One of the central factors stirring the economic maelstrom which has plagued the United States in recent years has been the inability of major American industries to compete with products produced more efficiently and economically by foreign competitors. A recent study of the American steel industry reveals major difficulties experienced by it because of high import penetration by Japanese, West German, and other foreign steel producers.1 Resulting unemployment and the bleak prospects for future employment in the steel industry prompted a Presidential Task Force in 1977 to recommend federal assistance through the Public Works and Economic Development Act of 1965 (PWEDA or the Act), 42 U.S.C. § 3121 et seq. (1976), to assist struggling domestic steel producers in modernizing their steelmaking capabilities and improving their potential for competing with cheaper foreign imports. 43 Fed.Reg. 16360.

The Economic Development Administration (EDA), the agency charged with administering grants under PWEDA, has proposed to guarantee $29,039,400 of the rental obligation to be incurred by Phoenix Steel Corporation (Phoenix) under a leveraged lease of capital equipment to be used in the modernization of its steel plate production facilities. Phoenix had been experiencing severe financial difficulties. Lukens Steel Company (Lukens), a competitor, filed an action against EDA in the district court to set aside the agency's decision to guarantee the loan. It alleged that this decision violated various provisions of PWEDA and its implementing regulations. The district court granted Phoenix's motion to intervene and thereafter granted the defendants' motions for summary judgment. We reverse.

The PWEDA was passed by Congress in 1965 to provide "areas and regions of substantial and persistent unemployment and underemployment (with federal assistance) to take effective steps in planning and financing their public works and economic development." 42 U.S.C. § 3121. In response to a growing concern over large layoffs in the steel industry, a Presidential inter-agency task force in December, 1977 recommended federal assistance for the purpose of assisting the industry. EDA, pursuant to its authority and with funds provided by PWEDA, thereupon established a "special program to guarantee loans and leases financed by private lending institutions to firms in the basic steel industry." This program was viewed as a necessary vehicle for modernizing the facilities of domestic steel producers, waging battle with cheaper foreign imports and aiding the steel industry in coping financially with the increased cost of regulatory environmental regulations and the high cost of raising capital through private financial institutions.

"Steel Industry Lending Guidelines" were drafted, submitted to Congress, and published in the Federal Register, 43 Fed.Reg. 16360. According to the guidelines, an applicant for financial assistance under the steel program must meet all of the requirements of PWEDA and its regulations. An applicant must not only be in a "redevelopment area" designated by EDA,2 but PWEDA also requires that the applicant be in such a financial position that the assistance it seeks from EDA is not available from other sources, whether private lenders or other federal agencies. 42 U.S.C. § 3142(b)(4).3 EDA reiterated this requirement in the Steel Industry Lending Guidelines.

To qualify for assistance under the program, a steel company must have been experiencing financial difficulties and be in need of funds to upgrade the efficiency and productive capabilities of existing steel facilities. EDA funds, however, may not be used to assist an applicant in increasing its share of the marketplace at the expense of other domestic producers. Section 702 of PWEDA prohibits assistance to a company when the result would be "to increase the production of goods . . . or the availability of services or facilities" when "demand" is insufficient to employ the existing "efficient capacity" of competitive commercial or industrial enterprises. See 42 U.S.C. § 3212; 13 C.F.R. § 309.2 (1980). Simply put, EDA assistance is not to be used to expand the production of steel products if there already is excess capacity in the domestic steel industry.

Phoenix Steel is a small Eastern Pennsylvania steel company which commands only one-half of one percent of the domestic steel market. It fabricates seamless steel pipe and tube at its Phoenixville, Pennsylvania plant and steel plate products at Claymont, Delaware. Phoenix's severe financial distress, caused by unfavorable economic and market conditions in the 1970's as well as other factors, made it evident that without extensive modernization of existing steel facilities and equipment, it could not survive. In April 1978, Phoenix applied for EDA assistance. After an exhaustive review of Phoenix's financial condition and needs, including written submissions in opposition by Lukens and its attorneys, EDA made extensive findings and agreed to guarantee 90% or $29,039,400 of rental obligations to be incurred by Phoenix under a proposed leverage equipment lease.4 The mechanics of the loan guarantee require a private lease lender to purchase interest-bearing notes from the equipment lessor. The lessor will use the proceeds of the purchase to buy modernization equipment which will then be leased to Phoenix at a rate sufficient to service the debt evidenced by the notes. The notes will be secured by a security interest in the equipment and the leases and by the EDA guarantee. EDA will be protected because if it should be required to meet the guarantee, it would be entitled to an assignment of the notes and the collateral.

