Opinion for the Court filed by Circuit Judge SILBERMAN.
SILBERMAN, Circuit Judge:We consider for the second time the Department of Labor’s new methodology for computing the adverse effect wage rate (“AEWR”), which is the minimum wage that employers who wish to hire aliens as temporary agricultural workers must offer American and foreign workers. The Department appeals the district court judgment rejecting its final rule adopting this new methodology. Because we find that the Department has adequately explained its change in regulatory policy and that its new policy is within statutory authorization, we reverse.
I.
Although the Department had, under the Immigration and Naturalization Act of 1952 (“INA”), long issued regulations which set forth the minimum wage at which foreign agricultural workers could be employed, after the passage of the Im*361migration Reform and Control Act of 1986 (“IRCA”), the Department took a new tack. It ended its policy of enhancing the minimum wage to compensate for past wage depression, choosing instead to base the minimum wage rate on the previous year’s average hourly agricultural wage. The AFL-CIO and other farmworker representatives (“appellees” or “farmworkers”) challenged the new regulations, which reduced the minimum wage by an average of 20 percent from that calculated under the old method, as failing to carry out the Department’s statutory mandate under IRCA of ensuring that domestic wages are not “adversely affected” by foreign labor. 8 U.S.C. § 1188(a) (1988).
Under INA, the Department established the AEWR — a minimum wage that all employers who wished to import alien workers must first offer to qualified U.S. workers, and then, if the job remained unfilled, to foreign workers. See 8 C.F.R. § 214(2)(h)(3) (1986). The methodology used to compute the AEWR was designed to offset both future and past adverse effect from the addition of legal and illegal foreign labor to the U.S. labor supply.1 The wage floor is obviously designed to prevent cheaper foreign labor from undercutting domestic wages in the future. By adding a factor calculated to offset the supposed degree to which past legal and illegal additions to the work force had depressed agricultural wages, DOL also intended that the AEWR “compensate” domestic farmworkers for what might be thought to be past failures to protect the wage scale.
Congress expressly incorporated the pri- or regulatory requirement that employing foreign workers would not “adversely affect the wages and working conditions of workers in the United States similarly employed,” 8 C.F.R. § 214(2)(h)(3) (1986), as part of IRCA’s amendments of INA governing temporary foreign worker (or H-2A) visas. See 8 U.S.C. § 1188(a).2 Congress did not, however, further define adverse effect and left it in the Department’s discretion how to ensure that the importation of farmworkers met the statutory requirements. The Department chose to protect domestic workers’ interests, as it had done under INA, through its longstanding AEWR program, but it modified the program in light of IRCA.3
Since IRCA was primarily designed to reduce illegal immigration, the Department thought it would be necessary and appropriate to ease temporary legal entry. Accordingly, it adopted a new, simpler methodology in which the adverse effect wage rate would be the previous year’s annual regional average hourly wage for agricultural workers (the USDA average wage) with no added adjustments.4 See 52 Fed. Reg. 20,496, 20,502-505 (1987) (to be codi*362fied at 20 C.F.R. §§ 654 & 655) (interim final rule) (June 1, 1987).
Appellees challenged the new regulations as violating an alleged statutory mandate to compensate for past wage depression and, furthermore, arbitrarily changing the methodology without adequate explanation. The district court held that the regulations were illegal “because the particular method chosen will not carry out the Secretary's statutory responsibility to address the depressed wages of American workers.” AFL-CIO v. Brock, 668 F.Supp. 31, 39 (D.D.C.1987). On appeal, this court reversed, AFL-CIO v. Brock, 835 F.2d 912, 913 n. 2 (D.C.Cir.1987) (“Brock ”), holding that the statute neither “explicitly [n]or implicitly mandates the Department’s AEWR policy____ [N]either the Department’s former policy of offsetting for past depression nor its present policy of ignoring this adverse effect, is statutorily required.” Id. at 917. But because the Department had not adequately explained why it was altering its method of computing the AEWR, the court remanded for a reasoned explanation, noting that the “[i]nability to secure persuasive data as to any effects of past wage depression might indeed justify ending the enhancement.” Id. at 919.
The interim rule was mooted by the Department's publication of the final rule, in which, after notice and comment, the Department adopted the same methodology. See Final Rule, 54 Fed.Reg. 28,037. Once again the farmworkers sought to invalidate the rule, and once again the district court agreed, holding that “where a policy has been in place for many years an agency must rely on something .more than admittedly inconclusive data if it chooses to vary its course.” AFL-CIO v. Dole, 745 F.Supp. 18, 22 (D.D.C.1990) (“Mem. Op.”). The government appealed, asserting that the choice of methodology was within its discretion. We granted a stay of the district court’s order which enjoined implementation of the new rule and reinstituted the pre-June, 1987 methodology.
II.
