Braden v. University of Pittsburgh

GARTH, Circuit Judge,

concurring.

I concur in Part I of the opinion of the Court, which concerns the power of a district court to vacate and to re-enter a certification order so as to allow an interlocutory appeal. I also concur in the result reached in Part II of that opinion, which, after discussing the issue of state action, leaves the ultimate determination of whether state action exists to the final hearing on the merits. I have written separately on the issue of state action, however, for two reasons. First, I believe several aspects of the opinion of the Court deserve special emphasis. Second, I wish to explain why I disagree with the Court’s analysis of the relevant state action precedents.

I.

In light of the opinion of the district court and the arguments advanced in this appeal by counsel for Dr. Braden, I think it is extremely important to emphasize the limited nature of this Court’s holding.

As the opinion of the Court makes clear, this case is before us on interlocutory appeal from an order of the district court which did no more than to deny Pitt’s motion to dismiss Dr. Braden’s complaint under Fed.R.Civ.Proc. 12(b)(6). Since the dis *966 trict court, in ruling on that motion, considered matters outside the pleadings, Pitt’s motion was converted by Fed.R.Civ.Proc. 12(c) into a motion for summary judgment. On appeal from the denial of that motion, it is obvious that the only question which is properly before us is the following: did the district court err in refusing to enter summary judgment in favor of Pitt on the issue of state action?

Despite the limited nature of the question presented by the denial of Pitt’s motion for summary judgment, counsel for Dr. Braden has called our attention to certain dicta in the opinion of the district court and has urged us to decide a much broader question. Although the district court did no more than to deny Pitt’s motion, its opinion concluded with the categorical statement that “Pitt was acting under ‘color of state law’ when it engaged in the activity challenged by Dr. Braden.” 392 F.Supp. at 126. See also id. at 124 — 25. In this Court, counsel for Dr. Braden has pointed to this language as the “finding” of the district court and has urged us to affirm it. Brief for Appellee at 67. See also id. at 6, 49-50. In effect, therefore, counsel for Dr. Braden has asked this Court to treat the order of the district court as one which not only denied-Pitt’s motion for summary judgment but as one which also granted partial summary judgment in favor of Dr. Braden, even though she did not file a cross-motion and could not have been granted that relief even if she had.

In view of this background, I believe it is important to emphasize that this Court has not treated the order of the district court as one granting partial summary judgment for Dr. Braden but merely as one denying Pitt’s motion for summary judgment. Maj.Op. 949, 954, 955 and 965. This Court has decided only that “the district court did not commit error in declining to dismiss [Dr. Braden’s] complaint for a purported want of state action.” Id. at 965. This Court has not decided either that state action is or is not present in this case. Id. at 955, n.35, 965. That question remains completely open. As I understand it, nothing in the opinion of this Court would preclude the district court on remand from determining either that state action is or is not involved in this case. Nor is there anything in that opinion which would preclude this Court from reaching either result should this case again come before us on appeal from the final judgment of the district court. It goes without saying, therefore, that, as the opinion of the Court notes, our affirmance of the order of the district court may not and should not be taken as any indication that we approve of “certain language in the opinion of the district judge” which “seems to indicate affirmatively that state action exists.” Id. at 955 n.35. ■

Finally, I think that one of this Court’s observations in Braden v. University of Pittsburgh, 477 F.2d 1 (3d Cir. 1973) [Bra-den I], also deserves emphasis. In Braden I, it seems to me, this Court recognized that the state action issue which this case presents is not suited for resolution by means of summary judgment. We concluded by stating:

This is an important case and we are entitled to a carefully prepared record after full hearing in the trial court and to the views of the trial court upon such a record.

(Emphasis added.) Id. at 8 . Accord Weise v. Syracuse University, 522 F.2d 397, 408 (2d Cir. 1975) (relying on Braden I).

I believe that this observation is as true today as it was when Braden I was decided. The Supreme Court has frequently warned that “[o]nly by sifting facts and weighing circumstances can the nonobvious involvement of the State in private conduct be attributed its true significance.” (Emphasis added). Burton v. Wilmington Parking Authority, 365 U.S. 715, 722 , 81 S.Ct. 856, 860 , 6 L.Ed.2d 45 (1961). See also Moose Lodge No. 107 v. Irvis, 407 U.S. 162 , 172, 9S.Ct. 1965, 32 L.Ed.2d 627 (1972). Since a district court, in ruling on a motion for summary judgment, is required to give the party opposing the motion the “benefit of *967 all reasonable doubts and inferences,” 1 the court is not entitled to and cannot “sift” the facts it believes to be important from those it regards are insignificant. Nor can the court attribute to each fact the weight which it believes it deserves. Every fact must be “sifted” and “weighed” against the moving party. As a result, it seems obvious to me that summary judgment procedures are poorly suited for deciding state action questions such as that posed by this case. Where the nub of a state action claim rests upon statutes, case law, or regulations, summary judgment may well be proper. 2 But where, as in this case, the practical working relationship between the state and the private entity is at issue, it seems clear that summary judgment will be proper only in the most obvious cases. This is not such a case.

II.

I believe that it is also important to emphasize that the summary of the relationship between Pitt and the Commonwealth contained in Part IIC of the opinion of the Court must not be read out of context.

