under the deferential standard of review, the plaintiffs would need to do more for the Court to find the Department's standards technologically impossible. Based on the findings contained in the record -- of ASHRAE, the National Institute of Standards and Technology ("NIST"), and DOE -- the Department has met its burden under this first prong of the Steelworkers test by showing "that modern technology has at least conceived some industrial strategies or devices which are likely to be capable of meeting" the standards. American Iron and Steel, slip op. at 8. For example, the Department indicated its conclusion, based on studies of current technology and predictive models, that electric storage water heaters "built with 2.75 to 3 inches of foam insulation surrounding the tank and R-8 fiberglass on the controls, or 6.5 inches inches of fiberglass surrounding the tank and controls, will meet the criteria." 55 Fed. Reg. at 23867. Similarly, the Department was able to identify certain improvements in gas water heater design that would, according to the studies and extrapolations, comply. See 55 Fed. Reg. at 23857-58. Because the evidence and reasoning that the Department employed in making such findings is documented on the record, the Court finds that DOE was not arbitrary or capricious in concluding that compliance with the interim regulations is technologically feasible.
The requirements under the second prong -- the economic practicability prong -- of the Steelworkers test are more complex, but the deference to be accorded the Department is nonetheless substantial: "A standard is economically feasible if the costs it imposes do not 'threaten massive dislocation to, or imperil the existence of, the industry.'" American Iron and Steel, slip op. at 8 (quoting Steelworkers, 647 F.2d at 1265).
Under the particular statute at issue -- as applied to commercial storage water heaters -- no technologically possible standard would seem to violate this second prong. The standards that the Department has promulgated are binding only on the construction of federal buildings. Yet there has been no showing that federal construction comprises the bulk of, or even a substantial part of, total commercial storage water heater sales by the industry. In fact, plaintiffs seem more concerned about the possible adoption of the interim standards by states and localities, and about the money that will have to be spent by plaintiffs trying to persuade the states and localities not to adopt the standards. See Pl. Memo. at 13-14. Possible adoption by nonfederal entities, however, is not currently within the jurisdiction of this Court, even if it were ripe for judicial review. As a result, plaintiffs have not established that even the loss of the entire federal government as a customer would cause a meaningful enough dislocation of the industry to require the Court to remand or vacate the revised interim standards.
After all, if the standards truly are impracticable, as the plaintiffs contend, then the federal government will not be able to find any water heater suppliers for its new buildings, and DOE will be besieged by federal officials in charge of federal construction projects demanding that the standards be changed. In the mean time, plaintiffs will be able to continue selling to nonfederal purchasers.
Based on the above analysis, the Court cannot find that the revised interim standards are substantively unreasonable. Plaintiffs, however, also claim that the promulgation of the standards was procedurally defective.
In particular, they argue that DOE has failed to analyze the standards adequately "in terms of energy efficiency, . . . economic cost and benefit, and impact upon affected groups." 42 U.S.C. § 6839. Some of these arguments are similar to those discussed above. For example, plaintiffs claim that DOE cannot have performed a reliable cost-benefit analysis, because DOE never obtained or produced a complying test model on which to base its evaluations. Furthermore, plaintiffs argue, the theorizing DOE did on hypothetical models was riddled with errors and false assumptions. Plaintiffs especially take issue with the computer model DOE employed to make many of its predictions. According to the plaintiffs, the computer results were so inaccurate that DOE had to add in a "fudge factor" to get everything to come out right. See Pl. Memo. at 38-40, 56-58.
As plaintiffs correctly point out, the Court needs only ascertain whether or not the Department's cost-benefit analysis was irrational when considered against the backdrop of the evidence that was or reasonably should have been before it, see Steelworkers, 647 F.2d at 1266; in other words, whether "the agency has developed substantial evidence" on which to base its conclusions. Id. Based on the record that DOE has compiled since the remand in GAMA I, the Court now finds that the Department has sufficiently complied with its procedural obligations.
It is important to remember, in reviewing DOE's actions, that although the standards are final for purposes of judicial review, they are nonetheless only interim standards. Congress seems specifically to have contemplated that this initial set of standards would not be perfect. Thus, in drafting the statute, Congress required that the interim standards be adopted only in federal buildings. See 42 U.S.C. § 6835. Moreover, the legislators mandated that, after adoption of the interim standards, DOE perform demonstration projects to ascertain more exactly the costs and benefits of the standards. See 42 U.S.C. § 6833(1), (2). Following those projects, DOE is to report back to Congress, and only then are final standards to be promulgated. If Congress anticipated that DOE would be able to perform a complete cost-benefit analysis right away -- which it presumably would have been able to do if complying models were already in significant use or readily available -- then "interim" standards would not have been necessary. Although this statutory scheme does not give the Department license to act arbitrarily and capriciously, it does seem to envision special deference to DOE's interim set of standards. Congress expected that they would require further refinement.