DOL summarized its rationale for changing its approach as follows:
(1) Best available studies fail to provide evidence of any adverse effect (much less any quantification of such effect) at State/regional/national levels; this lack of evidence of adverse effect is supported by theoretical analysis; ... (2) any localized adverse effects are probably not large enough to show up in USDA series; (3) thus, DOL has no basis and no right to impose an arbitrary enhancement factor; (4) to the extent there are localized pockets [of depressed] wage rates [which] are below the State/regional average, requiring the average as a minimum will “cure” this adverse effect;
Final Rule, 54 Fed.Reg. at 28,047 (1989) (Analysis of Farmworker Comments).
The evidence presented to the Department showed that the data on wage depression was, at best, equivocal. See id. at 28,043-44.5 It could be observed, if at all, only at the local, crop-specific level. See id. at 28,042-43. Some economists, moreover, challenged the underlying hypothesis that increasing alien labor depresses wages; given “the great elasticity of demand for labor, particularly in the highly internationally competitive agricultural markets of the 1980's ... wage rates would have been about the same, with far less production and employment.” Id. at 28,047.
Even if domestic wage depression resulted from the importation of foreign farm-workers, studies showed that it could not be precisely quantified at the aggregate level at which regulation is administratively feasible. DOL's review of its past methodologies for adjusting for wage depression revealed that there is no correlation between its past upward adjustments to the AEWR and any putative level of wage depression. See id. at 28,041, 28,044. The Department concluded that, without ade*363quate measurement techniques, its previous attempts to compensate for past wage depression had been no more than shots in the dark and it could not justify continuing to augment wages by an arbitrary factor. Continued use of the old methodology would create severe wage disparities and impede IRCA’s goal of replacing illegal aliens with documented foreign workers.
The Department examined and rejected the alternative methodologies suggested by both the growers and the farmworkers. The growers’ proffered option of using the prevailing wage (which would be lower than the AEWR) would not, in the Department’s view, offset the crop-specific wage depression that had been shown to exist; continuing to employ the old methodology, as urged by appellees, would arbitrarily perpetuate wage anomalies and create artificially high wages that could cost jobs in the long run.
The Department concluded that the USDA data series that it had been using to index the old AEWR would best serve statutory purposes as the AEWR. The Department believed that the USDA series provided the most reliable wage data covering relevant jobs. Evidence showed, moreover, that the USDA average hourly wage is largely untainted by wage depression, and that the documented incidents of localized adverse effect were too small to affect the USDA data. Use of the USDA average wage was thus the best option to accommodate the interests of both growers and farmworkers. See id. at 26,044.
While the growers accepted the new approach, the farmworkers continued to oppose the Department’s change in policy on the grounds that the Department’s explanation understated the radical nature of the change in the method of computing the AEWR, that DOL had not viewed the record evidence fairly, and that without conclusive evidence that the USDA survey was free of all wage depression, the Department could not abandon its past practice of enhancing the base rate to compensate for past depressive effect.
This disagreement between the Department and appellees (and the district court) over how much the new policy departs from prior AEWR methodologies is somewhat irrelevant, however, because, as we concluded in Brock, the Department is entitled to change its policy so long as it supplies a reasoned explanation for its choice. See 835 F.2d at 917 (quoting Atchison, Topeka & Santa Fe Ry. Co. v. Wichita Bd. of Trade, 412 U.S. 800, 808, 93 S.Ct. 2367, 2375, 37 L.Ed.2d 350 (1973)). We think that the Department’s detailed and rational explanation, with its accompanying consideration of and answers to the criticisms of farmworker representatives, amply satisfies our instructions on remand and meets the standard of “reasoned analysis” enunciated in Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 42, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983).
Appellees’ other arguments are also off target. They quarrel with the Department’s evaluation of the studies cited in the administrative record, particularly those of Dr. Peter Martin (“Martin”), and the General Accounting Office (“GAO”). Appellees argue that the Martin study was based on assumptions to which they were not privy. Since appellees had access to the study itself during the notice and comment period we do not see how they were harmed. More to the point, appellees admit, as they must, that the Martin study supports the Department’s opinion that wage depression is not precisely quantifiable. To be sure, Martin states that there is no conclusive evidence on whether or not the USDA average wage is depressed by alien workers (it may even be raised by their purported higher productivity). But that one could not be certain of the effect of past immigration on the USDA average rate does not make its use unreasonable. And that, after all, is the issue.
The GAO study is no more helpful to appellees. The Department does not dispute, as the GAO determined, that there are local instances of wage depression. But, like Martin, the GAO study recognized that a precise measurement of adverse effect is “technically and administratively infeasible,” that the Department must there*364fore find a method to measure and prevent a hypothetical adverse effect, that one such device is the USDA average wage, and that DOL’s AEWR regulations “necessarily include a judgmental component choosing among reasonable methodologies to prevent adverse effect.” GAO, The H-2A Program: Protections for U.S. Farmworkers 64, 68 (1988); J.A. at 935, 939.