At the beginning of its discussion of the state action question, the opinion of the Court recognizes that “since the dismissal motion was submitted by Pitt, Dr. Braden, as the opposing party, is entitled to the benefit of all reasonable doubts and inferences which may arise in connection with the consideration of such motion.” Maj.Op. at 955. As I understand Part IIC of the opinion of the Court, it represents an application of this rule to the facts of this case. As a result, in that portion of its opinion, the Court gives Dr. Braden the benefit of all reasonable doubts and inferences with respect to every fact which bears upon the relationship between the Commonwealth and the University. I believe it is important, therefore, to emphasize that Part IIC of the opinion of the Court must be read only as a summary of that relationship as we are obliged to view it at this stage of the proceedings. At the final judgment stage, the district court may and undoubtedly will be required to view the facts quite differently. In addition, should this case again come before this Court on appeal from the final judgment of the district court, we will then be constrained to consider the facts from a very different perspective. . A few illustrations will make this point clear.

A.

The opinion of the Court, giving Dr. Bra-den the benefit of all reasonable doubts and inferences, apparently attaches considerable significance to the language of Pa.Stat. Ann. Tit. 24, § 2510-202 (Supp.1976), which characterizes Pitt as an “instrumentality of the Commonwealth” and a “State-related institution.” Maj.Op. at 959. At the same time, the Court ignores the testimony of the trustees concerning the effect which this provision has on the actual administration of the University, presumably because that testimony was highly damaging to Dr. Braden. 3 See 392 F.Supp. at 121-22. While this treatment of the evidence is unobjectionable at this stage of the proceedings, at the final judgment stage the district court will obviously have to view the evidence quite differently. It will have to consider both the language of the statute and the testimony of the trustees which Braden I required the district court to take. 4 In addition, since Dr. Braden will *968 not automatically be given the benefit of all reasonable doubts and inferences at that stage, the district court will be free to give the language of the statute and the testimony of the trustees the weight which it believes they deserve.

B.

The opinion of the Court places great emphasis upon the Commonwealth’s authority to appoint one-third of Pitt’s board of trustees (Maj.Op. at 959), while failing to mention the evidence concerning the trustees which Braden I required the district court to assemble. 5 In Braden I, this Court stated:

We cannot decide this case in a semi-vacuum. We do not know, for example, the number of trustees which the University actually has, their names, their tenure of office, and who and what they are, their occupations and connections, and whether or not they are paid for their services or meetings or receive salaries or emoluments, and if so, by whom are these emoluments or salaries paid and what do they amount to.

477 F.2d at 5 .

On remand, the district court took evidence on all these points. That evidence revealed that the backgrounds and occupations of the trustees appointed by the Commonwealth were remarkably similar to those of the other trustees. 6 It is also showed that most of the trustees in both groups were prominent private citizens with few ties to the Commonwealth. In addition, as the district court stated:

4. The voting record of individual trustees is not kept, but, according to the testimony, the trustees appointed by the Commonwealth have never acted in a bloc.

5. All trustees, except the Chancellor, serve without compensation.

392 F.Supp. at 121.

At the final judgment stage, the district court will be required to take all this evidence into account and to give it the weight which it merits. 7

*969 C.

The opinion of the Court concludes that “Pitt may have surrendered much of its fiscal autonomy to the state, with the Commonwealth holding both the purse-strings as well as extensive financial control over the University.” Maj.Op. at 960. It must be emphasized, however, that the Court reaches that conclusion only because it gives Dr. Braden the benefit of all reasonable doubts and inferences, as it is required to do at this stage of the proceedings.

The opinion of the Court, for example, infers that the Auditor General’s authority to review the University’s expenditure of non-state moneys gives him considerable control over the University’s finances. Maj.Op. at 960. At the same time, the Court gives little weight to the testimony in the district court which suggested that the Auditor General “reviewed” the expenditure of Pitt’s private funds only in order to determine how much the Commonwealth should contribute. Appendix at 101a. At the final judgment stage, the district court will be required to weigh all this evidence and to make a finding on this disputed issue of fact. 8

The opinion of the Court also infers from the fact that Pitt is required to make various annual reports to the Commonwealth that the Commonwealth exercises significant control over Pitt’s finances. Maj.Op. at 960. At the final judgment stage, however, when Dr. Braden will not automatically be given the benefit of all reasonable doubts and inferences, the district court will be required to weigh the evidence and to make findings with respect to whether the Commonwealth has utilized these reporting requirements to “control” Pitt. 9

The opinion of the Court fails to give any significance to Pa.Stat.Ann., tit. 24, § 2510— 205 (Supp.1976), which vests the “entire management, control and conduct of the . financial affairs of the university in the board of trustees.” Maj.Op. at 960 n.59. It also fails to mention the testimony in the district court which revealed that the Commonwealth places no restrictions on Pitt’s freedom to use its private funds as it sees fit. Appendix at 478a-79a. At the final judgment stage, this evidence will obviously have to be taken into account.

Finally, the opinion of the Court places heavy reliance upon the percentage of Pitt’s annual budget which the Commonwealth contributes. While I believe that it is permissible to rely upon that factor at this stage of the proceedings, at the final judgment stage the district court will have to consider this Court’s twin warnings in Hollenbaugh v. Carnegie Free Library, 545 F.2d 382 (3d Cir. 1976). In Hoilenbaugh, this Court cautioned that “ ‘receipt of mon *970 ey from the State is not, without a good deal more, enough to make the recipient an agency or instrumentality of the Government.’ ” Id. at 385 , quoting Grossner v. Trustees of Columbia University, 287 F.Supp. 535, 547-48 (S.D.N.Y.1968). Accord, Maj.Op. at 960 n.58. This Court also warned that state action questions ought not to turn upon the precise percentage of an institution’s budget which is contributed by the state. Hollenbaugh v. Carnegie Free Library, supra, at 384-85. At the final judgment stage, these two principles will, in my view, preclude extensive reliance upon the percentage of Pitt’s operating budget contributed by the Commonwealth. 10

In short, it must be stressed that the Court’s conclusion with respect to the financial control which the Commonwealth exercises over the University will not necessarily be valid at subsequent stages of this litigation. That conclusion as well as all the factual statements contained in Part IIC must not and cannot be read out of the context of a 12(b)(6) or summary judgment motion.