Based on the loan guarantee, Phoenix planned to upgrade its facilities primarily in three areas: (1) Vacuum Degassing: Phoenix needed equipment which would enable it to produce more cheaply low hydrogen and oxygen content steel. The desired equipment is essentially a vacuum process by which the gaseous content in a ladle of steel is reduced. (2) Heavy Plate Equipment: Phoenix had the capability of producing large plates of steel only by welding two sheets together at a higher cost. Modern equipment would allow larger plates to be produced in single form and thus more economically. (3) Special Plate Finishing Equipment: Phoenix can produce heat treated steel but only by "thermal cycling," i. e., varying the temperature in the furnace. Proper finishing equipment would eliminate this inefficient process.

EDA, in approving Phoenix's application for assistance, concluded that (1) modernization was reasonably calculated to provide more than a temporary alleviation of under-employment or unemployment; (2) there was a reasonable assurance of repayment of Phoenix's lease obligation if it were to modernize its facilities; (3) Phoenix was unable to obtain adequate financing from alternative sources without EDA assistance; and (4) modernization would not violate the Act by causing a significant increase in Phoenix's production or enable it to manufacture any "new" products.

Lukens Steel is a substantially larger and more profitable competitor of Phoenix. It is concerned over the proposed EDA assistance to Phoenix because it fears Phoenix would gain a competitive advantage, particularly in specialty steel plate product lines which are the major source of competition between the two firms. Lukens commenced suit in the United States District Court to block the EDA guarantee, alleging that it would (1) create unfair competition in violation of section 702 of the Act; (2) violate the 15% equity contribution requirement of the Act; (3) violate section 202(b)(4) of the Act because Phoenix had the ability to modernize without EDA assistance; and (4) violate section 202(b)(4) because Phoenix could not assure repayment.

The district court sustained EDA and Phoenix concluding that EDA acted within the scope of its authority and that its findings concerning the need for the vacuum degasser and heavy plate equipment were not arbitrary and capricious. See Lukens Steel Company v. Kreps, 477 F. Supp. 444, 447 (E.D.Pa.1979). The court, however, found that EDA in making its section 702 findings, had failed to include projections of stainless steel conversion levels pertaining to Phoenix's proposed specialty plate finishing project. Also, the district court needed additional clarification of Phoenix's 15% capital contribution to the projects. The court, therefore, remanded the case to EDA to develop a more extensive record on this phase of the application. On all other issues, including Phoenix's ability to obtain funds from other sources, the district court found in favor of EDA and Phoenix.

Phoenix submitted additional data to EDA on remand to show that small projected production increases in the specialty plate finishing line would not be attributable to EDA funds for modernization. EDA thereafter resolved the section 702 question in Phoenix's favor along with the 15% contribution issue. The district court affirmed the findings of EDA. Lukens appeals.

Although Lukens challenged EDA's findings and conclusions in the district court on a number of grounds, its argument on appeal is that the district court erred in upholding EDA's conclusions that (1) assistance to Phoenix would not contravene the unfair competition provisions of section 702 and (2) that Phoenix did not have available other independent sources for long-term financing.

Before we may assess the merits of Lukens' argument, it is critical to note that the scope of the district court's and our review over EDA's action is exceedingly narrow, under which the agency's decision is entitled to a presumption of regularity. American Iron and Steel Institute v. EPA, 568 F.2d 284, 296 (3d Cir. 1977). Action taken by a federal agency under the authority of PWEDA is "informal" and the review of such decision making must be contrasted with de novo review and the substantial evidence test of 5 U.S.C. § 706(2)(E) for more formal agency proceedings. Thus, in reviewing EDA's findings, the court is limited by the Administrative Procedure Act, 5 U.S.C. §§ 551 et seq. (1976), to a determination whether the agency action is "arbitrary, capricious, an abuse of discretion, or otherwise, not in accordance with law." 5 U.S.C. § 706(2)(A). The district court viewed its limitations of review as follows:

Whether the EDA's decision was a wise one, or even a desirable one, is not for the Court to decide. The district court does not sit as a super-agency empowered to substitute its judgment for that of the agency. Evidence weighing must be left to the agency making the policy decision.

