict on the RICO counts which formed the basis of the district court’s judgment. We agree.

We assume, arguendo that the evidence warrants a finding that Blain and Nix committed or were responsible for certain RICO predicate acts as outlined in the court’s charge, such as mail or wire fraud in the interstate solicitation of deposits for Empire by false representations, or making misrepresentations to the FSLIC which insured Empire’s deposits, or commercial bribery by causing Empire to make loans on favorable terms to those who gave favors to Blain or Nix personally. The targets and victims of these acts were the depositors, other stockholders, and insurers of Empire. Zervas was none of these.

However, none of the charged predicate acts for which there is any support in the evidence were shown to be directed at Zer-vas, and no evidence shows he was a target or victim of any of them. Recovery under

section 1964(

c) is allowed only for injury to the plaintiff’s business or property “by reason of a violation of

section 1962.” In Sedima v. Imrex, 473 U.S. 479, 105 S.Ct. 3275, 3285, 87 L.Ed.2d 346 (1985), the Court noted that the RICO plaintiff “can only recover to the extent that he has been injured in his business or property by the conduct constituting the [RICO] violation” and that “ ‘a defendant who violates

section 1962 is not liable for treble damages to everyone he might have injured by other conduct’ ” (quoting Haro- co, Inc. v. American National Bank & Trust Co., 747 F.2d 384, 398 (7th Cir.1984), aff'd, 473 U.S. 606, 105 S.Ct. 3291, 87 L.Ed. 2d 437 (1985)). Sedima goes on to say that “the compensable [RICO] injury necessarily is the harm caused by predicate acts sufficiently related to constitute a pattern” and that “[a ]ny recoverable damages occurring by reason of a violation of § 1962(

c) will flow from the commission of the predicate acts.” Id. (emphasis added) (footnote omitted). Several decisions interpreting this language have apparently concluded that only those injuries which directly flow from the predicate acts are compensable under RICO. In Morast v. Lance, 807 F.2d 926, 933 (11th Cir.1987), RICO recovery was denied because the plaintiffs injury “did not flow directly from the predicate acts” (emphasis added). Morast was followed in Nodine v. Textron, Inc., 819 F.2d 347, 349 (1st Cir.1987). The Seventh Circuit in Marshall & Ilsley Trust Co. v.

Pate, 819 F.2d 806 (7th Cir.1987), observed that “the plaintiff should recover for whatever damages are directly caused by any part of the acts that add up to a [RICO] violation.” Id. at 810 (emphasis added). Pate had previously rejected the argument that the damages must be caused not by a single predicate act but rather by a sufficient number of predicate acts to constitute a pattern, stating in this connection: “We do not believe that a plaintiff, in order to state a claim under

section 1964(c), must allege an injury to its business or property either caused by at least two predicate acts, or caused by all the acts adding up to a pattern. Rather, we believe a plaintiff must prove only: (1) a violation of

section 1962 (for example, a person furthering an enterprise through a pattern of racketeering activity), and (2) an injury directly resulting from some or all of the activities comprising the violation.” Id. at 809 (second emphasis added). This language was quoted with approval in Town of Kearny v. Hudson Meadows Urban Renewal Corp., 829 F.2d 1263, 1268 (3d Cir.1987). A requirement that the nexus between the injury and a predicate act be “direct” may, at least in some circumstances, be overly restrictive.

In Sedima, the Court, in a footnote to its statement that recoverable damages are those which “flow from the commission of the predicate acts,” stated that “[s]uch damages include, but are not limited to, the sort of competitive injury for which the dissenters would allow recovery. See post, at [105 S.Ct.] 3303.” Id. 105 S.Ct. at 3285 n. 15.

The cited portion of the Sedima dissent gives as an example of competitive injury an instance where a businessman caused by extortinate threats to hire certain unneeded workers may recover under RICO not only for the added cost of the workers but also for the “toll on the competitive position of the prospective plaintiff by increasing his costs of doing business” (presumably as compared to his competitors who have not been so threatened or have not succumbed to the threats). Id. at 3303.

