Lead Opinion

*1377BALDWIN, Circuit Judge.

This appeal is from a judgment of the United States Claims Court1 granting the request of Kay Manufacturing Company (Kay) for fees and expenses pursuant to the Equal Access to Justice Act (EAJA). 28 U.S.C. § 2412 (Supp. V 1981). We reverse.

Background

The action that generated fees and expenses was brought in accordance with the Renegotiation Act of 1951, 50 U.S.C.App. § 1211 et seq. (1976). In September, 1973, the Renegotiation Board ruled that Kay realized excess profits of $150,000 in 1969 from subcontracts for the machining of 105-mm cartridge case bases. Kay brought an action in the Court of Claims in December, 1973, seeking a determination that it had not realized any excess profits in 1969. Trial was held in 1978 after extensive pretrial discovery. The trial judge’s opinion on the merits, issued September 29, 1980, found that Kay had realized excess profits of $111,332. On review, the Court of Claims rejected the trial judge’s decision and held that the government failed to meet its burden of proving Kay’s profits were excessive. Kay Mfg. Co. v. United States, 676 F.2d 555 (Ct.Cl.1982). Kay’s subsequent request for fees and expenses pursuant to the EAJA was granted in part by the trial court. Both parties appealed.

At the trial on the merits, Kay made a prima facie showing of reasonable profits by comparing its 1969 profits to the profits of competitors. The government contended differences between Kay and its competitors made direct comparisons of profits between Kay and its competitors inappropriate. For instance, Kay did finishing work on steel material furnished by its customers while Kay’s competitors purchased the materials on which they worked. The government then attempted to prove Kay’s profits were excessive by a comparison of Kay’s profits in two historical base periods with its profits in 1969.

The trial judge found that differences between Kay’s operations during the base periods and the review year rendered inadequate the government’s base year comparison. Despite finding the government’s proof to be unpersuasive, the trial judge determined that Kay realized excessive profits by using hypothetical discussed at trial and assumptions made by the judge to alter the comparisons provided by Kay. The trial judge’s resort to hypothetical and assumptions to compensate for the differences between Kay and its competitors was an approach which may have been permitted by the Court of Claims decision in Major Coat Co. v. United States, 543 F.2d 97 (Ct.Cl.1976). Kay Mfg. Co., 676 F.2d at 566. In Major Coat, the Court of Claims stressed the need for the government to provide comparative evidence of the profits of companies similar to the reviewed company to prove excessive profits. The court explained Major Coat in its review of the Kay Mfg. Co. case as follows:

Because Major Coat imposed a stricter standard of proof than had been required previously, we [the Court of Claims] provided that “the court will undertake, however reluctantly, and only for a limited span of time, to engage in judgmental adjustments and reworkings of the data placed in the record in order to arrive at a finding of a reasonable profit of the renegotiated contractor — where the state of the evidence permits this to be done in a responsible manner, ...”

Kay Mfg. Co., 676 F.2d at 559 (emphasis in original). After reviewing the decisions which followed Major Coat, the Court of Claims announced that the limited span of time granted by Major Coat was at an end, making the trial judge’s adjustments and decision based on those adjustments, improper. Since the government failed to prove Kay’s profits for 1969 were excessive, the court entered judgment for Kay.

In this appeal, the government argues that Kay should recover nothing under the EAJA because the government’s position in *1378the litigation was substantially justified. It also contends that no fees or expenses incurred before the effective date of the EAJA, October 1, 1981, can be recovered.

DISCUSSION

We first consider whether the EAJA applies to fees and expenses incurred before October 1, 1981, the effective date of that act. The government argues that sovereign immunity was not waived for fees and expenses incurred before October 1, 1981, because no specific statutory language or legislative history indicates Congress intended fees or expenses generated before October 1, 1981, to be recoverable.

The EAJA authorizes a court to “award reasonable fees and expenses of attorneys * * * to the prevailing party in any civil action brought by or against the United States * * * in any court having jurisdiction of such action.” 28 U.S.C. § 2412(b) (Supp. IV 1980). The act applies to “any adversary adjudication * * * and any civil action * * * which is pending on, or commenced on or after” October 1, 1981. 5 U.S.C. § 504 note (Supp. IV 1980). We read the latter provision of the EAJA as a clear directive from Congress making the EAJA applicable to fees generated before the effective date of the act as well as after that date. The touchstone for determining whether the act applies is whether the action was pending on October 1, 1981, not whether the services were rendered before or after that date. The statement that a statute applies to cases “pending” on its effective date means precisely that, and unless Congress indicates otherwise, the statute applies in full to all such cases.