477 F.Supp. at 449.

The Supreme Court in Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416-17, 91 S. Ct. 814, 823-24, 28 L. Ed. 2d 136 (1979), assessed the task a court faces in determining whether agency findings are "arbitrary and capricious":

To make this finding the court must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment. . . . Although this inquiry into the facts is to be searching and careful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency.

Under the "arbitrary and capricious" standard, the district court sits neither as a fact finder nor as a judge of the weight of the evidence. The task of the court, however, is not to routinely endorse agency action but rather to review thoroughly the administrative record to insure that a rational basis exists for the agency's action. If such a rational basis exists the inquiry must end there and the wisdom of the agency's action is of no further judicial moment. See 477 F. Supp. at 451. We must also keep in mind the well-settled principle that an agency is entitled to deference in interpreting the statutory and regulatory provisions which it is charged to administer. See Udall v. Tallman, 380 U.S. 1, 85 S. Ct. 792, 13 L. Ed. 2d 616 (1965). Unless the interpretation is "plainly erroneous or inconsistent with the (statute or) regulations," Lucas Coal Co. v. Interior Board of Mine Operations Appeals, 522 F.2d 581, 584 (3d Cir. 1975), the court owes deference to an agency's administrative interpretation.

With the narrow scope of review thus delineated, we turn to Lukens' challenge to EDA's proposed aid to Phoenix. Lukens argues that the federal aid in the three challenged product lines will result in an increase of Phoenix's production of specialty steel plate in violation of section 702 of PWEDA. Section 702 provides:

No financial assistance under this chapter shall be extended to any project when the result would be to increase the production of goods, materials, or commodities, . . . when there is not sufficient demand for such goods, materials, commodities, . . . to employ the efficient capacity of existing competitive commercial or industrial enterprises.

42 U.S.C. § 3212. The parties do not dispute EDA's preliminary finding that there is insufficient demand for the kind of products made by Phoenix to utilize the efficient capacity of existing domestic steel enterprises. See 477 F. Supp. at 453. EDA, however, made no formal finding on that issue because it found that federal aid would not significantly increase Phoenix's production level. Id. Thus, we are called upon to decide whether EDA had a rational basis for its determination that the aid it had agreed to provide would not significantly increase production level at Phoenix in violation of section 702.

In the context of Phoenix's request for assistance, EDA interpreted "increase in production" as meaning a " 'significant increase in production over recent historical levels.' " 477 F. Supp. at 453. The year chosen as the recent historical level was 1973, Phoenix's last peak year of production, when its position was competitively favorable. Comparing production figures with the 1981 production levels projected to follow the infusion of federal aid, EDA concluded that no significant increase in Phoenix's production level would result from the aid. Lukens challenged EDA's interpretation of section 702 as "at variance with the plain meaning of the statute, which it contends prohibits anything other than a de minimis increase over the current level of production." 477 F. Supp. at 453. The district court rejected this argument, deferring to EDA's gloss on section 702:

The agency's interpretation of § 702, which forbids assistance that would result in a significant increase in productive capacity, while permitting quality-improving and cost-reducing modernizations which insignificantly increase production when measured against profitable production levels, is a reasonable interpretation which reconciles existing tensions in the Act.

Id. at 455.

Lukens does not argue in this court that the district court erred in deferring to EDA's use of the 1973 peak year in making its section 702 analysis. Rather, Lukens contends that (1) the district court's basis for sustaining EDA's aggregate production approach was clearly erroneous, and (2) that the district court's finding that Phoenix's production would not be significantly increased by EDA financial assistance was arbitrary and capricious. EDA regulations permit financial assistance when such assistance will not result in a significant increase in production of goods and materials.5 Lukens maintains that EDA funds will indeed result in a significant increase in production levels for specialty steel plate, its main source of competition with Phoenix. In making its section 702 determination, EDA did not focus specifically on the effect EDA funds would have on specialty plate production. Instead, it used an aggregate production figure approach. This figure reflected the total amount of all steel produced at Phoenix's Claymont plant.

Lukens challenges the use of aggregate production figures, suggesting that it is inconsistent with the statute. EDA used the aggregate production approach because it concluded that the effects of the proposed modernization equipment "cut across product lines" and could not be isolated to a specific product line such as specialty plate. In addition Lukens argues that, even if aggregate production figures are used, the record shows that Phoenix's production of specialty plate will significantly exceed its 1973 specialty plate production levels by virtue of the EDA assistance. Thus, Lukens argues that EDA's decision to grant the assistance is "arbitrary and capricious" and not otherwise in accordance with law and thus that, even under the limited standard of review afforded to this court, it must be disapproved.