III.

While I agree with the majority of the Court that the district court was correct in denying Pitt’s motion for summary judgment (see Part IV of my opinion), I do not agree with the state action standards employed by the Court in reaching that conclusion. I therefore feel compelled to express the reasons for my disagreement.

A.

I am, to be frank, rather perplexed by the Court’s analysis of the state action standards which apply in this case. After identifying Burton v. Wilmington Parking Au *971 thority, 365 U.S. 715 , 81 S.Ct. 856 , 6 L.Ed.2d 45 (1961), and Jackson v. Metropolitan Edison Co., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974), as the two most pertinent Supreme Court precedents and after summarizing the holdings of those two cases (Maj.Op. at 956 — 957), the Court observes that Burton was not overruled by Jackson. Id. at 957. The Court then virtually ignores Jackson and proceeds to apply Burton to the facts of this case. I believe there are two flaws in the majority’s analysis.

First, I think the Court has greatly oversimplified the evolution of the Burton theory of state action by flatly equating Burton with the “overall relationship” test. The majority constructs that equation by focusing upon a single sentence in the Burton opinion. 11 But the Burton Court stated specifically and at some length that it was not announcing a state action “test” of general applicability. 365 U.S. at 725-26 , 81 S.Ct. 856 . More importantly, the facts in Burton were such that state action could certainly have been found under either the “overall relationship” test or the “nexus” test. While the state and the restaurant undoubtedly had a close overall relationship, the state was also deeply implicated in the very acts of the restaurant which were challenged, i. e., its refusal to serve blacks. Burton states that “it [cannot] be ignored especially in view of Eagle’s affirmative allegation that for it to serve Negroes would injure its business, that profits earned by discrimination not only contribute to, but also are indispensable elements in, the financial success of a governmental agency.” Id. at 724 , 81 S.Ct. at 861 . As a result, it is not at all “apparent that the dispositive factor in Burton . . . was the extent and nature of' the overall relationship between the state agency and the private enterprise.” Maj.Op. at 957. 12

The second flaw in the majority’s analysis is its failure to consider the gloss which subsequent opinions have placed upon Burton . The majority opinion, after noting that Burton has not been overruled by subsequent cases, apparently leaps to the conclusion that Burton can now be applied without even considering the interpretations of Burton contained in more recent Supreme Court opinions. Although Moose Lodge No. 107 v. Irvis, 407 U.S. 163 , 92 S.Ct. 1965 , 32 L.Ed.2d 627 (1972), and Jackson both discussed Burton at length, the *972 majority for the most part ignores that discussion. I must take exception to this approach. We cannot turn the clock back 16 years and read Burton as if Moose Lodge and Jackson had never been decided. Any discussion of Burton in 1977 must consider the extent to which subsequent Supreme Court cases have clarified or modified the meaning of that decision. I believe that a brief examination of those cases reveals that the question which Burton left open— whether the “overall relationship” test or the “nexus” test is the correct one — has now been answered in favor of the “nexus” standard.

*971 In Jackson , the Supreme Court suggested that the “actual holding” of Burton might be limited to lessees of public property. Even so, as previously indicated, many of the buildings within the Pitt complex were built by the state or its agencies with public funds and are leased to the University, including the very building within which Dr. Braden worked. It follows that even if Burton and its “relationship” theory of state action should be limited to a “lease” situation, the present appeal still may be governed by the teachings of that landmark decision.

*972 The first Supreme Court case to analyze Burton closely was Moose Lodge No. 107 v. Irvis, 407 U.S. 163 , 92 S.Ct. 1965 , 32 L.Ed.2d 627 (1972), in which the Court reversed Irvis v. Scott, 318 F.Supp. 1246 (M.D.Pa.1970) (three-judge court). 13 Irvis v. Scott, supra, had relied upon Burton in holding that the acts of racial discrimination of a private club were infused with state action. Writing for the three-judge court, Judge Freedman had noted that Pennsylvania had granted the club a liquor license and had regulated its activities extensively. He had concluded that “[i]t would be difficult to find a more pervasive interaction of state authority with personal conduct.” 318 F.Supp. at 1250 . As a result, the district court had held that “[a]s in Burton the state has ‘insinuated itself into a position of interdependence’ with its club licensees . .” Id. at 1251. In reversing the district court, the Supreme Court did not dispute the pervasive nature of the “overall relationship” between the state and the club. The Court wrote:

The District Court was at pains to point out in its opinion what it considered to be the “pervasive” nature of the regulation of private clubs by the Pennsylvania Liquor Control Board. .

However, detailed this type of regulation may be in some particulars, it cannot be said to in any way foster or encourage racial discrimination. .

We therefore hold that, with the exception hereafter noted, the operation of the regulatory scheme enforced by the Pennsylvania Liquor Control Board does not sufficiently implicate the State in the discriminatory guest policies of Moose Lodge to make the latter “state action” within the ambit of the Equal Protection Clause of the Fourteenth Amendment. (Emphasis added.)