Nothing in the Act or its legislative history indicates that Congress intended to distinguish between fees and expenses generated before or after October 1, 1981, in cases that were pending on that date. Congress made the act fully applicable to all pending cases. Although waivers of sovereign immunity must be strictly construed, the grant of authority to award attorney’s fees in actions pending on October 1, 1981, waives sovereign immunity to permit the award of fees and expenses incurred before the EAJA became effective.

We do not agree with the government’s contention that our analysis in Brookfield Construction Company v. United States, 661 F.2d 159 (Ct.Cl.1981), requires a different conclusion. In Brookfield, the issue was whether a contractor who elected to proceed under the Contract Disputes Act (CDA) could recover interest for periods of time the case was pending prior to the effective date of the CDA. The statute interpreted in Brookfield provides that interest shall run “from the date the contracting officer receives the claim pursuant to section 6(a).” 41 U.S.C. § 611. The statutory provision at issue in Brookfield did not clearly permit the recovery of interest for periods before the effective date of the CDA. It was contended, for instance, that the contractor’s claim could not be received “pursuant to section 6(a)” until the CDA was effective. The court in Brook-field therefore denied recovery of pre-CDA interest because the language of the statute was ambiguous and the legislative history did not indicate Congress anticipated the liability that would attend the award of pre-Act interest. No similar ambiguity is' present in the provision of the EAJA at issue here.2

Although the statute permits recovery of fees and expenses incurred before as well as after October 1, 1981, the EAJA permits recovery only when the government’s position in the litigation is not substantially justified. 28 U.S.C. § 2412(d)(1)(A) (Supp. V 1981). The trial court concluded that the position of the government was not substantially justified in light of the Court of Claims’ decision on review of the merits. The trial judge emphasized that the government’s comparison of Kay’s 1969 profits to *1379base year profits was inadequate and that the government’s attack on Kay’s comparative data was an argument that the data was not meaningful rather than a refutation with contradictory data. The analysis of the trial judge was in error.

Justification for the government’s position at trial and on appeal must be measured against the law as it existed when the government was litigating the case, not against new law enunciated by the Court of Claims as a result of the appeal. Broad Avenue Laundry and Tailoring v. United States, 693 F.2d 1387, (Fed.Cir.1982). The government’s behavior, however, need not be reprehensible for fees and expenses to be awarded as suggested by the government. Under these standards the record shows that the government’s positions at trial and on appeal had a reasonable basis.

In the decision on Kay’s request for fees and expenses, the trial judge noted that the government may well and reasonably have believed Kay’s level of profits to be excessive. This statement alone does not establish the reasonableness of the government’s position, but it is one indication that there was a factual basis for the government’s position in the litigation. In support of the trial judge’s award of fees and expenses, Kay argues that the fundamental economic error made by the government was in viewing Kay as a. typical manufacturing company when in fact, it was a service rendering company only. Kay’s sales represented the value its services added to the materials. The typical manufacturing company competing with Kay, purchased the materials upon which it worked. Since Kay’s competitors purchased their materials, their profits included profits or markups on such purchases. Competitor’s profits were therefore higher than they would be if they reflected only the value of services performed in finishing the material.

The Court of Claims opinion on the merits of this case indicates that the government did not ignore this economic reality, rather it presented the best evidence it could in view of this circumstance. The Court of Claims said:

Defendant [the government] chose to use the “base years” comparison because it recognized plaintiff was unique in receiving customer furnished materials from Jarnberg instead of having to obtain the raw materials on the marketplace as did its competitors.

Kay Mfg. Co., 676 F.2d at 558, note 7. As this quotation demonstrates, the government did not ignore economic reality but instead submitted evidence it judged to be the most probative in view of the economic conditions. The government’s position at trial was further justified by case law allowing the use of comparisons similar to that proffered in this case to prove excessive profits. Gibralter Mfg. Co. v. United States, 546 F.2d 386 (Ct.Cl.1976) (comparison of contractor’s performance with his own performance in prior years); Camel Mfg. Co. v. United States, 572 F.2d 280 (Ct.Cl.1978) (comparison of contractor’s renegotiable business with his commercial business during the review period). The government therefore had a reasonable basis to pursue the case at trial.

On appeal, the government defended the trial court’s opinion even though the trial court relied on adjustments to Kay’s evidence. As the Court of Claims said when it decided the merits of this case, the adjustments made by the trial judge were apparently permissible under Major Coat. Kay Mfg. Co., 676 F.2d at 566. The government’s position in the appeal was also reasonably grounded in Court of Claims precedent.