EDA advances two arguments in support of its decision to use aggregate production figures in making its section 702 analysis. First, EDA asserts that because an increase in specialty plate production would be caused as well by the strategy of Phoenix's management, which is to shift the production "mix" toward the production of more specialty steel, it is unable to determine whether projected production increases, if any, will be caused by its assistance. The district court characterized this as a "weak argument." Lukens Steel Co. v. Kreps, 477 F. Supp. 444, 457 n.41 (E.D.Pa. 1979). We agree that this argument is not persuasive. As the district court stated, "without the new equipment which will enable Phoenix to reduce production costs, there is no reason to expect that a new 'marketing strategy' would be more successful than the present one." Id. Indeed, Phoenix itself, in its application for EDA assistance, indicated that its goal in seeking the assistance was, inter alia, "to supplement its position as a leading specialty steel plate and plate products supplier." Thus, EDA's assertion that it is unable to determine a causal relationship between its assistance and Phoenix's competitive position is to undermine the very basis for EDA's decision to grant that assistance. Presumably EDA would not have agreed to provide Phoenix with assistance if it did not believe that assistance would play an important role in restoring Phoenix to a competitive position. Thus, if EDA's decision to use aggregate production figures is to be sustained, it must be on another basis.

The second justification offered by EDA for its use of aggregate production figures is that, because its assistance will be used in the production of all steel products produced by Phoenix, EDA is incapable of measuring the changes in production which its assistance may cause in individual product lines. It was on this basis that the district court sustained EDA's use of aggregate production figures. However, we do not believe that EDA's choice of aggregate production figures as the appropriate unit of comparison can be sustained on this basis either.

First, we have serious doubts whether, in a case such as this, a decision to use aggregate production figures is consistent with the purpose of the statute. The Act is designed to prevent EDA assistance from being used to improve the market share of any American producer at the expense of other American producers. The Act accomplishes this purpose through its prohibition of assistance that would significantly increase production when there is excess capacity in American industry.

The EDA acknowledges that when its assistance would result in a significant increase in production of an identifiable product line, it is forbidden to give assistance under the provisions of section 702. Thus, in evaluating Phoenix's application for assistance, EDA denied Phoenix's request for assistance in obtaining head spinning equipment because of its determination that the equipment would be used to produce only one particular product and, hence, when combined with the existing equipment, would result in a significant increase in production.

Lukens argues in its principal brief to this court that based on Phoenix's projections, upon which EDA relied, the modernization program also would have resulted in substantial increases in production levels of Phoenix's specialty product lines. Lukens contends that the record demonstrates the following projected percentage changes by product lines between 1973 and 1981:

Plate Product 1973 1981 % Change

Carbon 151,946 117,845 - 22% (decrease)

HSLA 33,974 66,575 k 96%

Alloy 13,133 27,230 k 107%

Heads 12,986 20,470 k 58%

Clad 1,418 7,290 k 414%

Stainless Conversion 1,362 22,000 k 1615%

EDA correctly asserts that while 1973 was Phoenix's peak aggregate output year for all types of plate, the record indicates that 1973 figures do not necessarily represent the peak output of particular kinds of plate.

Even when the aggregate total of all products produced remains the same, there can be no dispute that a shift in the mix of products produced can have significant effects in the marketplace. For example, if company A produced 100 airplanes and 1000 automobiles in one year and 1000 airplanes and 100 automobiles in the second year, such a shift in the products manufactured could be expected to have a significant impact on a competitor of that company which produced airplanes. This effect would occur even though company A's aggregate figures for the production of airplanes and automobiles remained at 1100 for the year.

In this case, Phoenix projects that it will produce more specialty steel plate as a result of EDA assistance. On the other hand, Phoenix projects that carbon steel will decline in its percentage of the total production. Thus, applying the foregoing analogue to this case, by shifting the mix of production, it would be possible for Phoenix to reduce its carbon plate production, a low profit item, from the 1973 production level of 151,000 tons to 100,000 tons and increase its stainless conversion, a high profit product, from the 1973 production level of 1362 tons by 25,000 tons and its HSLA product line from the 1973 level of 33,974 tons by 25,000 tons and in no way affect its aggregate production. Utilizing an aggregate production figure approach may effectively conceal substantial improvement in production levels of specific product lines which would be inappropriate and inconsistent with the purposes of section 702 of the Act.