Id. at 176-77, 92 S.Ct. at 1973 . If, as the majority suggests, state action can now be found based solely upon a pervasive “overall relationship,” then the Supreme Court’s decision in Moose Lodge is hard to understand. 14

In Gilmore v. City of Montgomery, 417 U.S. 556 , 94 S.Ct. 2416 , 41 L.Ed.2d 304 (1974), the Court stated flatly that the Burton test requires a nexus between the state and the very activity challenged. In Gilmore the Court affirmed an order enjoining the city from allowing private segregated schools to have the exclusive use of municipal recreational facilities. Discussing Burton and Moose Lodge in his opinion for the Court, Justice Blackmun wrote:

*973 Because the city makes city property available for use by private entities, this case is more like Burton than Moose Lodge. The question then is whether there is significant state involvement in the private discrimination alleged. Reitman v. Mulkey, 387 U.S. 369 , 87 S.Ct. 1627 , 18 L.Ed.2d 830 (1967); Burton v. Wilmington Parking Authority, supra; Evans v. Newton, 382 U.S. 296 , 86 S.Ct. 486 , 15 L.Ed.2d 373 (1966); Moose Lodge No. 107 v. Irvis, supra. (Emphasis added.)

The Court’s most extensive discussion of Burton occurred in Jackson v. Metropolitan Edison Co., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974), which held that state action was not present when a privately owned and operated utility company terminated a customer’s electric service without giving her notice, a hearing, or an opportunity to pay amounts due. Jackson reached this conclusion despite the fact that the state was involved “in virtually every phase of the company’s business.” 419 U.S. at 368 , 95 S.Ct. at 462 (Marshall, J., dissenting). In .addressing the Burton theory of state action, Jackson first pointed out the factual distinctions between that case and Burton. Then, after summarizing the substantial connections between the state and the utility, it concluded:

Under our decision this is not sufficient to connect the State of Pennsylvania with respondent’s action so as to make the latter’s conduct attributable to the State for purposes of the Fourteenth Amendment. (Emphasis added.)

Id. at 358, 95 S.Ct. at 457 . Jackson also observed that the proper “inquiry must be whether there is a sufficiently close nexus between the State and the challenged action of the regulated entity so that the action of the latter may be fairly treated as that of the State itself.” (Emphasis added.) Id. at 351 , 95 S.Ct. at 453 .

In conclusion, it appears to me that Moose Lodge, Gilmore, and Jackson place a gloss upon Burton which we are not free to ignore. The teaching of those cases is that “neither general government involvement nor even extensive detailed state regulation is sufficient for a finding of state action. Rather, the state must affirmatively support and be directly involved in the specific conduct which is being challenged.” (Footnote omitted.) Cannon v. University of Chicago, Nos. 76-1238/9 (7th Cir., filed August 27,1976) at 5.

B.

With the exception of Hollenbaugh v. Carnegie Free Library, supra, which is discussed below, this Court’s recent state action opinions have all employed the “nexus” test rather than the “overall relationship” approach. Broderick v. Associated Hospital Service of Philadelphia, 536 F.2d 1, 4-6 (3d Cir. 1976) (per Garth, J.); Magill v. Avonworth Baseball Conference, 516 F.2d 1328, 1332-33 (3d Cir. 1975) (per Aldisert, J.); Gibbs v. Titelman, 502 F.2d 1107, 1111-12 (3d Cir. 1974) (per Hunter, J.); Jackson v. Metropolitan Edison Co., 483 F.2d 754, 756 (1973) (per Weis, J.), aff’d., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974).

In Broderick this Court stated: “For state action to be found where the impetus for the discrimination is private, the record must demonstrate that the state has become significantly involved with the discriminatory action alleged.” (Emphasis added.) 536 F.2d at 4 . Broderick never even referred to the “overall relationship” test.

In Magill we grouped the Supreme Court’s state action cases into three general categories. 516 F.2d at 1331 . Magill placed Burton in the second group — cases in which “the state ‘significantly’ involved itself with the private party.” Id. at 1331-32. Describing the test applicable to cases falling into the second group, Magill stated that “the polestar of our analysis must be whether the state involvement in the challenged action is ‘significant.’ ” (Emphasis added.) Id. at 1332 . Magill refers to the “nexus” test [Id.], but it never even mentions the “overall relationship” test. Gibbs and this Court’s opinion in Jackson treated the issue of state action similarly. The majority makes no attempt to reconcile its *974 state action analysis with the language or the holdings in Broderick, Magiil, Gibbs, or this Court’s opinion in Jackson.

In short, I believe that the recent state action cases decided by this Circuit, as well as those decided by the Supreme Court, all establish that in order for state action to be found “the record must demonstrate that the state has become significantly involved with the discriminatory action alleged.” (Emphasis added.) Broderick v. Associated Hospital Service of Philadelphia, supra, at 4. In most cases, it will be impossible to conclude that state action is present under this standard unless there is proof which connects the state with the very activity challenged. However, it seems obvious to me that such proof would not be necessary in one limited situation, viz., where the state’s involvement with a private entity is “so dominant as to afford basis for a contention that the state is merely utilizing private trustees to administer a state activity.” Powe v. Miles, 407 F.2d 73, 81 (2d Cir. 1968). See also Cohen v. Illinois Institute of Technology, 524 F.2d 818 , 825 & n.19 (7th Cir. 1975) (per Stevens, J.); Wahba v. New York University, 492 F.2d 96, 101 (2d Cir. 1974); Kerr v. Enoch Pratt Free Library of Baltimore, 149 F.2d 212 (4th Cir. 1945). In a situation of that nature, I believe that state involvement in the challenged activity could simply be assumed.

