In this securities fraud class action, the district court granted Bankers Trust Company’s (BTC) motion for summary judgment. Appellant Thelma Metge appeals on behalf of herself and 635 others similarly situated, arguing that the district court erred on the facts and law regarding BTC’s liability as a controlling person and as an aider and abettor, and that the district court abused its discretion in dismissing pendent state claims against BTC. We affirm on the controlling person issue, but reverse and remand on the issues of aiding and abetting and pendent jurisdiction.
I. BACKGROUND.
Investor’s Equity, Inc. (IEI), is an Iowa corporation which originally was involved in the business of buying, selling, and servicing real estate contracts on low-cost homes. Fluctuating business conditions in 1969 caused IEI to form Investor’s Mortgage and Finance Company (IMF) to raise capital for IEI by selling unregistered, unsecured, one-year renewable securities called thrift certificates (certificates), which resembled promissory notes. Subsequently, IEI also issued thrift certificate guaranty bonds, which unconditionally guaranteed the certificates. Initially IMF transferred proceeds from certificate sales to IEI in exchange for IEI real estate contracts; but by the end of 1970, IMF was loaning the certificate proceeds to IEI in exchange for IEI’s unsecured promissory notes. Certificate sales were brisk and by mid-1974, the outstanding balance had risen to $1.5 million.
IEI invested the certificate proceeds in speculative real estate ventures which, by 1971, began to erode IEI’s financial standing. Over time, certificate proceeds grew, representing a larger share of IEI’s financing. Meanwhile, BTC had become heavily involved in financing IEI, and engaged in a series of banking strategies which kept IEI in business. By mid-1974, however, the State of Iowa prohibited the sale and renewal of unregistered thrift certificates. In August, 1974, IEI declared bankruptcy; as a result, certificate holders received only twelve and one-half cents on each dollar of accrued interest and no repayment of principal. BTC lost $543,703 in principal and $107,807 in accrued interest on its loans to IEI and its subsidiaries.
The appellant class of certificate holders sued BTC and seventeen individual appellees in federal district court, alleging state *624common law fraud and violations of federal securities laws stemming from misrepresentations and nondisclosure of relevant facts in connection with the sale of securities. In their federal securities law count, the appellants maintained that BTC is liable for violations of section 10(b) of the Securities Exchange Act of 1934 (15 U.S.C. § 78j) and Rule 10b-5 (17 C.F.R. § 240.-10b-5) on theories of aiding and abetting, conspiracy and controlling person liability. The district court granted summary judgment for appellees, and the appellants appeal on all grounds except the conspiracy theory. We consider each of these arguments in turn.
II. THE AIDING-AND-ABETTING THEORY OF LIABILITY UNDER SECTION 10(b) AND RULE 10b-5.
A. Legal Principles of Liability.
This Court has previously set forth a three-pronged test to establish aiding- and-abetting liability. These requirements include:
(1) the existence of a securities law violation by the primary party (as opposed to the aiding and abetting party);
(2) “knowledge” of the violation on the part of the aider and abettor; and
(3) “substantial assistance” by the aider and abettor in the achievement of the primary violation.
Stokes v. Lokken, 644 F.2d 779, 782-83 (8th Cir.1981). Because the theory of aiding- and-abetting liability is a matter of common law, the courts have not yet elaborated the full meaning of these factors. Nevertheless, certain aspects of the practical applications of these tests are clear. The factors — particularly the second and third— are not to be considered in isolation, but should be considered relative to one another. Id. at 784. Practically speaking, this means that “where there is a minimal showing of substantial assistance, a greater showing of scienter is required.” Id., citing Woodward v. Metro Bank of Dallas, 522 F.2d 84, 95 (5th Cir.1975). Thus, the two factors vary inversely relative to one another and where, as here, the evidence of substantial assistance is slim, the requirement of knowledge or scienter is enhanced accordingly.