EDA itself recognizes that a shift in product mix will result from its assistance. In EDA's Action Memorandum of March 16, 1979, it stated:

The product mix changes thus have a significant effect on profits. Phoenix's modernization is planned to ultimately enable a shift to higher margin products as a percentage of its total output and thus to generate profits to ensure its long term viability.

For the private lender, a shift in the product mix from low profit production lines to higher profit products will undoubtedly enhance the attractiveness of the loan. A change in production techniques which will "generate profits to ensure (the borrower's) long term viability" is a very strong endorsement on the merits of the loan and the ability of the borrower to repay. EDA is, of course, interested in repayment under the provisions of PWEDA but there is tension between the provisions for repayment and adequate security for the loan and the provisions and purposes of section 702. As we have already noted, section 702 forbids financial assistance to a project when the result would be to increase the production of goods when there is not sufficient demand to employ the efficient capacity of existing competitive enterprises.

As we have demonstrated, changes in product mix may have a significant effect upon other competitors which would be inconsistent with the purposes of section 702 of the Act. On the record before us, there is evidence that a shift to higher profit margin products such as specialty steel is inevitable. See Action Memorandum of March 16, 1979 at Appendix No. 47. When, as in this case, EDA assistance is specifically requested for the acquisition of modern equipment which is intended to facilitate the production of particular products, and thus effect a change in product mix, an aggregate approach is not consistent with the Act. We think the statute requires, absent compelling reasons to the contrary, that changes in production be measured in terms of those products for which the EDA assistance is requested.

Here, EDA's primary justification for its choice of the aggregate production approach is that it is unable to measure the increases in levels of production of specific product lines caused by its assistance. We believe, however, that this reason is without support in the record. Indeed, the administrative record warrants the opposite conclusion, suggesting that EDA is able to estimate production increases resulting from its assistance. Again, in its Action Memorandum of March 16, 1979, EDA stated with respect to Phoenix's projected production:

Plate is expected to significantly grow in market position from 1.6% to 2.8% of the market, a 102% growth, in the period 77-81. Phoenix projects this total market growing by 27% in the 77-81 period, therefore, the company is projecting a 3.8 times greater growth rate than total market growth. The justification for growth of such magnitude is contained in Phoenix's modernization program.

(Emphasis added). In addition, in justifying its assistance to Phoenix, EDA has elected to compare increased production expected to be generated by the modernization program with 1973 production levels rather than current levels. This implicitly recognizes that EDA assistance will result in increased production over current production levels. Furthermore, the financial analysis of Phoenix's projected income as contained in the Action Memorandum asserts:

The key assumption to the sales forecast is market growth overall, ability of the plate products to penetrate deeper into their market, the attributes imported to plate products by modernization and the shift to a higher value sales mix in plate products.

Phoenix itself, in its application materials, submitted production projections in which it attempted to isolate the effect of EDA assistance from all other relevant variables. Specifically, in its answers to items 15, 16, and 17 of Form ED-250, which it submitted to EDA on September 12, 1978, Phoenix stated as its "Key Assumptions":

These forecasts isolate the effects of the projects on Phoenix's financials from independent market variables. We have assumed that the impact of these independent market variables (i. e., macro market variables) on Phoenix would be the same in terms of tonnage and mix with or without the projects. If the forecasts included these independent variables, it would not be possible to determine how much of the total effect was contributed by the projects alone. Therefore, these variables have been eliminated from the forecasts, and the difference between the 1978 base (package) and the forecasts that include the projects is attributable only to the projects. Since independent variables are eliminated, the 1978 base serves as the "base" for each successive year.

These projections and statements reveal that Phoenix is counting on increased production and a shift to a higher value sales mix in plate products. This suggests increased production in specialty steel product lines where these higher profit margins prevail.