This Court recently considered a situation of this type in Hollenbaugh v. Carnegie Free Library, supra, in which a library received approximately 90 per cent of its financial support from the Commonwealth of Pennsylvania and its subdivisions and in which a large segment 15 of the library’s board of trustees was composed of government officials. In that case this Court reviewed a district court order which dismissed a complaint raising Fourteenth Amendment claims for failure to allege facts constituting state action. In reversing, this Court suggested that the extreme state involvement which was alleged in that case could without more constitute state action. It went on to say that “it is impossible to distinguish the state’s involvement in the allegedly discriminatory employment determination from the state’s involvement in the general operation of the library.” Id. at 385. Hollenbaugh is consistent with the analysis above. I read it to say that where a state has funded an institution to the extent of almost 100 per cent of its financial requirements, it is reasonable to assume that with such massive funding the state must be chargeable with all activities undertaken by the institution. However, short of the type of involvement alleged in Hollenbaugh , I believe that proof of state involvement in the very activity challenged is required.

In sum, I interpret the Supreme Court’s and this Circuit’s recent state action decisions to require either (a) state involvement with a private entity which is “so dominant as to afford basis for a contention that the state is merely utilizing private trustees to administer a state activity” 16 and therefore has necessarily participated in the challenged activity, or (b) proof that* the state has affirmatively supported or is directly involved in the specific conduct challenged.

C.

I must also express my disagreement with the discussion contained in Parts II D and E of the opinion of the Court. Those sections, in my view, constitute a wholly inadequate analysis of the manner in which other federal courts have dealt with state action claims lodged against private universities.

First, the Court’s heavy reliance upon Isaacs v. Board of Trustees of Temple University, 385 F.Supp. 473 (E.D.Pa.1974), appears to me to be misplaced. As the Court notes, Isaacs was decided prior to the Supreme Court’s decision in Jackson. There is good reason to believe that the analysis in *975 Isaacs would have differed significantly had that case been decided after Jackson. The analysis in Isaacs rests on the premise that “[t]he leading case in the entire ‘state action’ area is Burton v. Wilmington Parking Authority. . . 385 F.Supp. at 487 . In light of Jackson, the notion that Burton is the “leading case in the entire ‘state action’ area” is plainly untenable.

Second, the Court’s treatment of Powe v. Miles, 407 F.2d 73 (2d Cir. 1968), simply will not bear analysis. In Powe the Second Circuit held that the suspension of students enrolled in Alfred University’s four private colleges was not state action but that the suspension of students enrolled at the New York State College of Ceramics at Alfred University, a state college affiliated with Alfred, was action under color of state law. The Court attempts to analogize the relationship between the Commonwealth and Pitt to the relationship between the State of New York and the New York State College of Ceramics. In doing so, the Court neglects to mention several significant facts. Most important, the Court fails to note that New York State paid 100 per cent of the direct expenses of the New York State College of Ceramics. 17 Id. at 76 . The Court also fails to note that the New York State College of Ceramics, which was founded as a state institution, 18 remains a “state school” under New York law, 19 as distinguished from a “state-related institution” such as Pitt. 20

Although the Court relied upon Powe to support its state action analysis, that case may actually be quite damaging to the Court’s position. In Powe , the Second Circuit held that the suspension of students enrolled in Alfred’s private colleges was not state action despite the fact that during the year in question the State of New York paid 20.75 per cent of the total Alfred budget. 407 F.2d at 76 . As one commentator has noted, Powe also concluded that the suspension of the Alfred students was not state action even though “the state paid about a third of the salary of the Dean of Students who imposed the suspension.” (Emphasis in original). O’Neil, Private Universities and Public Law, 19 Buffalo L.Rev. 155, 160 (1970). In addition, the state action test employed in Powe by Judge Friendly was a precursor of the “nexus” standard enunciated in Moose Lodge and Jackson. In Judge Friendly’s words, under his test:

The state must be involved not simply with some activity of the institution alleged to have inflicted injury upon a plaintiff but with the activity that caused the injury. Putting the point another way, the state action, not the private action, must be the subject of complaint.

407 F.2d at 81 . This ’ test, as Professor O’Neil has noted, “put aside [Alfred’s] interdependence and close involvement with the state.” O’Neil, supra, at 159.

Third, I agree with the Court that Cohen v. Illinois Institute of Technology, 524 F.2d 818 (7th Cir. 1975) (per Stevens, J.), is highly relevant, but I believe that Judge (now Justice) Stevens’s analysis of the appropriate state action standards is exactly the same as mine and consequently at odds with the view expressed by the majority of the Court. In Cohen , Judge Stevens employed a two-part state action test. As the Court acknowledges, one part of his formulation was the “nexus” test of Jackson. Maj.Op. at 965. The second part of the test, as the Court notes, is whether the ostensibly private institution is “so heavily dependent on the State as to be considered the equivalent *976 of a public university for all purposes and in all its activities.” 524 F.2d at 825 , quoted in Maj.Op. at 965. To explain what he meant by that statement, Judge Stevens cited Judge Friendly’s observation in Powe v. Miles, supra, at 82 , that state action could be found, even absent a “nexus” between the state and the specific conduct challenged, “where the wholly state-supported activity is so dominant that the private activity could be deemed to have been swallowed up.” 524 F.2d at 825 n.19, quoting Powe v. Miles, 407 F.2d 73, 82 (2d Cir. 1968). See Hollenbaugh v. Carnegie Free Library, supra. This language was simply a restatement by Judge Friendly of the point he had made earlier that state action could be found where the state’s involvement with the private entity was “so dominant as to afford basis for a contention that the state is merely utilizing private trustees to administer a state activity.” 407 F.2d at 81 . As I noted earlier, in a case of that nature, I agree that proof of a nexus need not be shown. But that is quite different from saying, as the Court does, that a nexus need not be shown whenever there is simply a close overall relationship between the state and the private entity.

