This action challenges a regulation promulgated by defendant-appellant Department of Labor (“DOL”) establishing a methodology for determination of annual percentage changes in a wage rate required to be paid by agricultural employers who import temporary foreign workers. We granted DOL's application under 28 U.S.C. § 1292(b) for review of an order of the District Court for the District of Vermont (Albert W. Coffrin, Chief Judge) denying DOL’s motion for summary judgment.1 We now reverse.
Background
The challenged regulation, 20 C.F.R. § 655.207(b) (1984), is part of a framework of statutory provisions and regulations known as the “H-2 Program,” under which aliens are granted temporary visas to work in the United States. The plaintiffs-appellees represent New England apple growers who employ foreign workers to pick apples.2 These workers are admitted to the United States under provisions of the Immigration and Nationality Act, 8 U.S.C. §§ 1101-1524 (1982), that establish categories of “nonimmigrant aliens,” id. §§ 1101(a)(15), 1184, including aliens who come “temporarily to the United States to *137perform temporary services or labor, if unemployed persons capable of performing such service or labor cannot be found in this country,” id. § 1101(a)(15)(H)(ii). Under the Act, employers must petition the Attorney General for permission to hire such aliens. Id. § 1184(c). The Act authorizes the Attorney General to adopt regulations fixing the conditions under which such petitions may be granted and directs him to take this action “after consultation with appropriate agencies of the Government.” 3 Id. § 1184(a), (c). In carrying out this mandate, the Attorney General has sought the advice of DOL.
The Immigration and Naturalization regulations require a petitioning employer to obtain
certification from the Secretary of Labor ... stating that qualified persons in the United States are not available and that the employment of the [foreign worker] will not adversely affect wages and working conditions of workers in the United States similarly employed____
8 C.F.R. § 214.2(h)(3)(i)(A) (1985). Pursuant to this regulatory directive, DOL has in turn adopted regulations establishing procedures that agricultural employers must follow in order to obtain the necessary certification. See 20 C.F.R. §§ 655.200-655.212 (1984). The regulations pertinent to this case condition certification upon an employer’s acceptance of a specially computed wage rate called the “adverse effect rate” (“AER”) as the minimum hourly wage to be paid to both foreign and domestic employees. See id. §§ 655.200(b), 655.-202(a)(9), 655.207.
DOL has designed the AER to ensure that importation of H-2 workers will not depress the wages of similarly employed domestic laborers. Id. § 655.200(b). If DOL finds that employment of aliens in a particular area or occupation has not depressed the wages of domestic workers, it will set the AER at a level equal to the prevailing wage in that area or occupation. Id.; see id. § 655.207(a). On the other hand, if DOL determines that employment of aliens has depressed the wages of domestic workers, it will designate a “wage rate higher than the prevailing wage rate” as the AER. Id. § 655.200(b); see id. § 655.207(b). Underlying this controversy is DOL’s determination that the wages of domestic agricultural workers in a number of states have been adversely affected by importation of H-2 workers.4 Accordingly, DOL has imposed an AER higher than the prevailing wage on agricultural employers seeking to hire aliens to work in those states.
Historically, DOL has set AERs on an annual basis. From 1968-1981, DOL computed annual changes in AERs by using data contained in a United States Department of Agriculture (“USDA”) survey, which covered wages paid to farm workers during one week in each calendar quarter. To calculate the current AER for a particular state, DOL would adjust the previous year’s AER for that state by the same percentage as the change in the annual average wage rates as reported by USDA. Until 1981, USDA had calculated the annual average rates on the basis of quarterly surveys. In July 1981 USDA announced that it was abandoning its quarterly survey in favor of an annual survey measuring only one week in July. DOL believed that it could not rely on the annual survey in calculating AERs because the survey would not measure wages paid during the harvest seasons for the principal crops in *138which H-2 workers are employed.5 The action taken by DOL in developing a substitute methodology for AERs has embroiled the agency in many lawsuits,6 including this one.