Another competitive injury example given in the dissent is that of a business which conducts its affairs through a pattern of racketeering activity “to enhance its profits or perpetuate its economic power,” in which event “competitors of that enterprise could bring civil RICO actions alleging injury by reason of the enhanced commercial position the enterprise has obtained from its unlawful acts.” Id. But the evidence here shows no analogous competitive injury to Zervas, nor do we understand any to have been claimed. Neither Blain, Nix, Empire, nor Stateside were shown to be competitors of Zervas.

Moreover, there is no evidence that any of the charged predicate acts put Zervas in a less favorable position than any of his competitors, or put his competitors in a more favorable position than him, with resulting injury to Zervas. Any finding of competitive injury to Zervas resulting from the charged predicate acts could be based on nothing more than speculation or conjecture. While a “direct” nexus between an injury and a predicate act may not always be required, nevertheless the vast majority of appellate decisions do require something more than mere “but for” factual causation.

The Seventh Circuit in Haroco observed that

section 1964(c)’s “ ‘by reason of language ... imposes a proximate cause requirement on plaintiffs. The criminal conduct in violation of

section 1962 must, directly or indirectly, have injured the plaintiff’s business or property.” Id., 747 F.2d at 398. In Sperber v. Boesky, 849 F.2d 60 (2d Cir.1988), the Second Circuit, though assuming that some indirect injuries from predicate acts were recoverable under RICO, nevertheless held that “legal liability should not extend as far as factual causation,” id. at 63, and “that both direct and indirect injuries must be proximately caused.” Id. at 64. Sperber further held that mere cause in fact and foreseeability of the injury were not necessarily sufficient to establish the requisite proximate cause.

Id. at 65-66. Rather, the question of proximate cause involved “social policy decisions based on shared principles of justice.” Id. at 65. Although it was assumed that the injury there was a foreseeable result in fact of the predicate acts, RICO recovery was nevertheless denied because the connection between the two was overly “attenuated.” Id. A similar approach was taken by the Fourth Circuit in Brandenburg v. Seidel, 859 F.2d 1179 (4th Cir.1988). This recent opinion states: “[A] cause-in-fact connection, standing alone, does not suffice to establish liability.

Civil RICO is of course a statutory tort remedy — simply one with particularly drastic remedies. Causation principles generally applicable to tort liability must be considered applicable.

These require not only cause-in-fact, but ‘legal’ or ‘proximate’ cause as well, the latter involving a policy rather than a purely factual determination: ‘whether the conduct has been so significant and important a cause that the defendant should be held responsible.’ Prosser and Keeton Torts, § 42 p. 272 (general principle) (5th ed. 1984)_ As such, the legal cause determination is properly one of law for the court, taking into consideration such factors as the foreseeability of the particular injury, the intervention of other independent causes, and the factual directness of the casual connection.” Id., at 1189.

Brandenburg concluded that the claimed predicate acts “were not so significant and important a cause of the plaintiffs’ loss” as to authorize RICO recovery. Id. at 1189. In National Enterprises, Inc. v. Mellon Financial Services Corp., 847 F.2d 251 (5th Cir.1988), we adopted a somewhat similar approach, although we neither expressly referred to proximate cause nor explicitly decided whether RICO extended to any indirect injuries.

We there denied RICO recovery to National, a supplier-creditor of G & N, a construction company which was allegedly rendered bankrupt, and hence unable to pay National, by the RICO bribery and fraud perpetrated on G & N by its bank, the defendant in National’s RICO action. In affirming a dismissal of National’s complaint for failure to state a claim, we reasoned that National’s loss was not caused by the predicate acts perpetrated, as required by Sedima. 847 F.2d at 254.

Recognizing the argument that the RICO fraud on G & N was a cause of G & N’s bankruptcy and hence of its failure to pay its creditor National, we held this theory went “far beyond the pale of rational causation” and that “the only sound perspective is to view National’s injury as flowing from” its supplier relationship with G & N and “not from the remote corrupt acts of a virtual stranger ... to that relationship.” Id. We are in general agreement with the approach taken in Sperber and Branden burg, and believe those principles are fully consistent with and logically flow from National Enterprises.