Fourth and finally, I must disagree with the Court for failing to appreciate the relevance of our sister Circuit’s recent state action cases involving private universities. It seems to me that, in addition to Cohen v. Illinois Institute of Technology, supra, the following cases are particularly pertinent: Cannon v. University of Chicago, Nos. 761238/9 (7th Cir., filed August 27,1976) at 5; Berrios v. Inter American University, 535 F.2d 1330 (1st Cir. 1976); Greenya v. George Washington University, 167 U.S.App.D.C. 379 , 512 F.2d 556 (1975); Spark v. Catholic University of America, 167 U.S.App.D.C. 56 , 510 F.2d 1277 (1975). While I will not discuss these cases at length, their general perspective can perhaps best be summarized by the First Circuit’s words in Berrios : “we approach with some caution any labelling that might incidentally contribute to the erosion of the autonomy and diversity of private colleges and universities.” 535 F.2d at 1333 .

IV.

Although I disagree with the state action analysis contained in the opinion of the Court, I believe that affirmance of the order of the district court is required even under the correct state action “nexus” standard described in Part III supra. I reach this conclusion because I believe that genuine issues of material fact exist with respect to the state action requirements. In addition, as I stated above, I am convinced that summary judgment procedures are poorly suited for deciding state action questions when the practical rather than the formal relationship between the state and the private entity is at issue. When the practical working relationship is at issue, it seems obvious that, except in the most clear-cut instances, the issue of state action should not be decided until after the district court has made its required findings of fact. For example, the district court opinion summarizes some of the evidence which has already been introduced but as to which it has made no findings. That testimony in part is:

6. Concerning the views of the trustees respecting the import of the word “instrumentality” in 24 P.S. § 2510-202(6), the following testimony is typical:

“I do not regard the University as an agency of the State. I consider it purely autonomous. I don’t believe the State has any role in any of the personnel practices of the University.” (Harbaugh Miller, Esq., former Commonwealth Trustee. Tr. 205).

“I do not view the University as an agency of the State, nor do I view the State as having any role in dealing with the management of the University. I view this as the prerogative of the Board of Trustees, and the Chancellor who is appointed by that Board of Trustees.” (Alexandria W. Chauss, Alumni Trustee. Tr. 160).

“I think we look on our relationship with the State as a contractor for services to the State for which the University is compensated and for which the Universi *977 ty has some responsibility or has — let me correct that — has responsibility for accounting to the State for state funds, as would any contractor.” (William H. Rea, Charter Trustee, Chairman of the Board of Trustees, member of the State Board of Education, Chairman of the State Council of Higher Education. Deposition 20).

“Well, I have always considered that relationship both as Chancellor of the University and as the Secretary of Education, that the University is primarily a private institution, with a relationship to the State under which they agreed to meet certain requirements as far as giving Pennsylvania residents, for example, for admission, and having State representation on the Board of Trustees. . [T]he instrumentality is that, as I interpret it, it is a commitment that the University has made to perform certain services for the State.” (Dr. David H. Kurtzman, former Pitt Chancellor and Commonwealth Secretary of Education. Tr. 46).

Dr. Kurtzman further testified that the Commonwealth does not exercise any control over the hiring, promotion, tenure, or firing of either faculty members or other personnel. (Kurtzman, Tr. 48-49, 63-64). Similarly, Warren E. Ringler, Commonwealth Deputy Commissioner for Higher Education, testified that the Department of Education does not review individual Pitt faculty appointments. (Ringler, Deposition 77; Tr. 22).

It is evident to me that on remand the district court will be hard pressed to find state involvement if limited to the testimony taken to date 21 — which of course it will not be. The testimony to date may be supplemented or possibly contradicted. But at least on this record with the showing heretofore made by the defendants, it would appear unlikely that the required state involvement can be found — thus revealing the wisdom of making no state action determination until the entire record is closed.

I must therefore agree that the district court did not err when, on this record and at this stage of the proceeding, it denied Pitt’s motion for summary judgment.

. 10 C. Wright and A. Miller, Federal Practice and Procedure § 2725 at 510 (1973).

. E. g., Jackson v. Metropolitan Edison Co., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974); Moose Lodge No. 107 v. Irvis, 407 U.S. 163 , 92 S.Ct. 1965 , 32 L.Ed.2d 627 (1972); Broderick v. Associated Hospital Service of Philadelphia, 536 F.2d 1 (3d Cir. 1976).

. The essence of the trustees’ testimony was that in practice Pa.Stat.Ann. Tit. 24, § 2510-202 (Supp.1976) did not convert Pitt into an agency of the state; that the statute did not give the Commonwealth any role in the management of the University or in its personnel policies; and that the provision’s true meaning was simply to make Pitt “a contractor for services to the State for which the University is compensated.” 392 F.Supp. at 121-22.

. In Braden I, this Court was aware of Pa.Stat. Ann. Tit. 24, § 2510-202 (Supp.1976). Nevertheless, we remanded because we felt that “the *968 views of the trustees respecting- this unusual provision, i. e., how they interpret it — might be of great value in resolving the difficult issue presented.” Id. at 7. We believed, in other words, that the interpretation placed upon this provision by the persons who govern the University ought to be considered in ascertaining the statute’s practical effect.

