FRANK, Circuit Judge.

The issue before us is the extent to which stockholders and affiliated corporations are liable on a contract made by a so-called “leasehold” corporation. 1 The two individual defendants, Murray and Samuel Rosenberg, own substantially all of the capital stock of Murray M. Rosenberg, Inc., which was engaged in operating a chain of retail shoe stores. The shoes sold in these stores bore the name “Miles,” a trade-name owned by Miles Shoes, Inc., also substantially owned by the individual defendants, and apparently used by Murray M. Rosenberg, Inc., under some undisclosed arrangement.

In 1926, Murray Rosenberg approached the plaintiffs to negotiate the terms of a lease of certain premises in Paterson, New Jersey. Their version of the negotiations is as follows: “When we had agreed upon the terms of the lease, Murray M. Rosenberg told us that the tenant was to be the Mursam Shoe Corporation. I asked him who was the Mursam Shoe Corporation. Murray M. Rosenberg represented to me that the name Mursam was an abbreviation for Murray and Samuel, and that he and his brother were the corporation and ‘stood behind’ the lease. He told us that the store to be opened at the leased premises by them, was to be part of the chain of stores which he and his brother were then operating. Relying upon these representations, the plaintiffs entered into a lease with Mursam for a term of fifteen years.” The Mursam Shoe Corporation was organized b.y the Rosenbergs the day the lease was signed and sealed by the plaintiffs, and two days later it signed and sealed the lease as tenant. According to its books, the original capital investment in Mursam was $1; apart from paying legal and similar fees arising out of the organization of Mursam, the Rosenbergs paid nothing for their stock, and it does not appear that subsequently they made any contributions to capital. Mursam was, therefore, a corporation without assets. Its obligation under the lease was $10,000 annually, for the first five years, $11,000 for the next five and $12,000 for the last five years, or $165,000 for the entire term.

For fourteen years Mursam met its obligations on this lease. It was able to do so because of payments made to it by Murray M. Rosenberg, Inc:, which occupied the leased premises under short term subleases. In March 1940, Murray M. Rosenberg, Inc., terminated the sublease then in effect (made February 1, 1939), which was of unspecified duration on a monthly basis, and vacated the premises. Mursam having no assets, this action for damages caused by breach of the lease by failing to pay rent was brought against the other individual and corporate defendants as well.

It is alleged that Mursam is only an instrumentality through which these other defendants carry on their chain store operations, and that it is necessary to pierce its corporate “veil” in order to prevent injustice and circumvent fraud. There is evidence that Mursam’s corporate identity was often ignored by the other defendants themselves, and that funds were shifted about from one defendant to another without regard to formality and to suit their immediate convenience. Thus, the very first transaction under the lease was the payment of a $5,000 deposit by the personal check of Murray M. Rosenberg. Although this payment is asserted to be a loan to Mursam, its books are silent as to any arrangement with Murray M. Rosenberg for the borrowing or repayment of this amount. For several years Murray M. Rosenberg, Inc., paid to Mursam the amount of the taxes on the property (which Mursam was required to pay by its lease with the plaintiffs), although no provision in the subleases *346 required Murray M. Rosenberg, Inc., to pay these amounts. Rents due on a tobacco-stand on the premises which was subleased by Mursam to one Mourad Papas were, with a few exceptions, collected by Murray M. Rosenberg, Inc. Although by a surrender dated January 15, 1934, Murray M. Rosenberg, Inc., transferred title to the fixtures on the premises to Mursam, the fixtures were removed by Murray M. Rosenberg, Inc., when it vacated the store. Certainly Mursam was undercapitalized, and the payment of dividends and salaries soon relieved it of any surplus which it managed to accumulate by virtue of favorable subleases with Murray M. Rosenberg, Inc. For a substantial period Mursam lacked a bank account and books of account, and it never paid Murray M. Rosenberg, Inc., or Miles for accounting services rendered to it or for its use of their offices.

Enough of the facts have been stated to indicate that on a full trial it might be found that Mursam was only a tool of the other defendants, deliberately kept judgment-proof, to obtain the benefits of a lease with the plaintiffs without assuming any obligations. The plaintiffs allege that this was done fraudulently, in that Mursam was generally represented as being an integral part of their chain store operations and as having the same business and financial responsibility as the other defendants, so that the public was led to believe that dealing with Mursam was the same as dealing with them. The plaintiffs also state, in the language quoted above, that specific representations of these facts were made to them, and that they relied thereon. The defendants explain or deny many of the facts relied upon by the plaintiffs, but since this case comes up on motion for summary judgment, 2 we must give the plaintiffs the benefit of every doubt.