The common law character of this three-part test requires further elaboration of the law before we apply the legal test to the facts of this case. Regarding the first prong of the test, we note that the existence of an underlying securities law violation by the primary party is not at issue in this case. Although BTC does not concede the existence of a violation, it recognizes that the appellants adduced sufficient evidence to give rise to a question of fact, which relieves us of further inquiry on this issue.
Regarding the “substantial assistance” factor, we note that the district court properly held that the appellants had the burden of showing that the secondary party proximately caused the violation. In approving this standard, we rely on decisions in other courts that have required a showing of “substantial causal connection between the culpable conduct of the alleged aider and abettor and the harm to the plaintiff[,]” Mendelsohn v. Capital Underwriters, Inc., 490 F.Supp. 1069, 1084 (N.D.Cal.1979), or a showing that “the encouragement or assistance is a substantial factor in causing the resulting tort[.]” Landy v. Federal Deposit Insurance Corporation, 486 F.2d 139, 163 (3d Cir.1974), cert. denied, 416 U.S. 960, 94 S.Ct. 1979, 40 L.Ed.2d 312 (1975) (quoting Restatement of Torts § 436.)
In further examining the “substantial assistance” requirement, we note that BTC’s involvement amounted to inaction rather than positive deeds of manipulation or deception. The distinction between positive action and deliberate inaction is elusive, and we do not wish to foreclose appellants’ strategies on remand, but the facts suggest that we examine the standard as it relates to substantial assistance through inaction.
Although courts are by no means unanimous in treating the question of substantial assistance in a case of inaction, *625most seem to agree that, if the aider and abettor owes the plaintiff an independent duty to act or to disclose, inaction can be a proper basis for liability under the substantial assistance test. E.g., Cleary v. Perfectune, 700 F.2d 774, 777 (1st Cir.1983) (citing cases which impose a duty to disclose where defendant possesses insider information, where a controlling person approves fraudulent practices, or where the law imposes special obligations on a profession). We find no such duty in this case. Nevertheless, the district court properly applied an exception to this rule in deciding the case under the so-called Woodward-Mon-sen rule. In Monsen v. Consolidated Dressed Beef Co., Inc., 579 F.2d 793 (3d Cir.1978), cert. denied, 439 U.S. 930, 99 S.Ct. 318, 58 L.Ed.2d 323 (1979), the Third Circuit evaluated the substantial assistance requirement in a case of inaction and concluded that inaction “may provide a predicate for liability where the plaintiff demonstrates that the aider-abettor consciously intended to assist in the perpetration of the wrongful act.” Id. at 800 (emphasis in original). The Fifth Circuit took a similar tack in Woodward v. Metro Bank of Dallas, 522 F.2d 84 (5th Cir.1975):
When it is impossible to find any duty of disclosure, an alleged aider-abettor should be found liable only if scienter of the high “conscious intent” variety can be proved. Where some special duty of disclosure exists, then liability should be possible with a lesser degree of scienter.
Id. at 97 (footnote omitted).
Simply put, in the absence of a duty to act or disclose, an aider-abettor case predicated on inaction of the secondary party must meet a high standard of intent. As applied here, Woodward and Monsen require that the aider-abettor’s inaction be accompanied by actual knowledge of the underlying fraud and intent to aid and abet a wrongful act. The requisite intent and knowledge may be shown by circumstantial evidence. Woodward, 522 F.2d at 96.1
B. Application of the Law to the Facts.
At the outset, we note the standard governing reliance on inferences in a summary judgment motion: direct proof is not required to create a jury question, but to avoid summary judgment, “the facts and circumstances relied upon must attain the dignity of substantial evidence and not be such as merely to create a suspicion. * * Furthermore, an inference which a jury, is entitled to draw must be based upon proven facts and not upon other inferences.” Impro Products, Inc. v. Herrick, 715 F.2d 1267, 1272 (8th Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 1282, 79 L.Ed.2d 686 (1984), citing Admiral Theatre Corp. v. Douglas Theatre Co., 585 F.2d 877, 884 (8th Cir.1978).