. The opinion of the Court relies to some extent upon the fact that the Governor, the Mayor of Pittsburgh, and the Secretary of Education of the Commonwealth are ex officio members of Pitt’s Board of Trustees. I am not sure that that reliance is justified even at this stage of the proceedings. The record shows that these officials very rarely attended trustees’ meetings. Appendix at 314a, 445a-46a. More important, it is a very old and very common practice for private universities to make one or two high state officials ex officio trustees. As Braden I noted, the distinction between public and private educational institutions under American law dates from Trustees of Dartmouth College v. Woodward, 17 U.S. (4 Wheat.) 518 , 4 L.Ed. 629 (1819). 477 F.2d at 6 . See generally Friendly, The Dartmouth College Case and the Public Private Penumbra, 12 Texas L.Q. 9 (1969). It is therefore interesting to note that the original charter granted to Dartmouth College in 1769 by King George III named as trustees the colonial governor and several other New Hampshire officials. 17 U.S. (4 Wheat.) at 523-24 .

. Of the 16 Commonwealth appointees who served between 1966 and 1976, six were corporate executives, one was general counsel to the United Steel Workers, four were privately employed professionals, one was a retired state judge, one was an active state judge, and three were members of the legislature, Supp.App. at 144a. This group is demographically quite similar to the group of trustees elected by the Board. For example, the 11 persons chosen as term trustees during the same period included four corporate executives, one lawyer in private practice, one doctor, the president of a private college, one judge, one Congressman, and one retired governor. Id. at 144aa-45aa.

. It would appear to me that on remand the district court should compare the statute which authorizes the Commonwealth to appoint one-third of the members of Pitt’s Board of Trustees with several provisions in the By-Laws of Blue Cross of Greater Philadelphia which were called to our attention in Broderick v. Associated Hospital Service of Philadelphia, 536 F.2d 1 (3d Cir. 1976). Under those by-laws, various Pennsylvania public officials are empowered to nominate seven of the 34 members of the Board of Directors of Blue Cross of Greater Philadelphia. By-Laws of Blue Cross of Great *969 er Philadelphia § 3.04(a). In addition, the bylaws require under the usual circumstances that at least two of the seven members of the organization’s executive committee be nominated by public officials. Id. at § 3.17(a). The executive committee is empowered to “exercise all of the powers and authority of the Board of Directors” with a few exceptions. Id. at § 3.17(b). All of these provisions were called to the Court’s attention in Broderick . Brief for Amicus Curiae American Civil Liberties Foundation of Pennsylvania in Broderick v. Associated Hospital Service of Philadelphia at exhibit A at 5. Nevertheless, tested against the requirements of Jackson v. Metropolitan Edison Co., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974), we found them insufficient to show the presence of state action. Broderick v. Associated Hospital Service of Philadelphia, supra, at 4 n.15.

. In my view, the district court on remand should compare the Commonwealth’s authority to audit Pitt with its extensive authority to audit Blue Cross [Pa.Stat.Ann., tit. 40, § 6125(b) (Supp.1976)] and public utilities such as the Metropolitan Edison Company [Pa.Stat. Ann., tit. 66, §§ 1211-18 (1959)]. In Broderick and in Jackson v. Metropolitan Edison Co., 419 U.S. 345 , 95 S.Ct. 449 , 42 L.Ed.2d 477 (1974), aff’g. 483 F.2d 754 (3d Cir. 1973), neither this Court nor the Supreme Court felt that the Commonwealth’s authority to audit the private entities indicated the presence of state action.

. It would appear to me that on remand the district court will have to compare these requirements with the detailed reporting requirements which Pennsylvania law imposes upon Blue Cross. Pa.Stat.Ann., tit. 40, § 6125(a) (Supp.1976). See Broderick v. Associated Hospital Service of Philadelphia, supra.

. Without these two salutary principles, federal courts would soon be drawn into a morass. First, courts would inevitably be required to determine precisely how much state aid is permitted before state action can be found. The District of Columbia Circuit has already held that governmental aid totalling approximately 25 per cent of a private university’s budget is not by itself enough to implicate the government in the actions of the institution. Spark v. Catholic Univ. of America, 167 U.S.App.D.C. 56 , 510 F.2d 1277 (1975). If Pennsylvania’s annual aid to Pitt — approximately 35 per cent of the University’s operating budget — is viewed as strong evidence of state action, that holding together with Spark would establish the maximum amount of aid permitted at somewhere between 25 and 35 per cent. It would not be long before courts would be required to determine the precise percentage where the critical line would be drawn.

Second, if the percentage of an institution’s budget contributed by the government were to be accorded that significance, courts would soon be required to scrutinize the institution’s financial affairs in order to ascertain the relevant figures for each relevant category. Courts would be required to decide whether to focus upon an institution’s operating budget, its capital grants, or its “hard core” budget, see Rackin v. Univ. of Pa., 386 F.Supp. 992, 996-97 (E.D.Pa.1974), and whether to include state aid given directly to students or their parents as part of their calculations. Cf. Comm. for Public Education v. Nyquist, 413 U.S. 756 , 93 S.Ct. 2955 , 37 L.Ed.2d 948 (1973) (granting tuition reimbursements and tax benefits to parents of students attending church-related schools implicates the state in the religious activities of those schools). For a description of state aid to college students and the benefits such aid provides for private colleges, see Carnegie Commission on Higher Education, Higher Education: who pays? who benefits? who should pay? 183-85 (1973); O’Neil, Private Universities and Public Law, 19 Buffalo L.Rev. 155, 174-75 (1970).