The trial judge granted the motion, saying that Wagner v. Manufacturers’ Trust Co., 237 App.Div. 175, 261 N.Y.S. 136, 140, was controlling and that the complaint here is only an “artful attempt” to avoid the ruling of that case. In holding that New Jersey law is the same as New York, the court below was in error, but even if it were correct, we are not sure that the holding of the Wagner case disposes of the case at bar. There, on similar facts, the Appellate Division, affirmed without opinion by the Court of Appeals, 261 N.Y. 699, 185 N. E. 799, held that the Statute of Frauds prevented the substitution of a new party to the lease, and said:

“If, upon the allegations of this complaint, defendant were held equitably es-topped from relying upon the statute of frauds, then in many cases a defendant might be impeded in obtaining the benefit of that statute. The courts would abound with cases where a plaintiff would claim that it might be more equitable to prevent the defendant from pleading and relying upon the statute of frauds, than to enforce that statute, and the statute would be, in large measure, nullified. Every litigant, of course, has as much right to rely upon that statute as upon any other legislative enactment and courts are not prone to deny its protection.”

The opinion hardly comes to grips with the real issue in such cases as these, which the Supreme Court has said to be whether or not ownership of the subsidiary “is resorted to, not for the purpose of participating in the affairs of the corporation in which it is held in a manner normal and usual with stockholders, but for the purpose of making it a mere agent, or instrumentality or department of another company.” United States v. Reading Co., 253 U. S. 26, 62, 63, 40 S.Ct. 425, 434, 64 L.Ed. 760 3 If this is found as a fact, 4 courts have seldom found the Statute of Frauds an obstacle to the imposition of liability, since it does not prevent penetration through an alias to hold the “real” signatory. 5 Even if, however, New York law were applicable (and, as we shall show, it is not), the rule of the Wagner case would not dispose of the case before us.

*347 For, in that case, the court laid stress on the landlord’s knowledge that it was accepting the subsidiary’s signature on the lease; there was, the court said, no concealment by the tenant. Here, however, a jury might well have found a fraudulent concealment by the responsible individuals of the fact that the ostensible tenant, although named “Mursam Shoe Corporation,” was only a leasehold corporation which, contrary to representations deliberately made by the other defendants, was never intended to operate a shoe store. This alone would, in our opinion, create a triable issue.

Furthermore, Schulte Real Estate Co. v. Pedemode, Inc., 241 App.Div. 732, 270 N.Y.S. 919, affirmed 266 N.Y. 550, 195 N.E. 195, decided after the Wagner case, may indicate a retreat from the doctrine there established. Upon similar facts, the New York Supreme Court held in an unreported opinion that the Wagner case could be distinguished and that the plaintiff’s allegations stated a cause of action. No appeal was taken, but thereafter the defendant moved for summary judgment, which was denied. The denial, in the first instance, apparently was based on the theory that, since plaintiff’s examination before trial was incomplete, the moving affidavits were not the best evidence of the facts. Judge Finch’s dissent in the Appellate Division, however, was squarely on the ground that it would be a waste of time to proceed with an examination since the complaint was inadequate, and implicit in the majority’s denial of summary judgment would seem to be a holding that the allegations sufficiently stated a cause of action. The Court of Appeals affirmed without opinion. The allegations in the present case are identical with those in the Pedemode case. As we interpret the results in the Pedemode case, we are constrained to believe that the New York Court of Appeals necessarily held that neither the sealed instrument rule nor the Statute of Frauds barred recovery on the allegations there made. On identical allegations, we should similarly deny summary judgment. Even if the denial of summary judgment in the Pedemode case were, as defendants argue, on the procedural issue alone, the fact would still remain that, at an earlier stage of the same proceeding, a Judge of the New York Supreme Court had held the allegations sufficient. We note also that in Mangan v. Terminal Transportation System, 157 Misc. 627, 631, 284 N.Y.S. 183, 188, the New York Supreme Court, subsequent to the Wagner case, held broadly that “when one corporation controls another, and uses it as the means, agency, and instrumentality by which the former carries out and performs its business, it is liable for the torts of the latter.” In that case, the defendant was under contract with terminals to furnish adequate cab service. It contracted with four ■operating companies, in which it had a majority interest, to operate their cabs out of the terminals designated by it. It furnished them with various supplies and services and supervised the hiring of their drivers. Factually, the subsidiaries were no more under the parent’s thumb than Mursam; furthermore, there was no suggestion that the operating companies were undercapitalized or unable to pay the plaintiffs judgment. If anything, therefore, we should be more ready to hold the defendants before us than the operating companies sued there. 6