Our conclusion that the government’s position in this case was substantially justified makes it unnecessary for us to consider other issues raised by the parties concerning the propriety of particular fees awarded or denied by the trial court.

For the foregoing reasons, the judgment of the trial court granting Kay’s request for fees and expenses is reversed.

REVERSED.

Notes

. Pursuant to order of this court dated October 4, 1982, Judge Yannello entered a judgment on October 8, 1982, corresponding to the decision recommended in this case.

. Our opinion in this regard is consistent with interpretations of several other courts: United States for Heydt v. Citizens State Bank, 668 F.2d 444 (8th Cir.1982); Photo Data, Inc. v. Sawyer, 533 F.Supp. 348 (D.D.C.1982); Wolverton v. Schweiker, 533 F.Supp. 420 (D.Idaho 1982); contra Allen v. United States, 547 F.Supp. 357 (N.D.Ill.1982).

Dissenting Opinion

NICHOLS, Circuit Judge,

dissenting.

I agree with the conclusion of the court that Kay could recover an award for legal *1380services rendered before October 1, 1981, if the case was then not finally decided, and the claim for counsel fees qualified otherwise. I respectfully differ with the conclusion of the court that the government position was “substantially justified.”

The test is, as we have held in Broad Avenue Laundry and Tailoring v. United States, 693 F.2d 1387 (Fed.Cir.1982), and Gava v. United States, 699 F.2d 1367 (Fed.Cir.1983), decided this same day, whether conduct by government counsel was “reasonable.” That their conduct may have been “reprehensible,” whatever that meant, is abandoned as the test. But it still remains up to this court to determine and flesh out in concrete cases what it conceives to be “reasonable.” In this regard, it must deem “unreasonable” conduct by government counsel that is the evil the statute, 28 U.S.C. § 2412, is aimed at. This we know is not simply that they took a stance which did not prevail. It is not even that they took a stance which could not prevail unless precedents were to be overruled. The most readily identifiable case is, I think, one when the chances of success depended chiefly on the adverse party running out of litigation resources, on the ignorance or inattention of the adjudicating tribunal, or on sheer bias and prejudice. Congress, I think, thought there were such cases, and so do I, though I also think they are far more atypical than some persons imagine. The usual and typical behavior of government counsel in the former Court of Claims, with which I was familiar, was responsible and reasonable. Counsel were in general well aware of their duty, as officers of the court, to assist in the “just, speedy, and inexpensive determination of every action,” as called for by Rule 1 of that court. When there was a vast difference between the resources at their command, and those of their adversary, they tried not to exploit the fact unfairly. When they knew of legal precedents favorable to their adversary’s position, but unknown to him, they disclosed them. They cooperated in necessary discovery, only demanding cooperation in return. But there were occasional unfortunate exceptions, often due to inexperience and excessive zeal.

It is not, as the panel stated in Broad Avenue, per se necessarily reasonable for government counsel to support and rely on a decision below in his favor, by a lower court or administrative tribunal. Such a decision must sometimes, though I should think but rarely, be abandoned and error confessed, to avoid conduct that is not substantially justified, and therefore unreasonable. It would be most presumptuous for a government attorney to refuse to support a decision in his favor, because he supposed himself smarter and better qualified to decide the matter, than the clients. Yet some government agencies exist to be exponents of particular interest groups in the community. Courts exist to hold on to their coattails and impose standards of fairness and legality. The government counsel is in the middle and must not forget he is an officer of the court. That the decision below is palpably and demonstrably unreasonable is not per se sufficient to make government counsel’s defense of it unreasonable too, but that is a factor which with other factors may lead to that conclusion.