In 1982 DOL announced that the 1981 AERs would remain in effect for 1982 and that it was considering alternative methodologies for use in future years. See 47 Fed.Reg. 37,980 (Aug. 27, 1982). Litigation brought in the District Court for the District of Columbia by domestic farm laborers located in four states resulted in court orders directing DOL to establish new AERs for use in 1982.7 Accordingly, DOL adopted an interim methodology under which it calculated new AERs for the four states covered by the court orders, using USDA data from the first two quarters of 1981.8 See 48 Fed.Reg. 232-35 (Jan. 4, 1983). Commenters on the 1982 methodology supported DOL’s decision that the USDA annual survey did not provide an adequate basis for computing annual changes in AERs. For example, USDA commented that it saw no justification for using the annual survey to calculate AERs and that it knew of no appropriate data that were being collected by any agency. Some employers commented favorably on the interim measure but also stated that a new methodology should be developed for use in future years. See id. at 233.
In publishing a final rule establishing 1982 AERs, DOL stated that it was not adopting a methodology for use beyond 1982, but the final rule itself was open to the interpretation that DOL intended that 1982 AERs for the four affected states and 1981 AERs for all other states would remain in effect indefinitely. See id. at 232, 235. This interpretation formed the basis for successful challenges by domestic farm workers to the 1982 regulation. See NAACP, Jefferson County Branch v. Donovan, 566 F.Supp. 1202, 1206-07 (D.D. C.1983). On June 28, 1983, the district court in NAACP v. Donovan, supra, ordered DOL to develop a new methodology to calculate 1983 AERs for all states, holding that DOL’s regulations required setting AERs on a yearly basis. In complying with that order, DOL proposed and finally adopted an amendment to 20 C.F.R. § 655.-207(b), which established a new methodology for calculating annual changes in AERs. That amendment is challenged in this case.
In a notice of proposed rule-making published on July 22, 1983, DOL invited comment 9 on an AER methodology that would rely on data supplied by the Bureau of Labor Statistics (“BLS”) through the Employment and Wages Program (“ES-202 Program”). The ES-202 Program is conducted by BLS in cooperation with state employment security agencies. See 1 BLS Handbook of Methods 32 (Bulletin No. 2134-1 Dec. 1982). The program generates wage data in the following manner: Each quarter, the state agencies receive reports *139from employers covered by unemployment insurance, setting forth number of employees, total wages, taxable wages, and unemployment insurance contributions. In turn, the state agencies provide BLS with quarterly reports showing the number of covered employers, employment during the midweek of each month, and total wages paid during the quarter. The state agencies supply this information to BLS under codes identifying categories of employers. See id. at 32-34. For its new AER computations DOL proposed to use ES-202 data reported under codes covering those employers hiring the bulk of foreign agricultural workers. See 48 Fed.Reg. 33,685 (July 22, 1983). In essence, DOL proposed an AER methodology identical to that used in years when the USDA quarterly survey data were available but substituting the relevant ES-202 data: Each year DOL would adjust AERs by the annual percentage change in average weekly wages reflected in the ES-202 data under the appropriate agricultural codes for the state. In announcing this proposal, DOL observed that “[t]he broad universe coverage, continuity, and currency of the ES-202 program make its data one of the most useful data bases for determining wage movements in all United States industries.” Id.
DOL received over 200 comments to the proposed rule. Growers uniformly objected to use of the ES-202 data and urged use of the USDA annual survey.10 USDA also objected to reliance on ES-202 data, repeated its observation that no agency was collecting appropriate data and advocated continued use of its annual survey as an interim measure. On the other hand, some farm workers commented favorably on the proposed methodology, noting that the ES-202 data base provided a realistic measure of farm wage trends.11
In a notice of final rule-making issued on September 2, 1983, DOL announced its decision to adopt the proposed methodology with modifications, published an amended version of 20 C.F.R. § 655.207(b),12 and set AERs for 1983. See 48 Fed.Reg. 40,168-75 (Sept. 2, 1983). In announcing the final rule, DOL discussed a list of flaws in the ES-202 data base identified by commenters, made some adjustments in the proposed methodology based on commenters’ concerns,13 and explained why it rejected commenters’ alternative proposals. See id. at 40,170-72. Noting that the rule changed the data series by which it had tracked wage movements for purposes of adjusting AERs, DOL indicated that such a change was within its discretion in responding to the “critical need” to establish 1983 AER’s created by the impending harvest season and the order in NAACP, Jefferson County Branch v. Donovan, supra. See 48 Fed.Reg. 40,173-74 (Sept. 2, 1983).