In evaluating the record under this standard, we must determine whether the bank understood that the thrift certificates were worthless and, as a result, prolonged the business lives of IEI and IMF for its own benefit at the expense of the certificate holders. Although we express no view of the outcome of this case on the merits, careful review of the record convinces us that the appellants adduced sufficient evidence to avoid a summary judgment.
To explore the question of BTC’s knowledge, we initially examine the nature of the relationship between BTC and IEI. The relevant history began in 1970 when BTC acquired by foreclosure a controlling block of stock in Mid-Iowa Lakes Corporation (MIL), which developed recreational real estate outside Des Moines, Iowa. Later *626that year, BTC, along with an Omaha bank, loaned MIL nearly $1.5 million in an attempt “to keep everything afloat.”
In February, 1971, BTC sold the MIL stock to IEI under innovative terms, which figure prominently in this litigation. The total value of the transaction was $950,000, of which IEI paid $130,000 in cash. In addition, IEI gave BTC a $575,000 collateral pledge note for the acquisition loan, 50,000 shares of IEI stock, an irrevocable proxy to vote the MIL stock (the MIL stock was registered to IEI with blank stock powers back to BTC), and a guaranty of the acquisition loan by IEI directors and officers totaling $440,000. The 50,000 shares of IEI stock (the largest block of outstanding IEI stock, amounting to seventeen to eighteen percent of total shares) were transferred to Bell and Company, which was a partnership of senior BTC officers.
The proxy held by BTC in the MIL stock was unlimited and irrevocable. It was qualified, however, in that the purchase agreement included a proviso that BTC could only vote the stock when IEI was in default on its payments under the purchase agreement. IEI was frequently behind in its payments, which gave BTC at least potential control virtually from the outset of the agreement. Because the proxy allowed BTC to vote a controlling share of MIL stock, it gave BTC potential control over the general business affairs of MIL; for instance, it could call shareholder meetings, influence the composition of the board of directors, and amend articles and by-laws.
The purchase agreement also included a provision which empowered BTC to “put” or sell IEI stock back to IEI at a rate higher than BTC had paid for the stock. Thus, BTC was protected against the vagaries of the market, for if the stock value did not increase, IEI could be compelled to repurchase it at a higher price, but if the stock’s value rose, BTC could sell it in the open market at a profit.
As part of its lending relationship with IEI, BTC also arranged to receive regular financial reports on IEI and MIL. Although in 1972 they were initially receiving quarterly statements (which showed quarterly balances of outstanding thrift certificates), by 1973, BTC was also receiving unaudited financial statements which treated IEI on a separate consolidating basis and revealed large thrift certificate balances.
An account of BTC’s relationship with IEI must also consider the measures that BTC undertook to perpetuate IEI’s business life. Viewed separately, most of these banking transactions are unremarkable events in the turbulent economy of the early seventies; but viewed in conjunction with the other evidence we discuss, they suggest an unusual pattern of extraordinary attempts to prolong IEI’s financial viability.
Under the terms of the MIL acquisition loan, IEI was obligated to make a $100,000 downpayment, followed by $23,000 quarterly payments until September, 1972, when the loan would be repaid with a large balloon payment. When the first quarterly payment came due, however, IEI’s poor cash flow compelled it to default, as it also did on the second payment. After extending the deadlines several times, BTC recognized that IEI was in part hampered by its involvement in another recreational real estate project, and so to recover its quarterly payments and to secure prepayment of the third quarterly payment, BTC arranged a “take-out loan” of $200,000 for the other project and IEI made its first three payments on the original acquisition loan.2
In April, 1972, BTC acted as escrow agent for IEI in an exchange of other obligations for thrift certificates and thrift certificate guaranty bonds. IEI had previ*627ously purchased the assets and liabilities of the Triple Eye Corporation, which obligated it on certain outstanding notes. Working through BTC and with the approval of BTC’s president, IEI offered to exchange its thrift certificates and guaranty bonds for the notes. Two noteholders, who are members of the appellant class, accepted the offer, sent their notes to BTC and received IEI’s thrift certificates and guaranty bonds from BTC in exchange. In deposition, bank officers stated that this was a standard service offered by the bank’s trust department. Nevertheless, the bank’s role in the exchange is relevant at least in demonstrating its awareness of IEI’s reliance on the thrift certificate program in April, 1972.