Several cases in the Court of Claims were brought to obtain redeterminations of excessive profits determinations under 50 U.S.C. app. § 1211 and ff, made originally by the Renegotiation Board that existed under that legislation. There were several factors that should, I think, have led a reasonable government attorney to scrutinize the decisions of that agency with a critical eye, and give serious consideration to whether he should support them in the Court of Claims, or abandon them. These factors were, not necessarily in the order of importance, the following: (1) By express provisions of the statute, the board holdings enjoyed in court no presumption of correctness, and court redeterminations were de novo. (2) A perfectly reasonable board determination might be indefensible in court by reason of the differences in the applicable rules of evidence and burden of proof. (3) For reasons unknown to me, the Justice Department was unable or unwilling to call on the board to assist in defending its deci*1381sions by furnishing witnesses or otherwise, and as a result thereof. (4) The starting point of every renegotiation redetermination in the Court of Claims was a dollar figure of the amount of excessive profits, arrived at by the board for unknown and unascertainable reasons. To be sure, in a few cases, there were lurking in the records perfunctory board statements of reasons. I do not recall a single case where counsel based any argument upon them and they could not be given weight in view of the absence of anyone who could be questioned about them. Their sole utility was as a benchmark to which the court could compare its determination, after having made it on other grounds. (5) As was Kay Manufacturing Company, so most of the other Court of Claims petitioners under the Act were small business concerns, the very type the Equal Access to Justice Act (EAJA) was meant to protect. Since the judicial determination of excessive profits was inordinately prolonged and costly, even as compared to other contemporaneous civil litigation, there was always an excellent chance that any typical petitioner would run out of resources to litigate before the litigation was concluded. While the board had tarred these parties with being profiteers, few of them had the alleged excessive profits when the time came to refund them and some never had possessed them, save on paper. (6) While Kay Manufacturing and many other redetermination petitions were pending, Congress allowed the legal life of the board and its powers to lapse out of evident dissatisfaction with the work it had been doing. Thereby the cases pending here became dead-end cases, in which no significant principle or precedent could be established.

I think these reasons were more than sufficient to require a reasonable government attorney to scrutinize an assignment to try one of these cases with a more hesitant and skeptical eye than he would use on a typical administrative determination, subject to substantial evidence review. But there is more.

I take it I need not argue that no one has a right to tell a government contractor to refund part of his gains on the basis of a mere gut feeling that the contractor made too much money. Suppose Congress had constitutionally any power to endow anyone with such a prerogative, it never intended to exercise it. In the 1951 Act as in its predecessors, statutory factors were clearly set forth, to be considered and applied when the reasonableness or excessiveness of a contractor’s profits came to be determined, in the light of careful and detailed financial and performance analysis. These requirements imposed a heavy burden on those who were to make the determinations, whether in the executive or judicial branches, and, of course, an appreciation of this burden was the reason why trials were so costly and prolonged. The board was ostensibly composed of experts, and whether it was or not, it had the advice of expert staff. It therefore did not require the testimony of expert witnesses. When a court, however, determines a reasonable level of profit, it makes a fact finding on a matter of opinion, as to which it has no expertise. It can do this on the basis of proof of the underlying facts, plus the help of expert testimony to show it how to use these facts to construct an excessive profits determination within the statutory guidelines.

The Court of Claims, in the exercise of its newly imposed duty of making excessive profits redeterminations, was also guided by the fact that Congress was dissatisfied with the allocation of burden of proof by the Tax Court previously. We held that once the contractor had established the relevant financial data it relied on, and made a prima facie case, the burden shifted to the government to show how to construct an excessive profits determination from these figures, from comparable figures for other contractors or for the contractor itself at other times, and from performance information. Lykes Bros. Steamship Co. v. United States, 459 F.2d 1393 (Ct.Cl.1972). In Major Coat Co. v. United States, 543 F.2d 97 (Ct.Cl.1976), the court confronted a situation where, in the case before it, and in many others, the parties, in ignorance of the standards the court would apply, had *1382conducted trials in a manner that failed to satisfy the proper burdens of proof. To avoid failure of justice in those cases, or demanding repeats of the prolonged and costly trials, the court said it would itself rework the raw data “for a limited period.” I was and remain dubious about this because I doubted if it was proper for the court to constitute itself its own expert witness and build its structure of reasoning as a do-it-yourself operation. Only government sought a rehearing in Major Coat, as the outcome of the court’s deliberation was a clear government defeat. Had the contractor sought rehearing, I might have supported it. See also my dissent in Mason & Hanger-Silas Mason Co. v. United States, 518 F.2d 1341, 1365 (Ct.Cl.1975). Having had more personal experience with statutory renegotiation than our other judges, perhaps I was more conscious than they of my personal inability, trained only as a lawyer, to make an excessive profits determination. However, by Major Coat the right and proper rule was to be established respecting trials conducted post-Major Coat, and on this we all agreed. Meanwhile, we all enjoyed the satisfaction of playing God in cases tried pre-Major Coat, and I must admit I did it too.

Now, Kay Manufacturing was tried long after Major Coat. A reasonable government counsel, I think, after Congress had let renegotiation lapse, should have seriously considered making the best settlement he could without trial, for all the various reasons set forth above. He should have realized, if he went to trial, it would be futile unless he could produce expert testimony showing the court how to apply all the statutory factors. He was not entitled to ignore any of them unless he could show it was irrelevant, as for example, the capital and net worth factor might be in renegotiating a contractor whose line of business was not capital — intensive. If he hoped to succeed with anything short of that, he was hoping to succeed by the ignorance or inattention of the tribunal.