The appellees disagreed with DOL’s view of its discretion and brought this suit in the *140District Court under the Administrative Procedure Act, 5 U.S.C. §§ 701-706 (1982), alleging that DOL’s action in adopting the new methodology was arbitrary and capricious, see id. § 706(2)(A). The appellees point to defects in the ES-202 data that they believe are so serious as to render DOL’s use of the data an abuse of discretion. They further assert that, in promulgating the final rule, DOL ignored comments criticizing its proposal and failed to consider alternative methodologies. DOL moved for summary judgment, contending that adoption of the regulation was within its discretion. Believing that the administrative record did not support the reasonableness of DOL’s action, the District Court denied the motion.14
Discussion
The parties agree that the action taken by DOL in this case constitutes “rule making,” see 5 U.S.C. § 553, and that the scope of our review is governed by the standard set forth in id. § 706(2)(A). Accordingly, we must uphold DOL’s action unless it was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Id. It is well settled that under this standard a reviewing court must decide if the agency’s 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.” Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 824, 28 L.Ed.2d 136 (1971) (citations omitted); see Connecticut v. Environmental Protection Agency, 696 F.2d 147, 155 (2d Cir.1982); Friends of the Earth v. Environmental Protection Agency, 499 F.2d 1118, 1123 (2d Cir.1974). An agency engaging in rule-making is not required to make “specific and detailed findings and conclusions of the kind customarily associated with formal proceedings.” National Nutritional Foods Ass’n v. Weinberger, 512 F.2d 688, 701 (2d Cir.), cert. denied, 423 U.S. 827, 96 S.Ct. 44, 46 L.Ed.2d 44 (1975). However, the agency is obligated to examine the available evidence and to articulate a “ ‘rational connection’ ” between that evidence and its exercise of discretion. Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866-67, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 245, 9 L.Ed.2d 207 (1962)). This explanation must be sufficiently detailed to permit effective judicial review, see Camp v. Pitts, 411 U.S. 138, 142-43, 93 S.Ct. 1241, 1244, 36 L.Ed.2d 106 (1973) (per curiam); National Nutritional Foods Ass’n v. Weinberger, supra, 512 F.2d at 701; see also Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., supra, and, to some extent, should take account of evidence placed be*141fore the agency by interested parties, Consumers Union of United States, Inc. v. Consumer Product Safety Comm’n, 491 F.2d 810, 812 (2d Cir.1974).
At the outset, we note that this is not a case in which an agency changed its policy with no explanation of the reasons for that action. See Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., supra. Under those circumstances, the Supreme Court has declined to uphold agency action even if appellate counsel later offers reasons for the agency’s decision and the decision is one as to which the agency has developed expertise. See id., 463 U.S. at 48-50, 103 S.Ct. at 2869-70. The Court has cautioned that reviewing courts should not attempt “to make up for ... deficiencies” in the administrative record but rather should uphold agency action, if at all, on the basis articulated by the agency. Id. at 43, 103 S.Ct. at 2867. Far from being deficient, the administrative record in this case amply explains DOL’s reasons for abandoning the prior methodology and for deciding to use ES-202 data. In both the 1982 and 1983 rule-making proceedings, DOL offered cogent reasons explaining why it could no longer rely on farm wage data collected by USDA. See 47 Fed.Reg. 52,198 (Nov. 19, 1982); 48 Fed.Reg. 232 (Jan. 4, 1983); id. at 33,684 (July 22, 1983); id. at 40,168, 40,172 (Sept. 2, 1983). Moreover, in 1983, DOL explained the timing of its decision to revise the AER methodology; it is undisputed that the upcoming harvest season and the existing court order made it necessary for DOL quickly to adopt a methodology by which, at the very least, it could compute AERs for 1983.