In late September, 1972, after other late quarterly payments and discussion of exercising the “put” on the IEI stock and of seeking protection from the guarantors of the MIL acquisition loan, BTC agreed with IEI to renew and amend the terms of the loan agreement. Under the revised agreement, IEI was obliged to make an immediate $30,000 payment on the MIL acquisition loan, with the due date on the balance extended to October, 1974, on a new nine percent note. In addition, the bank attached an eleven percent service charge to the “put” because, according to BTC president Robert J. Sterling, the “put” represented a debt on which the bank made no interest or earnings if the “put” was not exercised. The “put” was additionally secured by stock of another IEI subsidiary and by mortgages on some IEI real estate. IEI also agreed to provide monthly, rather than quarterly, consolidating statements for itself and each of its subsidiaries, and it also agreed to refrain from borrowing more than $100,000 from any other lenders without BTC’s consent.
Refinancing became necessary in May, 1973, after IEI defaulted on two more note payments and on two of the “put” service charge payments. (It also defaulted on the next six payments due on both of these obligations, through at least April, 1974.) According to Sterling, the May, 1973, refinancing plan was intended “to keep all of this thing afloat.” Under this new plan, IEI borrowed an additional $71,000 from BTC, which was used to satisfy delinquent loan payments and service charges. As collateral for the loan, IEI secured a $71,-000 note from MIL. In addition, BTC required IEI to pledge additional MIL stock as collateral for a $100,000 obligation which had previously been unsecured. Although the $71,000 was a “wash” transaction, the application of the loan proceeds to delinquent payments brought IEI’s obligations up to date and apparently protected BTC with more collateral. Unfortunately, in real terms, the refinancing plan did little to help keep IEI “afloat.” On December 31, 1973, BTC was obliged to loan IEI an additional $69,000 on an “emergency” basis to pay delinquent taxes.
Yet by March, 1974, IEI was once again in serious financial trouble which required BTC’s intervention in the form of a final refinancing plan resembling the plan which went into effect ten months before. When asked if BTC was very concerned about its position on certain loans to IEI, loan officer Chris Pappas replied, “That would be an understatement.” As a result, IEI was obliged to sell a $13,000 receivable note and to pledge its interest in a plastics company to secure a previously unsecured loan. In addition, MIL issued a $25,000 note to IEI, which IEI pledged to cover the balance of the $71,000 loan in May, 1973. Finally, MIL borrowed $46,000 from BTC (which was secured by receivables) which it paid to IEI which in turn paid to BTC (along with $16,000 in other funds) to bring IEI’s obligations up to date. Thrift certificate sales ceased in May, 1974, and by the end of summer, IEI had filed for bankruptcy under Chapter 11.
We can only conclude from the above that several times BTC apparently rescued IEI and its subsidiaries from the brink of financial disaster with the declared objective of preserving the enterprises. The relevance of these banking decisions to the aiding-and-abetting issue becomes apparent when we review the extent to which the bank knew the extent of IEI’s financial *628woes and of the extent to which it knew that IEI depended for cash on the sale of then-worthless thrift certificates.