We are not, however, bound by New York law in our decision of this case. For the contract, a lease of New Jersey real estate, was made and recorded in New Jersey. 7 While the trial court held that New Jersey followed the doctrine of the Wagner case, the only case cited for this assertion is hardly in point. 8 It would, indeed, be strange if New Jersey did foi *348 low New York law, since even before the harsh doctrine of the Wagner case was established, the leading treatise 9 in this field said that New York law “is probably more conservative than the general run of most jurisdictions,” contrasting Berkey v. Third Avenue R. Co., 244 N.Y. 84, 155 N.E. 58, 50 A.L.R. 599 10 (which refused to impose liability on the parent of a street railway operated as part of a unified system) with Lehigh Valley R. Co. v. Dupont, 2 Cir., 128 F. 840, and Lehigh Valley R. Co. v. Delachesa, 2 Cir., 145 F. 617, in which, on similar facts, this court (in the old untrammeled days before Erie R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188, 114 A.L.R. 1487) imposed liability. New Jersey law is as yet relatively undeveloped, but some clue to the rationale which would be followed by its courts is afforded by Ross v. Pennsylvania Railroad, 1930, 106 N.J.L. 536, 148 A. 741, 743, where an administrator sued a parent corporation for damages caused by the act of its subsidiary. In holding the parent, the Court of Errors and Appeals said that liability must be imposed “where a corporation holds stock of another, not for the purpose of participating in the affairs of the other corporation, in the normal and usual manner, but for the purpose of control, so that the subsidiary company may be used as a mere agency or instrumentality for the stockholding company.” This language echoes that of the United States Supreme Court, quoted above, iñ the Reading case,- and suggests that New Jersey is not as reluctant as New York to hold corporate parents for the acts of their children. The holding of the Ross case is in conflict with the majority opinion in Berkey v. Third Avenue R. Co., supra, but is consistent with the dissent in that case and with the decisions of this court cited above. New Jersey law, therefore, is not identical with New York’s.

Outside of New York, the doctrine hardly seems challenged that a representation that the liability of the subsidiary is the same as thát of its parent or affiliate will result in a piercing of the corporate veil of a subsidiary which. is in fact under-capitalized, dominated and used as an instrument of the parent. 11 Stark Electric R. Co. v. McGinty Contracting Co., 6 Cir., *349 238 F.657; Platt v. Bradner Co., 131 Wash. 573, 230 P. 633; cf. Westinghouse Electric & Mfg. Co. v. Allis-Chalmers Co., 3 Cir., 176 F. 362, 367. Powell (loc.cit., p. 111) says that such a representation is “almost always fatal.” The Wagner case stands alone in disregarding such a representation. We cannot assume that New Jersey would follow it. A more reasonable assumption is that a New Jersey court would impose liability on the parent, as have most other courts, if the jury found as a fact that the parent ran the subsidiary as a division of its own business and represented that the subsidiary’s debts were its own obligations. This would be consistent with the rationale accepted in the Ross case. That same rationale would also indicate that New Jersey would not be troubled by the sealed instrument rule or the Statute of Frauds, since neither — as pointed out above — bars the imposition of liability upon the puppeteer who directed his marionette to sign.

We do not, of course, decide here that Mursam was, in fact, a dummy; we hold only that triable issues were created by plaintiff’s allegations and affidavits that Mursam was a tool of the other defendants, and that the lease was made because of representations, believed by plaintiffs, that its liabilities were the obligations of the other defendants.

The order granting summary judgment and dismissing the complaint is reversed, and the case is remanded.

The general problem before us is extensively dealt with in Douglas and Shanks, Insulation from Liability through Subsidiary Corporations, 39 Yale L.J. (1929) 193; Powell, Parent and Subsidiary Corporations (1931). The pioneer studies of Wormser, Piercing the Veil of Corporate Entity, 12 Col.L.Rev. (1912) 496, and of Machen, Corporate Personality, 12 Harv.L.Rev. (1911) 253, 347, are still of interest.