Federal courts considering DOL’s calculation of AERs for prior years have recognized that the agency enjoys considerable discretion in ensuring that importation of H-2 workers does not adversely affect domestic workers’ jobs. See Florida Sugar Cane League, Inc. v. Usery, 531 F.2d 299, 304 (5th Cir.1976); see also Rowland v. Marshall, 650 F.2d 28, 30 (4th Cir.1981). Neither the Immigration and Nationality Act nor the regulations promulgated thereunder specify any formula that DOL must use in carrying out the statutory purpose. Rather, under the “multi-tiered scheme of delegated statutory authority,” DOL has the ultimate responsibility to develop some reasonable formula that will protect the wages and working conditions of domestic laborers. Florida Sugar Cane League, Inc. v. Usery, supra, 531 F.2d at 300-01, 304. Our review of this administrative record, in light of the objections raised by the appellees, persuades us that DOL’s promulgation of the new AER methodology was a proper exercise of agency discretion.
The appellees point to certain flaws in the ES-202 base, which they believe demonstrate that DOL’s action was arbitrary and capricious. All of these alleged defects were urged on DOL by commenters to the proposed rule and DOL, in promulgating the final rule, explicitly considered each flaw raised by the appellees in this litigation and explained why the flaw did not render the data inappropriate. See 48 Fed.Reg. 40,170-72 (Sept. 2, 1983). For example, the appellees point out that, though the AER is an hourly wage, ES-202 provides a weekly earnings statistic, and that, though the AER is designed to protect wages of domestic farm workers, ES-202 reflects some information concerning non-agricultural employees. The appellees assert that the ES-202 base is not representative because it does not reflect figures for small farms.15 They also claim that the state agencies do not supply the data in a timely fashion.16 DOL meticulously re*142sponded to each of these defects and many others urged by commenters and gave reasoned explanations for believing that the ES-202 Program nonetheless provided a reliable statistical base.17 Under those circumstances, DOL plainly satisfied its obligation to supply a detailed and rational explanation of its decision to adopt the ES-202 system. DOL was not required, as appellees contend, to rebut every alleged flaw with empirical data.
The appellees further assert that DOL abused its discretion by failing to consider “relevant factors,” Citizens to Preserve Overton Park, Inc. v. Volpe, supra, 401 U.S. at 416, 91 S.Ct. at 823, when it decided to adopt the ES-202 system. The primary example of an allegedly overlooked factor is adverse economic impact of the new system on growers; the appellees believe that the system will create erratic and unreasonably large increases in AERs. This claim is without force. First, it is clear that DOL did consider this factor. For example, DOL explained that it found the ES-202 methodology to be superior to sys-terns based on hourly wages because studies showed that the latter might tend to produce erratic changes in AERs. See 48 Fed.Reg. 40,170 (Sept. 2, 1983). DOL noted that reports cited by some growers showed that average hourly earnings reflected greater percentage increases than average weekly earnings so that reliance on hourly earnings would in fact lead to larger increases in AERs than those produced using ES-202.18 Id. Second, DOL used the ES-202 methodology to calculate AERs for 1983, and there is no showing that the 1983 rates reflect erratic or unreasonably large increases over the prior rates.19 Third, comparison of 1983 AERs calculated using ES-202 data and those calculated using the USDA annual survey demonstrated that ES-202 did not produce an erratic or unreasonable increase; the record shows that ES-202 and the USDA annual survey, which was advocated by the appellees, yielded approximately the same aggregate percentage increases in the 1983 AERs.20 This comparison supports the rea*143sonableness of DOL’s action, see Rowland v. Marshall, supra, 650 F.2d at 30 n. 3, and undercuts the appellees’ argument that the chosen methodology places an unreasonable economic burden on growers.