In considering the financial data, the district court relied on IEI’s consolidated balance sheets (which reflected the status of IMF and other subsidiaries) prepared in March, 1974, and concluded that IEI had a net worth of $975,000 at the end of fiscal 1972 and $500,000 at the end of fiscal 1973. While these figures, as far as they go, appear to be correct, a proper consideration of the existence of a jury issue on this question requires a more thorough analysis of IEI’s financial outlook. When the net worth figures are considered along with certain qualifications and with other relevant figures, IEI’s future looks considerably bleaker, and BTC’s banking decisions are cast in a different light.
First, we review the assumptions and qualifications embodied in the net worth figures. At the deposition to which the appellees refer in their brief, the net worth figure is qualified by the assumption that the assets could be sold at the stated value and that the net worth figure is also subject to contingent liabilities cited in the balance sheets. Moreover, Coopers and Lybrand, IEI’s accountants, prepared the balance sheets on a “going concern basis” which “contemplates the realization of assets and liquidation of liabilities in the ordinary course of business. * * * [T]he Company has not generated sufficient revenue to cover all items of expense and to liquidate liabilities in accordance with the terms of such liabilities. The Company’s ability to retain the assets which are pledged as collateral for the notes which are in default and the continuation of operations depends upon achieving profitable operations and retaining additional and/or continued financing.” According to Coopers and Lybrand, the figures were also subject to “(1) the Company’s achieving profitable operations and obtaining additional and continued financing, and (2) any possible adjustment arising from the determination of the disputed liabilities and related litigation and final acceptance of a subsidiary utility company’s rates[.]” These assumptions throw the net worth figures in considerable doubt as regards the jury issue.
Yet perhaps more important, the net worth figures contrast with other less optimistic measures of IEI’s economic vitality which were included in the March, 1974, report and preceding reports. We note that IEI’s net loss was $151,346 in fiscal 1971, which more than doubled to $311,611 the following fiscal year, and increased to $365,101 for fiscal 1973. In addition, the March, 1974, report noted that IEI lost its office building for failure to keep the mortgage current and that IEI lost its rights on two leases on undeveloped property due to nonpayment of rent on the land leases.
We also note that, in previous years, BTC had performed its own analysis of IEI’s audited financial statements and found that IEI faced difficult economic problems. As early as February 10, 1972, BTC concluded that, for fiscal 1971, IEI exhibited an assets-to-liabilities ratio of .236 to 1 ($905,000 to $3,840,400), and a borrowing ratio of 8.70 to 1, based on liabilities of $10,933,900 and a capital structure of $1,257,300. By March, 1973, BTC’s spread sheets on IEI’s financial status revealed that the assets-to-liabilities ratio had improved to 1.09 to 1, but that its borrowing ratio had increased to 13.26 to 1. BTC concluded that the latter ratio showed “that the Company is under extreme pressure and Company leverage for creditors is extremely limited.” Significantly, BTC loan officer Chris Pappas testified in deposition that these ratios — assets to liabilities and debt to worth — are the “most important” ratios on the spread sheets.
In evaluating these data and the relationship between BTC and IEI, we analyze finally the status of the thrift certificates to determine the likelihood that BTC had actual knowledge of their importance to IEI as its financial fortunes worsened. Careful review of the deposition testimony of BTC officers reveals that they deny all knowledge of the thrift certificates, of the fact that they were secured by real estate, and of the possibility that thrift certificate *629proceeds were the essential means by which IEI stayed alive and was able to make any payments on its obligations to BTC.
Due to commingling of accounts and the difficulty of tracing funds to their sources, little evidence actually documents the transfer of thrift certificate proceeds to BTC in satisfaction of IEI’s obligations. Nevertheless, we find noteworthy two incidents from which a trier of fact might have inferred that BTC had actual knowledge of the existence and extent of the thrift certificate program. First is the Triple Eye Corporation’s note exchange in April of 1972, which we previously discussed. BTC president Robert J. Sterling approved the bank’s role as escrow agent in handling the exchange of the Triple Eye notes for IEI thrift certificates and BTC’s trust department handled the matter. Although only two members of the plaintiff class were involved in the exchange, the incident could be interpreted as confirming BTC’s knowledge of the program.