Zervas' theory seems to be that Blain and Nix were members of a conspiracy to artificially inflate land values in the 1-30 corridor, and that the predicate acts were among the means by which they amassed the funds to do so.

Though there is no direct evidence of a conspiracy involving Blain and Nix, we assume, arguendo, that the circumstantial evidence would have properly permitted a jury finding that Blain and Nix conspired with others to commit two or more predicate acts of mail or wire fraud or commercial bribery amounting to a pattern, and to conduct the affairs of Empire through that pattern of racketeering activity.

But there is simply no evidence whatever that any such conspiracy had as one of its objects the inflation of the value of the 14.55-acre tract in question; nor does the record contain anything more substantial than speculation and conjecture from which it could fairly be found that a purpose of the conspiracy was to inflate area land prices generally. Thus, any increase in the price of the 14.55-acre tract would at most be a remote, incidental, and nonproximate result of the charged RICO predicate acts and RICO conspiracy, and hence would not entitle Zervas to recovery in that respect under

section 1964(c). Moreover, the record also is lacking in any substantial evidence that either the RICO predicate acts or the conspiracy to conduct Empire's affairs through a pattern of such acts was in fact a cause of artificial increase in area land values, including the value of the 14.55-acre tract. The conclusion that Zervas was thus damaged, as a proximate result of the RICO violations, to the extent of $1,000,000 or to any other quantified extent, is based on no more than speculation and conjecture.

Zervas also seems to argue that his injury results from the overbuilding of condominiums in the 1-30 corridor, and that this in turn was caused by the RICO conspiracy and predicate acts. Again, however, essentially the same analysis shows this theory to be without merit. There was no substantial evidence that the charged predicate acts, or any conspiracy to thereby conduct Empire's activities, had as a purpose the overbuilding of condominiums, much less the purpose to injure Zervas (or anyone else) thereby. Any resultant overbuilding was, at most, a remote, incidental, and non-proximate result of the charged

section 1962 violations, and was hence not actionable under

section 1964(c). Moreover, there is no showing, beyond speculation and conjecture, that the overbuilding was in fact the result of the charged

section 1962 violations or that, if such violations had not occurred, Zervas would have been able to sell materially more of his condominium units or sell them materially faster. In sum, the record simply does not adequately support the conclusion that Zervas' injury was proximately caused by any of the charged

section 1962 violations on the part of Blain or Nix. Accordingly, the district court erred in overruling the motions for directed verdict on Zervas' first two claims and consequently in rendering judgment thereon. The judgment below therefore cannot be affirmed, as it is based only on those two claims. Nor is it appropriate for us to modify the judgment so as to rest it either on Zervas' claim of common-law conspiracy to defraud him in connection with his purchase of the 14.55-acre tract or his

section 27.01 claim of fraud in connection with that purchase. There is no evidence that Blain (or Empire) ever breached any promise either made to Zervas respecting his purchase of the 14.55-acre tract or ever made any factual representation to him in that connection. There is no evidence that Nix made any promise to Zervas, and the only arguable misrepresentation of fact she made to him was just before the closing as to the land's value.

In any event, the jury found that neither Nix nor Blain made (knowingly or otherwise) any factual misrepresentations to Zervas, specifically including any as to the market value of the 14.55-acre tract he purchased. Moreover, under the evidence, there is simply no rational basis on which the jury’s verdict on the fourth claim, for common-law fraud, can be reconciled with its verdict on the fifth claim, for

section 27.01 statutory fraud, nor with the hypothesis that the verdict on the third claim, for common-law conspiracy to defraud in connection with Zervas’ purchase of the land, rests on any factual misrepresentation by Blain or Nix (including those six listed in the charge in connection with the common-law fraud claim). Thus, the common-law conspiracy to defraud claim must rest on some other basis than any representation or promise made to Zervas by Blain or Nix. Zervas relies heavily on the “forged” appraisal. But this will not carry the day.