Finally, relying on Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., supra, the appellees contend that DOL abused its discretion because it gave no consideration to alternative methodologies. The administrative record flatly contradicts this position. Initially, this assertion ignores the history leading up to adoption of the ES-202 methodology. Deprived of the data series on which it had relied for more than ten years, DOL was informed by USDA that no government agency was engaged in collecting data appropriate for DOL’s purposes. With its choices thus very limited, DOL then attempted, concededly under pressure of litigation, to devise an alternative. In 1982, DOL considered and adopted an interim methodology as an alternative to leaving all 1981 AERs in effect.21 Believing that the interim rule was inadequate for use in future years, DOL then took steps to adopt a new rule. The notice of final rule-making establishes that DOL considered alternatives. DOL repeated its reasons for deciding that a methodology based on the USDA annual survey was inappropriate, 48 Fed.Reg. 40,172 (Sept. 2, 1983), rejected an alternative system advocated by farm workers that would have produced a wage rate higher than the ES-202 system, id. at 40,170, and explained that it rejected other alternatives suggested by employers, including a system that would base AERs on the federal minimum wage rate, id. at 40,-172. DOL gave considered responses to all of these proposals and explained why it continued to favor the ES-202 methodology. There is no broad requirement that an agency must consider “all policy alternatives in reaching decision,” Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., supra, 463 U.S. at 51, 103 S.Ct. at 2871.
We do not doubt that commenters offered some significant criticisms of the ES-202 methodology. But it is the business of the agency and not of the reviewing court to decide if commenters’ objections require modification or even abandonment of proposed action. Based on this administrative record, which contains clear and reasonably detailed explanations of the agency’s reasons for the choice made, we find no basis for holding that DOL failed to consider available evidence or otherwise made its choice arbitrarily. We hold that the District Court erred in denying DOL’s motion for summary judgment. Accordingly, the case is remanded to the District Court with instructions to enter summary judgment in favor of DOL.
. The District Court granted the plaintiffs’ cross-motion for partial summary judgment with respect to the validity of a regulation governing piece rates required to be paid by agricultural employers hiring foreign workers. The plaintiffs argued that DOL provided inadequate notice and opportunity for comment on an amendment to the piece rate regulation. Though DOL originally sought review of the District Court’s ruling on this issue, it later withdrew that portion of its appeal because it believed that a new rule-making proceeding would moot the controversy. See 49 Fed.Reg, 48,061 (Dec. 10, 1984).
. Plaintiff Shoreham Cooperative Apple Producers Association, Inc. is a cooperative of Vermont apple growers. Plaintiff New England Apple Council, Inc. is a Massachusetts corporation, representing New England apple growers. The remaining plaintiffs are individual apple growers with orchards located in Vermont.
. The Attorney General has delegated his rule-making authority under the Immigration and Nationality Act to the Commissioner of Immigration and Naturalization. See 8 C.F.R. § 100.6 (1985).
. Under the regulation now in effect, 20 C.F.R. § 655.207(b) (1984), DOL computes annual AERs for the following states: Arizona, Colorado, Connecticut, Florida, Maine, Maryland, Massachusetts, New Hampshire, New York, Rhode Island, Texas, Vermont, Virginia, and West Virginia. Id. § 655.207(b)(2). "[F]or purposes of wage movement, but not actual” ■ AERs, New York, Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont are treated as one state, and Maryland, Virginia, and West Virginia are treated as one state. Id. § 655.207(b)(1).
. Most foreign agricultural workers are employed as sugar cane cutters or as apple pickers. See 46 Fed.Reg. 4,569 (Jan. 16, 1981).