Still, appellees argue that the Department impermissibly abandoned its past policy of adding an upward adjustment for past wage depression, merely because such adverse effect cannot be precisely measured, rather than attempting to find another proxy for supposed wage depression. The farmworkers’ argument, however, assumes (as did the district court, see Mem. Op. 745 F.Supp. at 22) a statutory requirement to adjust for past wage depression, a view we rejected in our prior opinion in Brock. It is clear that it is within the Department’s discretion to end the enhancement policy. Since there is no conclusive evidence — indeed, appellees offer none — that the USDA average is tainted by wage depression, its use seems well within the range of reasonable methodological choices open to the Department.
The government’s choice of methodology is really a policy decision taken within the bounds of a rather broad congressional delegation. The Department is obliged to balance the competing goals of the statute— providing an adequate labor supply and protecting the jobs of domestic workers. See Rogers v. Larson, 563 F.2d 617, 626 (3rd Cir.1977), cert. denied, 439 U.S. 803, 99 S.Ct. 57, 58 L.Ed.2d 95 (1978). Striking that balance is a judgment call which Congress entrusted to the Department of Labor. See Brock, 835 F.2d at 915 n. 5; see also Shoreham Cooperative Apple Producers Ass’n v. Donovan, 764 F.2d 135, 141-143 (2d Cir.1985). The Department chose, in the face of imprecise and inconclusive data, the USDA hourly wage as the AEWR because, in its opinion, that rate will neither ratchet wages upward, driving growers out of business nor perpetuate wage depression. See 54 Fed.Reg. at 28,-044-45.
The district court dismissed the Department’s reasons for its policy choice as “post hoc rationalizations,” which demonstrated that DOL did not evaluate the evidence on remand with “an open mind.” Mem. Op. 745 F.Supp. at 21. Appellees echo these objections before us, but we do not see why the Department’s adherence to its policy position reflected in the interim rule is at all relevant to our review. Only in rather extraordinary circumstances do we examine an agency’s motivation. See, e.g., Sierra Club v. Costle, 657 F.2d 298, 408-410 (D.C.Cir.1981). And when an agency on remand is giving a fuller explanation of its prior decision, its reasoning is in some sense necessarily post hoc.
In the same vein, the district court accused the Department of “playing politics with the AEWR,” because it failed to protect its intended beneficiaries — the farm-workers — and was defended in court by those “normally adversely affected” by the rate — the growers. Mem. Op. 745 F.Supp. at 22. We had not thought that a government position could be discredited by the character or nature of organizations that support the government in court. But, in any event, the statute requires that the Department serve the interests of both farmworkers and growers — which are often in tension. That is why Congress left it to DOL’s judgment and expertise to strike the balance.
Appellees’ objections to the Department’s change in method are therefore meritless. Were we to require that DOL hew to its old methodology until it had conclusive data on which to base a change, see Mem. Op. 745 F.Supp. at 22, we would lock the Department into its previous policy. Instead, we are entitled to ask only that DOL demonstrate that the data is inconclusive and “identify the considerations it found persuasive in making its decision.” American Tel. & Tel. Co. v. FCC, 832 F.2d 1285, 1291 (D.C.Cir.1987) (citation omitted). Because the record clearly shows that the data on adverse effect is inconclusive, and the Department provided an ample explanation of the considerations *365it found persuasive, we reverse the district court and uphold the final rule.
It is so ordered.
. DOL has used a number of methods for calculating the AEWR over the years, first measuring and using the prevailing wage rate, then adopting various rough proxies for the supposed degree to which prevailing wages have been depressed by foreign labor. See Adverse Effect Wage Rate Methodology, 54 Fed.Reg. 28,037, 28,039-28,041 (1989) (to be codified at 20 C.F.R. § 655) ("Final Rule").
. Section 1188 in pertinent part requires the Secretary of Labor to certify that:
(A) there are not sufficient workers who are able, willing, and qualified, and who will be available at the time and place needed, to perform the labor or services involved in the petition, and
(B) the employment of the alien in such labor or services will not adversely affect the wages and working conditions of workers in the United States similarly employed.
. Under both INA and IRCA Congress delegated to the Attorney General the authority to grant nonimmigrant farmworker visas. Under INA, the Department’s AEWR regulations were issued pursuant to the Attorney General’s "adverse effect” regulation. In codifying the “adverse effect" standard of the regulation under IRCA, Congress also expressly delegated responsibility to the Department; the AEWR regulations are now promulgated under that statutory authority-
.The previous AEWR was based on a 1950. Census figure, augmented by the increase in manufacturing wages from 1950-1963 and by a flat $0.20, to compensate for past adverse effect. That base rate was then indexed upward each year by the percentage change in the statewide annual average hourly wage rate for agricultural workers, published quarterly by the United States Department of Agriculture (“USDA"). See Final Rule, 54 Fed.Reg. at 28,-040.
. Some studies found that domestic workers’ wages were reduced in areas with a high concentration of alien labor, while others suggested that domestic workers’ wages actually increased. Id. at 28,041-2.