We therefore believe that in light of the foregoing representation made by Phoenix and the projections and analysis made by EDA, much of which is reflected in its Action Memorandum, a conclusory assertion that increases in production levels cannot be measured by product lines is unsupported. Such an approach appears inconsistent with the provisions of the Act and EDA's reason justifying such an approach is not persuasive. Therefore, EDA's decision to apply the concept of aggregate production in measuring changed production levels resulting form its assistance to Phoenix is, on the record before us, arbitrary and inconsistent with the purpose of the Act. We hold, therefore, that the district court erred in granting the motions for summary judgment and we reverse and remand to the district court with directions to return the matter to EDA for further consideration of this issue consistent with this opinion.6

Although our disposition of the section 702 argument requires that EDA reconsider its decision to provide assistance to Phoenix in obtaining a vacuum degasser, we also consider Lukens' separate challenge to the vacuum degasser because of its possible relevance on remand to the agency. Lukens' second major argument is that the EDA grant violates section 202(b)(4) of the Act which bars an extension of a loan or guarantee "unless the financial assistance applied for is not otherwise available from private lenders or from other federal agencies . . . ." 42 U.S.C. § 3142(b)(4). Lukens originally pled that section 202(b)(4) would be violated by the EDA grant but withdrew this issue when Phoenix submitted adequate documentation of its heavy debt burden, negative net worth and inability to obtain alternative long-term financing. The district court nonetheless resolved the issue against Lukens on the merits finding that the record rationally supported EDA's conclusion that alternative financing was unavailable to Phoenix. See 477 F. Supp. at 462.

Following the remand, however, Lukens argued, based on allegedly new information, that the vacuum degassing project was completed by Phoenix pending these proceedings without EDA assistance, and thus the proposed project would violate section 202(b)(4). Phoenix had decided, due to the extensive delay in obtaining approval of EDA assistance, that it had become necessary in the interim to invade its existing working capital to undertake the vacuum degassing project. Lukens now argues that this use of Phoenix's own funds demonstrates that alternative financing was available for the vacuum degassing project and that the EDA had been arbitrary in its determination that section 202(b)(4) did not interdict assistance.

Initially, EDA and Phoenix argue that Lukens, by withdrawing its section 202(b)(4) complaint prior to remand, should be precluded from raising the issue again. There is evidence in the record to indicate that Lukens was aware of Phoenix's use of its own funds. Thus, EDA and Phoenix argue that the issue should have been raised before the district court initially and it was inappropriate to consider the issue again on remand. See Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 553-54, 98 S. Ct. 1197, 1216-1217, 55 L. Ed. 2d 460 (1978).7

Both the district court in its initial review and following the remand considered the merits of Lukens' section 202(b)(4) argument and rejected it. We need not be concerned with whether the issue was properly raised because we agree with the district court's resolution of the merits of the problem. The district court concluded: "The fact that Phoenix used working capital as interim financing for the vacuum degassing equipment does not compel the conclusion that there is no rational basis for the EDA's determination that private financing was not available for Phoenix' modernization program." Indeed, it is apparent that Phoenix was forced to invade its badly strained working capital to finance the highly needed vacuum degassing equipment at the expense of increased secured borrowing on a short-term basis and at high interest rates because of its inability to obtain long-term permanent financing from other sources without EDA assistance. Phoenix's use of its working capital for temporary short-term financing does not diminish the critical need for EDA assistance for the vacuum degassing project. Phoenix was forced to deplete its already strained working capital resources in an effort to consummate a project which would have been completed with government assistance but for the prolonged delays in obtaining final approval of the grant. The EDA funds will serve to restore necessary working capital to Phoenix which Phoenix would not have been forced to dip into in the first instance for the vacuum degassing project but for the long delay in processing its application. Thus, we agree with the district court that the completion of the vacuum degassing project by the use of working capital does not demonstrate that it could have been funded without EDA assistance on a sixteen-year term basis as proposed by the guaranteed lease. Moreover, Congress vested EDA with broad discretion to determine "in the opinion of the Secretary," 42 U.S.C. § 3142(b)(4), "when and under what circumstances other private or non-private financing is reasonably available." S.Rep.No. 193, 89th Cong., 1st Sess. at 29, 34. We, therefore, perceive no violation of section 202(b)(4)'s prohibition against federal funding when alternative sources of finances are available.

The complexity of this case underscores the vital role administrative agencies and their expertise play in the evaluation of projects which are authorized by Congress. It is an administrative and not a judicial task to assess the merits of a particular application for federal assistance and our role is merely to assure that the agency has considered the relevant criteria and acted on a rational basis and within the powers delegated to it by Congress. Nevertheless, when the agency has acted arbitrarily and capriciously, without support in the record, a reviewing court must not abdicate its responsibility. Accordingly, because we believe the EDA has acted in such a manner, we will reverse the judgment of the district court and remand the cause to it with directions to return the matter to the Agency for further consideration.

Each side to bear its own costs.