. See Bragg v. Donovan, No. 82-2361 (D.D.C. Aug. 25, 1982); NAACP, Jefferson County Branch v. Donovan, 566 F.Supp. 1202 (D.D.C.1983); Florida Fruit & Vegetable Ass’n v. Donovan, 583 F.Supp. 268 (S.D.Fla.1984); Virginia Agricultural Growers Ass’n v. Donovan, 597 F.Supp. 45 (W.D.Va.1984); Production Farm Management v. Donovan, No. 84-143 (D.Ariz. Oct. 17, 1984); see also NAACP, Jefferson County Branch v. Donovan, 558 F.Supp. 218 (D.D.C.1982) (challenge to piece rate regulation), subsequent injunction vacated, 737 F.2d 67 D.C.Cir.1984).
. See Bragg v. Donovan, No. 82-2361 (D.D.C. Aug. 25, 1982) (Maine, Vermont, and Florida); NAACP, Jefferson County Branch v. Donovan, 558 F.Supp. 218 (D.D.C.1982) (West Virginia).
. For all other states, 1981 AERs remained in effect. See 48 Fed.Reg. 232-35 (Jan. 4, 1983).
. The district court’s decision in NAACP, Jefferson County Branch v. Donovan, supra, issued on June 28, 1983, ordered DOL to promulgate 1983 AERs no later than July 29, 1983. See 566 F.Supp. at 1210. DOL therefore was required to act quickly. The notice of proposed rule-making, published on July 22, 1983, requested written comments no later than August 5, 1983. 48 Fed.Reg. 33,684 (July 22, 1983). Many parties wrote to DOL asking for more time to comment. DOL responded by extending the comment period to August 22, 1983. See 48 Fed.Reg. 35,667 (Aug. 5, 1983).
. Among the commenters was plaintiff-appel-lee Shoreham Cooperative Apple Producers Association, Inc.
. Most farm workers who commented on the proposed regulation focused their criticisms on DOL's decision to amend the piece rate regulation. See 48 Fed.Reg. 40,173 (Sept. 2, 1983).
. DOL also published an amended version of the piece rate regulation. See 48 Fed.Reg. 40,-175 (Sept. 2, 1983). As a result of the District Court’s order in this case holding that DOL’s procedures in adopting that rule were defective, DOL decided to withdraw the rule and engage in a new rule-making proceeding. See 49 Fed. Reg. 48,061 (Dec. 10, 1984).
. Significantly, DOL responded to one criticism by making an adjustment in the rule that resulted in lower AERs for most states. Under the proposed rule, DOL planned to compute the 1983 AERs by adjusting 1981 AERs “by the same percentage as the change in the appropriate ES-202 data during 1979-81.” 48 Fed.Reg. 40,-171 (Sept. 2, 1983). Some growers objected to this computation, pointing out that, "since the 1981 [AERs] had been determined using the changes in the USDA data for 1979-80,” DOL’s proposal would result in “ ‘double-counting’ wage inflation for 1979-80.” Id. DOL believed thát this criticism was “valid” and responded by modifying the final rule "to use changes in the best available ES-202 data for 1980-82,” thereby avoiding the problem raised by the commenters. Id. As a result of this modification, the 1983 AER for Vermont was reduced from the $4.69 computed under the proposed rule to $4.28. Compare 48 Fed.Reg. 33,686 (July 22, 1983) with 48 Fed.Reg. 40,170 (Sept. 2, 1983).