Zervas had not ordered and was not even aware of the appraisal until the time of closing, well after he had already agreed to purchase the property following extensive negotiation, and at closing he was not shown it by anyone, but merely glanced at it when he saw it lying among the closing papers. Most importantly, there is no substantial evidence that either Blain or Nix knew or suspected, or even had reason to suspect, that the appraisal was forged, or, indeed, that it was or would be an inducement to Zervas, or that either Blain or Nix had any

part in furnishing the appraisal, much less any

part in concocting it. Neither Blain (nor Empire nor Stateside) nor Nix was either a seller or a lender in the 14.55-acre purchase. There is no substantial evidence that Blain (or Empire) or Nix (apart from the normal title policy fee paid Dallas Title) directly or indirectly received anything from the closing, or any “kickback” of any kind in that respect. While Empire— through an officer other than Blain — made a conditional commitment to Lancaster, the lender in the sale, to loan $1,800,000 on completed condominium units, that is no evidence of either Empire’s or Blain’s complicity in any fraud against Zervas.

There is no indication this commitment was not fully honored, and given its conditional nature and relatively small size as compared to the $5,200,000 purchase loan, its materiality to the transaction is questionable at best. Nix’s role as closing officer for the title company is not even suggestive of a role in any conspiracy to defraud Zervas, nor is her answering — honestly, so far as the jury verdict on the fourth claim reflects — a general question posed by Zervas as to her opinion of the transaction.

For a defendant to be liable for civil conspiracy to defraud, it must be shown not only that there was such a conspiracy but that the particular defendant charged agreed with one or more of the other conspirators on the claimed illegal object of the conspiracy — here the defrauding of Zervas in his purchase of the 14.55 acres— and intended to have it brought about. As the Texas Supreme Court said in the somewhat analogous case of Schlumberger Well Survey Corp. v.

Nortex Oil and Gas Corp., 435 S.W.2d 854, 857 (Tex.1968), civil conspiracy requires “ ‘a meeting of the minds on the object or course of action, and some mutual mental action coupled with an intent to commit the act which results in injury; ... there must be a preconceived plan and unity of design and purpose’ ” (quoting with approval from 15A C.J.S. Consyiracy § 2 at 600).

The Texas court there likewise held that a civil “conspiracy to defraud” requires “ ‘a common yur-yose, supported by a concerted action to defraud, that each has the intent to do it, and that it is common to each ... and that each has the understanding that the other has that purpose.’ ” Id. (emphasis added by Texas Supreme Court) (quoting with approval from Brumley v. Chattanooga Speedway & Motordrome Co., 138 Tenn. 534, 198 S.W. 775, 776 (1917)).

Proof that a defendant had some collateral involvement in a transaction, and had good reason to believe that there existed a conspiracy among other parties to it, is insufficient of itself to establish that the defendant was a conspirator. Id. Although a conspiracy is usually proved by circumstantial evidence, nevertheless “vital facts may not be proved by unreasonable inferences from other facts and circumstances” or “by piling inference upon inference.” Id. at 858. See also Mack v. Newton, 737 F.2d 1343, 1350-54 (5th Cir.1984).

Moreover, “ ‘mere presence at the scene ... or close association with a coconspirator will not support an inference of participation’ ” in a conspiracy. United States v. Basey, 816 F.2d 980, 1002 (5th Cir.1987) (quoting United States v. Vergara, 687 F.2d 57, 61 (5th Cir.1982)). See also United States v. Stanley, 765 F.2d 1224, 1236-37, 1243-44 (5th Cir.1985); United States v. Jackson, 700 F.2d 181, 185 (5th Cir.), cert. denied, 464 U.S. 842, 104 S.Ct. 139, 78 L.Ed.2d 132 (1983).

A conclusion from this record that either Blain or Nix agreed with others to defraud Zervas in his purchase of the 14.55-acre tract can rest only on an overly attenuated chain of inferences and ultimately on no more than speculation and conjecture. Zer-vas simply did not sustain his burden of proof against either Blain or Nix on the civil conspiracy to defraud claim. Conclusion Accordingly, the judgment below is reversed and the cause is remanded to the district court for further proceedings consistent herewith. REVERSED AND REMANDED. .