The majority opinion here is deficient in that it ignores the very just view taken by the merits panel, which was that the technique followed in Major Coat to prove excessive profits had been acceptable only temporarily, and the time for it had long since passed. This panel now assumes a responsible government counsel could follow some parts of Major Coat and ignore other parts, but it seems to me if he used Major Coat at all, he had to respect all of it, including the warning that government presentations that forced upon the court a do-it-yourself role must end.

The majority also ignores that the government position on the merits actually was, and had to be, dependent on the sufficiency of the reasoning of the government expert, Dr. Mock. If his testimony did not support an excessive profits determination in some amount, no one’s did, for he was the government’s only expert. Yet Dr. Mock was, as it appeared to the panel and as it appears to me, utterly unreasonable in insisting that comparisons be confined to contractor’s own profit experience in what he was pleased to call the “base years,” and rejecting comparisons with other subcontractors who supplied the same item, the cartridge base that Kay supplied, at the same time. The conditions during the so-called “base years” were so different as to be useless for comparison, as the merits panel well and clearly spells out. The government, as the court pointed out in Major Coat and elsewhere, is the only one to be able to know and furnish data as to the prices and performance of other suppliers of the same item. The refusal to do so here was on the alleged ground that Kay Manufacturing used customer-furnished material and the others purchased their own. This difficulty is often encountered in renegotiation and is easily overcome with proper adjustments, as is shown in Major Coat and elsewhere.

My conclusion is that the government did not make any serious effort to prove its case according to the standards of proof laid down by the court. Rather, it relied for success on the ignorance or inattention of the trial tribunal.

*1383The simplicity of the situation, as it would be, is complicated by the trial judge’s effort at a do-it-yourself construction of an excessive profits determination. It is probably no worse than similar efforts by the court in earlier cases, but it is unsound for reasons clearly stated by the merits panel, and was untimely besides. From the point of view of evaluating the reasonableness of counsel’s conduct, the point about it is that it was a gratuitous sua sponte effort counsel had not asked for and had no right to expect. Whether his handling of the case was reasonable or unreasonable must be assessed without regard to this unforeseen and unforeseeable assistance. One who has thrown himself into the river cannot urge his conduct was reasonable because the Coast Guard happened along at the right moment and pulled him out.

It is suggested the reasonableness of counsel’s position is supported by the board determination. I have already dealt with that. The point would have more dignity if counsel had not asked for an increase in the board’s determination, an act reflecting scant confidence in the board’s wisdom or perspicuity. Even if counsel asked for the same figure the board did, this proves nothing unless we know that the Mock testimony reflected an approach to the case the same as or similar to that which the board had previously used. I would be more inclined to find counsel reasonable if counsel knew, understood, accepted, and tried to sell to the court, the reasoning the board had previously used. These are assumptions we have no basis for in the record.

I would further remind the court that we would not assess personally against government counsel the fees I would favor assessing in this case. I do not consider his conduct to have been in bad faith, or “reprehensible,” the term we have agreed to deep-six. I think the unreasonable prolongation of this meritless government claim is properly assessable, as the EAJA assesses it, to the client and not to the lawyer. The government lawyer can properly and justly say, “don’t blame me, I only work here.” The client created the conditions under which such cases as this are possible and sometimes occur. The journeyman level attomey does with the case what he is told to do. I assume we are not holding that “superior orders” to him would exonerate the government. I.e., when we talk about unreasonable conduct, we do not necessarily limit it to unreasonableness on the part of the attorney of record, the one who the Court of Claims insisted should actually appear and try the case, and exclude unreasonable orders by superiors.

The late Judge Kunzig of the Court of Claims usually knew how to say what he meant. His opinion for the unanimous merits panel here indicates that, in his eyes, the claim was a dog. To vary the simile, he pulled no punches. Government counsel had persisted in trying the case (again I say government counsel as an institution, not a person), in face of the plainest indications it could not be won, unless all the judges of the Court of Claims passed away and were replaced by others with an itch to overrule their predecessors. Or unless someone’s attention lapsed. Government counsel at the same time made no effort — and perhaps knew of no effort they could have made — to present the case in a manner sustaining their burden of proof as prescribed. If counsel fees are not assessed in this case, we might as well declare we have decided to nullify the EAJA, and counsel fees will never be assessed. Cases that warrant them more than this one will be rare indeed.