. Noting the difference of opinion among the district courts that have ruled on the validity of the ES-202 methodology, compare Virginia Agricultural Growers Ass’n v. Donovan, 597 F.Supp. 45 (W.D.Va.1984) (appeal pending) (denying DOL’s motion for summary judgment and holding ES-202 system invalid after trial), with Florida Fruit & Vegetable Ass’n v. Donovan, 583 F.Supp. 268 (S.D.Fla.1984) (appeal pending) (granting DOL’s motion for summary judgment after finding that administrative record supported reasonableness of DOL’s action); cf. Production Farm Management v. Donovan, No. 84-143 (D.Ariz. Oct. 17, 1984) (appeal pending) (granting DOL’s motion for summary judgment in action alleging DOL had failed to establish adverse effect on domestic farm workers), the District Court certified in its order denying the motion for summary judgment that the order involved controlling questions of law as to which there were substantial grounds for difference of opinion and that immediate review under 28 U.S.C. § 1292(b) would advance the litigation. We agreed with the District Court and therefore granted DOL’s motion for interlocutory review. Though the parties invite us to consider whether the District Court is entitled to hold an evidentiary hearing to determine the validity of the new AER methodology, see Camp v. Pitts, 411 U.S. 138, 93 S.Ct. 1241, 36 L.Ed.2d 106 (1973) (per curiam), we do not reach that issue, nor decide whether the ruling ordering a hearing is subject to our grant of section 1292(b) review, since we conclude that the District Court should have granted appellant's motion for summary judgment.
. With respect to comments asserting that AERs should not be adjusted by changes in average weekly earnings, DOL responded by noting that studies indicated that the ES-202 system was likely to produce more reasonable increases than a system based on hourly earnings. See 48 Fed.Reg. 40,170 (Sept. 2, 1983). Acknowledging comments pointing out that ES-202 reflects data concerning non-agricultural workers, DOL emphasized that the actual earnings of workers covered by the ES-202 Program were not being used as AERs. In DOL’s view, the wages of non-agricultural workers reflected in the base did not rise or fall at significantly different rates from wages of agricultural employees. Id. at 40,171. DOL considered comments asserting that the ES-202 Program was not representative in that it surveys only those farms covered by unemployment insurance. DOL responded by noting that BLS estimated that ES-202 covered 40% of agricultural workers. “Data on earnings paid to such a large portion of workers on agricultural payrolls is not insignificant. DOL has determined that it represents the best data available at this time on wage movements in agriculture.” Id. Moreover, DOL points out that a system based on the USDA annual report, which measures wages paid during one week in July, would not make it possible to adjust AERs more promptly.
. This assertion is based on the fact that BLS receives data from state agencies only for farms covered by unemployment insurance, which typically employ at least ten workers or have a $20,000 quarterly payroll. The appellees claim ' that commenters informed DOL that large farms represent a minority of agricultural employers and that wages on large farms are higher than, and move at rates different from, those on small farms.
. For example, the appellees claim that ES-202 data, for 1983 were not available until August 1984. Timeliness is important to the growers, who allege that they need to know the relevant *142AER in July, the month when they begin recruiting H-2 workers.
. We note that the prior methodology based on the USDA quarterly survey occasionally produced large increases, which were sustained by the courts. See, e.g., Rowland v. Marshall, supra (sustaining $3.50 AER imposed on Virginia employers for 1981, which represented more than 18% increase, see 45 Fed.Reg. 30,734 (May 9, 1980) (establishing AER of $2.96 for 1980)).
. Moreover, the 1984 AER for Vermont, which also was set under the ES-202 system, reflected an increase of only 1.4% increasing from $4.28 to $4.34. See 49 Fed.Reg. 31,784 (Aug. 8, 1984).
. The following table sets forth AERs computed on the basis of the ES-202 system and those computed on the basis of report: the USDA annual
State ES-202 USDA
Arizona $4.22 $4.11
Connecticut 4.05 3.99
Florida (sugar cane only) 5.37 5.15
Florida (except sugar cane) 4.34 4.17
Maine 4.15 4.10
Maryland 4.38 4.53
Massachusetts 4.05 3.99
New Hampshire 4.34 4.28
New York 4.20 4.15
Rhode Island 4.05 3.99
Texas 4.16 4.73
Vermont 4.28 4.22
Virginia 4.39 4.54
West Virginia 4.24 4.32
. We note that, in proposing the interim methodology for 1982, DOL discussed several alternatives to a rule based on USDA data for 1981, which was finally adopted. See 47 Fed.Reg. 52,199 (Nov. 19, 1982).