These entities are: John Zervas Construction Development Company; One Pentridge Place Condominiums, Ltd.; Two Pentridge Place Condominiums, Ltd.; Three Pentridge Place Condominiums, Ltd.; and Four Pentridge Place Condominiums, Ltd. Because it does not affect the result, in this opinion we shall use the name Zervas to refer both to the natural person and to the natural person together with the entities controlled by him. . Most of these facts are undisputed. To the extent that they are not, we view the evidence as a whole in the light and with all reasonable inferences most favorable to the verdict. .

Zervas was convicted on his plea of guilty to a mail fraud indictment involving a 1979 land transaction and the related procurement of a land loan by materially false misrepresentations. . At this time, the vast majority of Zervas’ net worth consisted of his interest in the Galveston condominium project. He assigned that project a gross value of $3,630,000; it had a loan against it with a $2,970,000 outstanding balance, leaving an equity of $660,000. Falgiamo and others also owned an interest in the project.

Construction on that project had commenced in October 1981, and by October 1982, some thirty or forty units had been completed; by August 1982, some fifteen units were under contract of sale. Closings were slow. In March 1983, Zer-vas deeded the project back to the lender, and it released him from liability except for his promise to reimburse it up to $100,000 for any further losses it suffered on the loan. . These items were: $22,450 for title policy, recording, and escrow fees; $1,100 for survey; $700 for attorneys’ fees (Zervas said he thought this was for a title company attorney).

Zervas signed “approved" an itemized listing of all the disbursements of the $5,200,000 loan. No charge is shown for the appraisal. Zervas also negotiated at closing a reduction of the Merrill Lynch brokerage fee from $175,000 to $100,000. . Tannehill also testified that the three comparable sales shown in the forged September 7 appraisal were valid and properly comparable sales with which he was personally familiar. He further identified a tract, within approximately a half mile of, and "less desirable” than, the 14.55 acres, which sold for ten dollars a square foot in "mid-1982.” .

The jury answered Questions 12 through 15, despite instructions that they were not to answer any of those questions if they answered Question 11 as they did; they similarly answered Question 21, despite instructions that they were to do so only if they answered Question 15 in a manner other than as they did. So far as the record reflects, the verdict was accepted without any objection on this basis. . The instructions for Question 16 included the following: "Plaintiffs’ fifth claim is asserted under

Section 27.01 of the Texas Business and Commerce Code. This act is a state statute which provides in relevant part as follows: . "(

a) Fraud in a transaction involving real estate or stock in a corporation or joint stock company consists of a "(1) false representation of a past or existing material fact, when the false representation is "(

A) made to a person for the purpose of inducing that person to enter into a contract; and "(

B) relied on by that person in entering into that contract; or "(2) false promise to do

an act, when the false promise is "(

A) material; "(

B) made with the intention of not fulfilling it; "(

C) made to a person for the purpose of inducing that person to enter into a contract; and “(

D) relied on by that person in entering into that contract.” . Nix and Blain each moved for directed verdict at the close of all the evidence (as well as at the close of Zervas’ case); Nix also moved for judgment n.o.v., but Blain did not.

As to Blain, then, 5A Moore’s Federal Practice ¶ 50.12 (2d ed.) states the applicable rule: “A party’s failure to move for judgment n.o. v. does not preclude the court of appeals from reviewing the denial of the party's earlier motion for a directed verdict, but in this situation the appellate court may not direct the entry of judgment n.o.v. for the appellant; the only relief the appellant may obtain is an order for a new trial.” Id. at 50-91 (footnote omitted). Our decisions are in accord with the above statement. Smith v. Trans-World Drilling Co., 772 F.2d 157, 160, 162 (5th Cir.1985); Gorsalitz v.