The second incident involves thrift certificate proceeds received by loan officer Chris Pappas on September 30, 1972, as part of the renewed and amended loan agreement between BTC and IEI. The record reflects that Pappas received a hand-delivered $30,000 check from IMF made payable to IEI, an IEI deposit slip for the $30,000 check, and an IEI check for $30,000 payable to BTC. He signed a receipt for the deposit and personally credited the IEI account for the following business day. Although Pappas argued in deposition that he did not make the connection between the deposits and thrift certificate proceeds until he was asked about it during the deposition, we regard the evidence of the transaction as potentially probative of BTC’s actual knowledge as we survey the totality of the evidence.
The other set of circumstances surrounding the thrift certificate program that we regard as important in evaluating the propriety of the summary judgment is the growth in thrift certificate sales relative to IEI’s declining financial health. We believe that records available to BTC confirm that these developments were inversely correlated — the worse things got for IEI, the higher were its thrift certificate proceeds. For instance, according to the balance sheet for fiscal 1971, while IEI’s financial condition worsened, outstanding thrift certificates totaled $392,100, which was over three times the balance of the previous fiscal year. By March, 1972, as IEI’s woes worsened, the total had risen to $482,950, and a September “Cash Flow Projection” assumed continued sales of at least $15,000 per month. By March of 1973, outstanding thrift certificates reached $694,970 and at the time of IEI’s bankruptcy, they totaled nearly $1.5 million. Thus, the impressive growth in thrift certificates is a nearly perfect reciprocal of IEI’s downward-sloping path, which we recounted earlier.
The final inquiry we make in evaluating the relationship between BTC and IEI involves the extent to which BTC can be showed to have benefited from postponing IEI’s bankruptcy. Although it is undisputed that BTC lost over $650,000 in principal and accrued interest due to IEI’s insolvency, Metge v. Baehler, 577 F.Supp. 810 at 814, we believe that this figure should not be the end of the inquiry, for the loss must be placed in context. First, we note that holders of thrift certificates, who were entitled to $1.5 million of principal, received none of their principal and only twelve and one-half cents of every dollar of accrued interest. Moreover, BTC’s initial exposure in its lending arrangement was $950,000, which, when compared with its loss, suggests that it received significant payments of principal and interest as well as the service charges it collected on the “put” arrangement after it renegotiated an eleven percent service charge in September, 1972. Thus, although the record is not very illuminating on the question of BTC’s benefit from IEI’s delayed bankruptcy, we find sufficient evidence at least to give rise to an inference that BTC may have benefited at the expense of the certificate holders from IEI’s renewed lease on life.
*630Although the facts we have recounted here at length are unremarkable taken in isolation, we find that taken together, they present what should have been a jury issue on the question of aiding-and-abetting liability. In sum, we have remarked on an unusually favorable banking relationship between the two parties at a time when BTC knew of IEI’s precarious financial position. Moreover, the evidence suggests that BTC had knowledge of the thrift certificate program and its importance to IEI’s ability to repay BTC’s loan to IEI. Finally, we find an inference arising from the evidence which suggests that, although BTC made no profit on its loans to IEI nor even recovered its outstanding funds by postponing IEI’s demise, it may have been able to lever itself into a more favorable position than the holders of thrift certificates. We express no view of any likely outcome on the merits, but hold only that summary judgment at this stage of the proceedings appears to have been premature.
III. CONTROLLING PERSON LIABILITY.
A. The Applicable Legal Standard.
The appellants also contend that the district court erred in granting summary judgment in favor of the appellees on the issue of controlling person liability. Careful review of the legal and factual issues under this theory persuades us that the district court did not err.