When viewed in this light, DOE's actions (subsequent to the remand) were within the realm of reason. Based on tests performed by itself and others, the Department developed a computer program to model the effects of increased insulation on standby loss. Because, as plaintiffs point out, the standards proposed by DOE will exceed the performance capabilities of existing units, the method employed by DOE of extrapolating from existing data by means of a computer simulation was a reasonable way to attempt to predict costs and benefits. Based on its modeling, DOE was able to estimate the combination of tank and control insulation and other conservation devices such as intermittent ignition devices ("IIDs") and flue dampers that would achieve compliance. DOE was not itself required to produce a complying model, but was directed only to provide a reasonably derived set of requirements. See 42 U.S.C. § 6832(9).
Plaintiffs properly raised objections during the rulemaking process to the Department's methodology. In the Court's view, DOE adequately responded to those objections. The Department was not obligated to convince plaintiffs that one testing method was superior to another; it was obligated only to consider and reasonably respond to plaintiff's comments. In GAMA I, the Court found that DOE's responses were "conclusory and vague and therefore insufficient to justify the rule." 722 F. Supp. at 796. Since the remand, however, DOE has shown on the record that it has given adequate consideration to the plaintiffs' comments. The Department does not have to be right; it must simply not be arbitrary, capricious, or irrational.
For example, one of the primary critiques plaintiffs made of the Department's methodology related to the what the plaintiffs described as a "fudge factor," the amount of standby loss that DOE added in to its calculations and attributed to nonjacket heat losses. Plaintiffs claim that the Department never identified any way of insulating against such losses in electric water heaters, and miscalculated the amount of nonjacket losses in gas and oil water heaters. DOE asserts in response that it performed loss studies solely on the jacket first, in order to get an undistorted idea of the amount of loss attributable to the jacket alone, see 55 Fed. Reg. at 23873, and then, in calculating what modifications would lead to compliance, added in the fittings and controls losses as derived from certain named studies as well as NIST data. See 55 Fed. Reg. at 23853-54, 23857-58. The Department's reasoning process is on the record, as are its extensive responses to plaintiffs' comments. Regardless of whether that reasoning later proves to have been faulty in one way or another, it seems to the Court not to be arbitrary, capricious, or irrational.
Plaintiffs argue in a similar vein that a number of the aspects of the Department's evaluation of gas water heaters are fatally flawed. plaintiffs assert, for example, that DOE's expectation that flue dampers and IIDs would be installed on low-input units is unreasonable, because the installation of those devices -- which would be necessary in many cases to bring the low-input units into compliance -- would be too expensive relative to the price of the units. In response, DOE points out that most of the higher output units are already equipped with IIDs and flue dampers, so the added cost for those units would be nonexistent. In regard to the lower input units, DOE concedes that some may simply not be able to be modified cost-effectively enough to be brought into compliance with the new standards. But as DOE correctly argues, a new standard is not unreasonable simply because certain product types will have to be discontinued. Under the Steelworkers test, the Department must demonstrate only that the costs of the new standards will "not threaten the existence or competitive structure of an entire industry, even if it does portend disaster for some marginal firms." 647 F.2d at 1272.
Throughout their briefs, plaintiffs assert that the Department has not quantified in enough detail the costs and benefits of the new standards. To a certain extent, DOE's estimates are often conjectural. But some speculation and prediction is inevitable when an agency is prescribing technology-forcing standards, particularly when, as here, the costs of compliance will likely be different for each individual manufacturer; indeed, it would be unreasonable to require the Department to come up with the exact retooling and modification expenses that each industry member can expect to bear in bringing its products into compliance. As the Steelworkers court recognized, "when the agency has proved technological feasibility by making reasonable predictions about experimental means of compliance, the court probably cannot expect hard and precise estimates of costs." 647 F.2d at 1266.
Under that standard, DOE has met its burden of making reasonable cost-benefit estimates. For example, in determining the cost of complying electric and gas commercial water heaters, DOE began with the amount of increase that had been determined in the residential market, and then modified that number according to predicted differences for the commercial market. 55 Fed. Reg. at 23853, 23861. Despite plaintiffs' claim that the Department's method neglected certain factors, such as the difference in size between the commercial and domestic markets (which makes expenditure spreading more difficult in the former), the Department's prediction is a reasonable way of estimating the increased costs. DOE does not need to produce exact figures; it simply needs to be able to say, based on substantial evidence, that the costs will not dislocate the industry.
Plaintiffs have taken issue with a number of DOE's other assumptions and predictions as well,
apparently in a effort to demonstrate that the cumulative effect of the Department's errors is so significant that the cost-benefit analysis must be considered fundamentally unsound. But after a careful review of the record, the Court concludes that, as in the case of the factors discussed above, the Department has provided adequate on-the-record justification for its actions and sufficiently explained responses to plaintiffs' comments. As a result, the Court finds that its remand in GAMA I has been complied with in all material respects, and thus declines either to remand again or to vacate the standby loss requirements at issue.
An appropriate Order accompanies this Memorandum.
ORDER - September 11, 1991, Filed
Upon consideration of the cross-motions and the entire record herein and for the reasons stated in the accompanying Memorandum, it is hereby
ORDERED that defendant's motion for summary judgment is granted; and it is further
ORDERED that plaintiff's motion for summary judgment is denied.