Olin Mathieson Chemical Corp., 429 F.2d 1033, 1037-38 (5th Cir.1970), cert. denied, 407 U.S. 921, 92 S.Ct. 2463, 32 L.Ed.2d 807 (1972); Yorkshire Indemnity Co. v. Roosth & Genecov Production Co., 252 F.2d 650, 657-58 (5th Cir.1958). See also Jackson v. Seaboard Coast Line R. Co., 678 F.2d 992, 1021 (11th Cir.1982) (citing Gorsalitz and Yorkshire); United States v. Valdosta-Lowndes County Hospital Authority, 696 F.2d 911, 912-13 (11th Cir.1983) (quoting Jackson and its citation of Gorsalitz and Yorkshire).

The scope of our review of the denial of Blain’s motion for directed verdict is the same as if he had also moved for judgment n.o.v., namely, that set out in Boeing Co. v. Shipman, 411 F.2d 365, 374-75 (5th Cir.1969) (en banc), but the relief we may order in favor of Blain is limited to the grant of a new trial. See, e.g., Trans-World Drilling at 160 (Boeing standard), at 162 (new trial only remedy); Jackson at 1022 (Boeing standard). The Boeing standard of review contrasts to the narrower “any evidence” standard applicable where no motion for directed verdict has been made. See Hinojosa v.

City of Terrell, 834 F.2d 1223, 1228 (5th Cir.1988). A motion for directed verdict alerts the opposing party to the alleged insufficiency before the case is submitted to the jury, thus affording an opportunity to cure defects in proof, id., and requiring such a motion has also been said (perhaps questionably) to prevent a party from gambling on the verdict and "later question[ing] the sufficiency of the evidence that led to his defeat.” Coughlin v. Capitol Cement Co., 571 F.2d 290, 297 (5th Cir.1978). ■ These considerations are not applicable to a motion for judgment n.o.v.

Under Boeing, we “consider all the evidence&emdash; not just that evidence which supports the non-mover’s case&emdash;but in the light and with all rea- sonable inferences most favorable to the party opposed to the motion." 411 F.2d at 374. "A mere scintilla of evidence is insufficient” to sup- port a verdict, and a motion for directed verdict may be granted even if there is not "a complete absence of probative facts to support a jury verdict”; rather, to sustain a verdict the evi- dence supporting it must be "substantial." Id. at 374-75.

Circumstantial evidence alone may suf- fice to prove conspiracy, which is rarely evi- denced by explicit agreements. Nevertheless, even in conspiracy cases, the jury’s freedom to draw inferences from circumstantial evidence does not permit either wholly unreasonable in- ferences or those which amount to mere specu- lation and conjecture. See Mack v. Newton, 737 F.2d 1343, 1350-51 (5th Cir.1984). Our review is under these standards. these standards. . However, in American National Bank & Trust Co. v.

Haroco, Inc., 473 U.S. 606, 105 S.Ct. 3291, 87 L.Ed.2d 437 (1985), decided the same day, the Court expressly rejected the argument "that the injury must flow not from the predicate acts themselves but from the fact that they were performed as part of the conduct of an enterprise." Id. 105 S.Ct. at 3292. . The Eighth Circuit may be an exception in this respect. See Terre Du Lac Ass'n, Inc. v. Terre Du Lac, Inc., 772 F.2d 467, 472-73 (8th Cir.1985). We at least implicitly rejected this aspect of Terre Du Lac in National Enterprises, Inc. v. Mellon Financial Service Corp., 847 F.2d 251, 253-54 (5th Cir.1988). .

Indeed, the jury's answer to Question 22-that $1,000,000 was the difference between the value of the 14.55 acres as represented to Zer-vas, which he said was $7,000,000, and its actual value as delivered to him-suggests that the land was worth $6,000,000 when Zervas bought it for $4,000,000 (or $4,500,000, counting the "points"), thus negating any damage to him by inflated land values. . We recognize that these are criminal cases, and we do not suggest that the beyond a reasonable doubt standard applies to civil conspiracy actions, RICO or otherwise.

Nevertheless, the general principles of circumstantial evidence and proof of conspiracy are similar.