Controlling person liability is predicated on 15 U.S.C. § 77o, which provides:
Every person who, by or through stock ownership, agency, or otherwise, or who, pursuant to or in connection with an agreement or understanding with one or more other persons by or through stock ownership, agency, or otherwise, controls any person liable under sections 77k or 111 of this title, shall also be liable jointly and severally with and to the same extent as such controlled person to any person to whom such controlled person is liable, unless the controlling person had no knowledge of or reasonable ground to believe in the existence of the facts by reason of which the liability of the controlled person is alleged to exist.
Congress has provided little guidance for the courts in construing these sections, however, believing that strict definitions of control would be “undesirable” because “[i]t would be difficult if not impossible to enumerate or to anticipate the many ways in which control may be exerted.” H.R. Rep. No. 1383, 73rd Cong., 2d Sess. 26 (1934). The SEC appears to have adopted a similar approach in formulating a broad definition of control as “[t]he possession, direct or indirect, of the power to direct or cause the direction of the management and policies of a person, whether through the ownership of voting securities, by contract, or otherwise.” 17 C.F.R. § 240.12(b)-2 (1983). Accordingly, we turn to the courts for guidance.
In Myzel v. Fields, 386 F.2d 718 (8th Cir.1967), cert. denied, 390 U.S. 951, 88 S.Ct. 1043, 19 L.Ed.2d 1143 (1968), this Court observed that “[t]he statute is remedial and is to be construed liberally. It has been interpreted as requiring only some indirect means of discipline or influence short of actual direction to hold a ‘controlling person’ liable.” Id. at 738 (emphasis added). We believe that the district court properly followed Myzel’s direction for interpreting the statute.
The district court relied on Stern v. American Bankshares Corp., 429 F.Supp. 818 (E.D.Wis.1977), to formulate its controlling person test. The Stem court declared that, to establish a prima facie case of controlling person liability, the plaintiff must demonstrate that the purported controlling persons “actively participated in the operations of Bankshares and possessed actual control over the transaction in question. Once that is established, the matters of knowledge and actual participation, which directly or indirectly induced the fraud, are matters for the defense.” Id. at 824.3
From Stern, and in compliance with this Court’s guidance in Myzel, the district *631court formulated a two-point test under which plaintiffs must establish, first, that the defendant lender “actually participated in (ie., exercised control over) the operations of the corporation in general; then he must prove that the defendant possessed the power to control the specific transaction or activity upon which the primary violation is predicated, but he need not prove that this later power was exercised.” Metge, 577 P.Supp. at 817-18. We approve this test because it complies with precedents which counsel broad remedial construction of the statute and because it complies with precedents that distinguish between actual exercise of control in the violator’s principal affairs and potential control over the violation.
Although other courts of appeals have adopted the more restrictive “culpable participation” requirement, which requires a showing that the lender actually participated in the alleged violation, Rochez Brothers v. Rhoades, 527 F.2d 880, 889-90 (3d Cir.1975); Gordon v. Burr, 506 F.2d 1080, 1085-86 (2d Cir.1974); see also Christoffel v. E.F. Hutton & Co., 588 F.2d 665, 668-69 (9th Cir.1978) (discussing without deciding the relevance of the culpable participation standard), we reject this more stringent standard for reasons announced by the Fifth Circuit in G.A. Thompson v. Partridge, 636 F.2d 945 (5th Cir.1981). That Court, citing the text of the statute and regulation we set out above, noted that the plain meaning of both texts does not require participation in the wrongful transaction. Id. at 957-58. In addition, good faith and lack of participation are affirmative defenses in a controlling person action, and requiring them as part of the plaintiff’s prima facie case while allowing them as affirmative defenses, confuses the parties’ responsibilities and unnecessarily burdens plaintiffs contrary to the plain meaning of the statute. Id.; see also Metge, 577 F.Supp. at 815-16. Accordingly, we find no error of law in the district court’s formulation of the controlling person test.