Various facets of the problem have been recently discussed in Taylor v. Standard Gas & Electric Co., 306 U.S. 307, 618, 59 S.Ct. 543, 83 L.Ed. 669 and in Pepper v. Litton, 308 U.S. 295, 60 S.Ct. 238, 84 L.Ed. 281.

By all defendants except Mursam.

Although the doctrine was there applied, as it had often been before, to a case involving the commodities clause of the Hepburn Act, 34 Stat. 584, 585, 49 U.S.C.A. § 1(8) and the Sherman Act, 15 U.S.C.A. §§ 1-7, 15 note, it has become a classic statement applied generally in cases piercing the corporate veil.

That this is a question for the jury, see Summo v. Snare & Triest Co., 166 App.Div. 425, 428-430, 152 N.Y.S. 29.

The policy of piercing the corporate veil on proper occasions is too strong to permit its circumvention by the defendants’ reliance on the Statute of Frauds or the sealed instrument rule; the holding of the Wagner case, if taken as broadly as the defendant before us urges, would, in effect, destroy any possibility of holding the parent in contract cases.

The only real differences between tort and contract cases in this field are (1) the defense of the Statute of Frauds and the sealed instrument rule, which should not he important, and (2) the possibility that, in contract cases, the plaintiff chose to deal with the subsidiary, as in New York Trust Co. v. Carpenter, 6 Cir., 250 F. 668, 673, 678, and in Roos v. Texas Co., 5 Cir., 43 F.2d 1. These cases are not applicable here, since the court or a jury might, on the trial, accept plaintiffs’ story that they were tricked into accepting the signature of Mursam.

In determining what law applies, we must, of course, follow New York’s principles of conflict of laws. Klaxon Co. v. Stentor Electric Mfg. Co., 313 U.S. 487, 61 S.Ct. 1020, 85 L.Ed. 1477. But however we construe the cause of action here, it falls into the class of cases as to which New York would apply the law of New Jersey.

Beck v. Eagle Brewery of Newark, N.J.Ch.1895, 30 A. 1100. There the landlord sought to hold a corporation not named in the lease merely because it had some undisclosed agreement with the ten *348 ant and because it had paid the rent for several months. It did not occupy the premises, there was no evidence of any connection between it and the person who signed the lease, and the reason why it had paid any of the rent did not appear.

Powell, Parent and Subsidiary Corporations (1931) 72.

Although even that case was regarded as unusually conservative (Powell, supra; 36 Yale L.J. (1927) 878; cf. 27 Col. 702 (1927) 705-706) and stimulated a strong dissent, it certainly did not go nearly so far as the' Wagner case in relieving the parent of liability.- See especially 244 N.Y. 84, at pages 93-95, 155 N.E. 58, 50 A.L.R. 599.

Although some commentators have distinguished between the “agency” and “instrumentality” (or “identity”) rationales for imposing liability upon the parent, few (if any) courts consistently follow one or the other. See Douglas and Shanks, supra, note 1. Normally, the theories are interchangeable, although the “identity” notion is often used to overcome such ancillary problems as the sealed instrument rule, the Statute of Frauds, election of remedies, service of process, and various pleading and practice difficulties. New York seems to have planted its own seeds of confusion i-n this jurisprudential jungle by adding the ground of “estoppel.” See Wagner v. Manufacturers’ Trust Co., supra. Who is estopped by what and against whom, is left in convenient obscurity. Apparently neither a general representation of unified operation (Berkey v. Third Ave. R. Co., supra), nor a specific statement that the parent is liable for its subsidiary’s acts (Wagner v. Manufacturers’ Trust Co., supra) is enough to work an “estoppel,” the New York courts having in this respect repudiated the broad holding of the Quaid case that “the defendant * * * is estopped from denying the truth of his representation to the plaintiffs of the substantial identity of the corporation and himself, and that its obligations were his obligations.” If such commingling and utter domination as was shown in Quaid v. Ratkowsky, 183 App.Div. 428, 170 N.Y.S. 812, 815, constitutes an estoppel, then the notion is only a form of the agency and/or identity rationales. Yet it is cleat that disregard for corporate formalities need not go so far as in the Quaid case. See Mangan v. Terminal Transportation System, supra. (Query: What was the “estoppel” there?). Douglas and Shanks, supra, note 1, acutely point out that court decisions depend on the specific facts of each case rather than on an application of the various theories of legal liability. And in 1911, Machen, supra, note 1, concluded that “no other guide is desirable than sturdy common sense.”