Finally, while this case concerns state aid, significant reliance upon that assistance at the final judgment stage would raise serious questions with respect to the federal government’s massive payments to major private universities. For a number of years, those payments have averaged more than 40 per cent of the annual budgets of the recipient institutions. W. Bowen, The Economics of the Major Private Universities 35 (1968). See also Grossner v. Trustees of Columbia Univ., supra at 547 n.17 (federal aid totalled approximately 42 per cent of budget for fiscal year 1967). Federal assistance, of course, is not evidence of action “under color of’ state law and would not be relevant in an action under 42 U.S.C. § 1983 . But sufficient federal involvement in private activity which violates constitutional rights forms the basis for a claim against the private entity for equitable relief and possibly for money damages. Greenya v. George Washington Univ., 167 U.S.App.D.C. 379 , 512 F.2d 556 , 562-63 n.13 (1975); Spark v. Catholic Univ. of America, 167 U.S.App.D.C. 56 , 510 F.2d 1277 (1975).

. Maj.Op. at 957. The sentence in Burton stated:

The State has so far insinuated itself into a position of interdependence with Eagle that it must be recognized as a joint participant in the challenged activity, which, on that account, cannot be considered to have been so “purely private” as to fall without the scope of the Fourteenth Amendment.

365 U.S. at 725 , 81 S.Ct. at 862 . This sentence is hardly an unambiguous adoption of the overall relationship test. While the phrase “insinuated itself into a position of interdependence” may suggest that approach, the phrase “joint participant in the challenged activity” equally may be read as the “nexus” test.

. The Court does acknowledge at one point that Jackson may have greatly limited Burton’s scope:

Maj.Op. at 962. This passage is cross-referenced to footnote 64, which states in relevant part:

The record indicates that Dr. Braden worked primarily in Salk Hall during her employment at Pitt, a building apparently owned by the Commonwealth and renovated with G.S.A. funds. See Brief for Appellee at 45. The Supplemental Appendix at 227aa contains an exhibit which specifically lists the renovation of Salk Hall as a G.S.A. project.

Maj.Op. at 961 n.64.

Despite these statements in the opinion of the Court, the portions of the record which have been brought to our attention show only that a renovation of Salk Hall was begun in 1961 —five years before the enactment of the Pittsburgh-Commonwealth Act. Supplemental Appendix at 227aa. These facts would seem to cast some doubt upon the ability of Dr. Braden to prevail on the issue of state action if Burton is limited to “lease” situations.

. Supreme Court opinions announced prior to Moose Lodge do not discuss Burton in any detail and do not suggest that Burton adopted an “overall relationship” approach.

. In Columbia Broadcasting System, Inc. v. Democratic National Committee, 412 U.S. 94 , 93 S.Ct. 2080 , 36 L.Ed.2d 772 (1973) (holding that broadcasters are not required to accept paid editorial advertisements), Chief Justice Burger, joined by two other members of the Court, concluded that a broadcast licensee’s refusal to accept editorial advertisements was not state action. The Chief Justice wrote:

In dealing with the broadcast media, as in other contexts, the line between private conduct and governmental action cannot be defined by reference to any general formula unrelated to particular exercises of governmental authority. When governmental action is alleged there must be cautious analysis of the quality and degree of Government relationship to the particular acts in question. “Only by sifting facts and weighing circumstances can the nonobvious involvement of the State in private conduct be attributed its true significance.” Burton v. Wilmington Parking Authority, 365 U.S. 715, 722 , 81 S.Ct. 856, 860 , 6 L.Ed.2d 45 (1961). (Emphasis added.)

Id. at 115, 93 S.Ct. at 2093.

. The opinion reveals that 11 of the trustees were government officials, but it does not indicate the total number of trustees in office at that time. A total of 24 trustees was authorized, but there were at least four vacancies. Hol-Ienbaugh v. Carnegie Free Library, supra, at 384.

. Powe v. Miles, supra, at 81 .

. In addition, New York State paid Alfred a fixed amount per credit hour for courses which ceramics students took in one of the University’s private branches. 407 F.2d at 76 . The state also reimbursed Alfred for the ceramics school’s pro rata share of the University’s overall administrative expenses. Id. Thus, New York’s funding of the New York State College of Ceramics was of the same character and as extensive as the public funding of the library involved in Hollenbaugh v. Carnegie Free Library, supra, which is discussed above.

. Powe v. Miles, supra, at 75 .

. New York Education Law § 6101 (McKinney 1972).

. Pa.Stat.Ann. Tit. 24, § 2510-202 (Supp. 1976).

. In upholding the district court’s dismissal of Pitt’s motion, the majority relies almost exclusively on Pennsylvania’s legislative enactments, ignoring almost wholly the testimony and evidence introduced at the hearing which we ordered in Braden I. As I have observed, however, these statutes had been enacted and were in force long before Braden I.

I believe as the panel in Braden I obviously believed, that a good deal more than mere reference to the Commonwealth’s statutes will be required to determine the existence of the necessary “nexus” and other mandated factors for a finding of state action. Doubtlessly the Braden I panel had in mind the determination of this issue on a final record with evidence and testimony fleshing out the legislative intent, and with findings of fact made by the district court. (It could not have believed otherwise for, as we have noted, the statutes in issue were the same both before and since Braden I). This circumstance, if no other, bolsters my views which are expressed here that the state action issue posed by this case should not be made on less than a complete record — and certainly not on a naked reading of inconclusive statutory enactments with no factual context on which findings of the court can be predicated.