B. Application of the Legal Standard to the Facts.
Due to the bifurcated nature of the district court’s two-pronged test, we first consider whether the appellants introduced sufficient evidence of BTC’s actual control over IEI’s general affairs. Because we find this evidence to be insufficient, we need not consider evidence regarding BTC’s potential control over the alleged violation.
The appellants insist throughout their brief that the facts that they cite must be considered against the backdrop that BTC was IEI’s primary lender, with the ability to foreclose on the loans whenever IEI was delinquent on payment. The appellants cite several factors which they argue suggest actual control of IEI. First, BTC held seventeen to eighteen percent of IEI’s stock, with the prerequisites that accompany stock ownership (including the right to attend stockholder meetings, to vote its stock, and to receive stockholder communications). Second, BTC influenced the management of IEI’s subsidiary, MIL, in that, as part of its security agreement on the acquisition loan, BTC held the proxy on a controlling block of fifty-one percent of MIL’s voting stock; third, the “put” arrangement in the security agreement on the acquisition loan gave BTC power over IEI’s policy regarding capital stock; fourth, BTC personnel attended a few MIL and IEI board meetings; fifth, BTC controlled IEI’s operations through its innovative refinancing arrangements through which BTC influenced IEI’s and MIL’s debt structure by causing IEI to issue notes, sell its receivables and loan money to MIL. (We also consider as part of the record BTC’s alleged receipt of financial reports and its performance of services as escrow agent in the exchange of the Triple Eye Notes.)
We agree with the district court that careful review of the record reveals no evidence which demonstrates actual con*632trol. At most, these facts suggest the potential for influence over some of IEI’s business decisions, but they are not probative of appellants’ allegation that BTC actually exercised control over the operation of the corporation in general. In fact, the evidence we have mentioned exemplifies the kind of evidence that might be adduced on the second prong of the test to demonstrate possession of power to control, rather than actual control.
IV. PENDENT JURISDICTION.
Because we find that the district court erred in granting summary judgment against the appellants on the aiding-and-abetting theory of liability, we also reverse on the dismissal of the pendent state claims of common law fraud. These state claims appear to arise out of a common nucleus of operative fact, United Mine Workers v. Gibbs, 383 U.S. 715, 725, 86 S.Ct. 1130, 1138, 16 L.Ed.2d 218 (1966), and, in the interest of judicial economy, should be resolved with the primary federal claims.
Accordingly, the judgment of the district court is affirmed on the controlling person theory of liability, but is reversed and remanded on the aiding-and-abetting theory and on the pendent state claims.
. Because the circumstances in this case dictate the high scienter standard, we leave open the question of whether other facts in another case with more substantial assistance might state a claim under a recklessness standard of scienter. See Ernst & Ernst v. Hochfelder, 425 U.S. 185, 194 n. 12, 96 S.Ct. 1375, 1381, n. 12, 47 L.Ed.2d 668 (1976). Other circuits have limited the applicability of recklessness as a scienter requirement to cases in which the secondary party owed the plaintiff a duty. Cleary v. Perfectune, 700 F.2d 774, 777 (1st Cir.1983); Edwards & Hanly v. Wells Fargo Securities Clearance Corp., 602 F.2d 478, 484 (2d Cir.1979), cert. denied, 444 U.S. 1045, 100 S.Ct. 734, 62 L.Ed.2d 731 (1980); Woodward, 522 F.2d at 97.
. Because of title problems and construction delays, the other IEI project was never funded, and that division of IEI went bankrupt. Depositions of bank officers indicate, however, that they did not trace the delinquent payments to the take-out loan; according to BTC loan officer Chris Pappas, it is not banking practice to question the source of the payments "when you get paid."
. In part, the Stem court relied on Holloway v. Howerdd, 377 F.Supp. 754, 761 (M.D.Tenn.*6311973): "A director's liability [as a controlling person] * * * presupposes some extent of actual participation in the corporation’s operation before the consequences of control may be imposed.”