OPINION

SENTER, Chief Judge.

On February 11, 1988, this court entered an order in this cause establishing that back pay was an appropriate remedy, set*118ting forth the method of computing the back pay award, and setting a hearing on April 11, 1988, to consider two matters relating to the computation of the back pay award. The hearing was held on that date and evidence as to each issue was taken.

Amounts Earnable With Reasonable Diligence

The court requested information from the parties to establish whether the plaintiffs as a class had exercised reasonable diligence and, if not, to set forth a sum which the average individual in Bolivar County would be able to earn with reasonable diligence in the year 1970. The following facts were adduced as evidence:

The average weekly duration of unemployment in the State of Mississippi by race and sex for 1978-19871 is:

Male Female White Nonwhite

8.73 8.83 8.90 8.61.

The median 1969 earnings for blacks in Mississippi generally and for members of the class in this cause are:

General Class Members

Black Males 2,298 1,910.49

White Females 1,417 1,586.00

1,598.17.

The average class are: 1969 earnings from the

Black Males $2,191.36

Black Females $1,696.71.

The unemployment rate among black males in Bolivar County in 1970 was 13.5%. The unemployment rate among black females in Bolivar County in 1970 was 17.8%. The plaintiff introduced evidence that the rate of unemployment in Bolivar County in 1970 approximated that of the United States in the depths of the Great Depression.

The minimum wage from February 1, 1968, to May 1, 1974, was $1.60.

The court concludes that the presence of minimal earnings in any given year does not give rise to a conclusion of failure to exercise reasonable diligence. Moreover, the presence of minimal earnings in three of the five years will result in the conclusion that a claimant failed to exercise reasonable diligence. A claimant has minimal earnings in any year in which the claimant earned less than 20% of a full-time minimum wage position, or $640.00 in annual income. If a claimant has minimal earnings in three of the five years for which income figures are available, the claimant will have $640.00 in earnings imputed to them for each year in which the claimant earned less, both in calculating the average interim earnings of the class and in computing the distribution of the award. For example, there are thirty-one active claimants on Job Order number 1. Of the thirty-one claimants, the following ten claimants have failed to exercise reasonable diligence:

1969 1970 1971 1972 1973

Tommie Louis Brown 0.00 0.00 0.00 1364.53 4594.79

Bernice Burten 3371.55 2961.69 0.00 0.00 0.00

Mary Lee Causey 0.00 1707.36 752.25 0.00 286.10 Linda O’Neal Cherry 48.78 671.33 1316.19 0.00 0.00

Sallie Mae Foster 0.00 566.97 0.00 0.00 889.70

Dorothy Ann Idleburg 0.00 300.00 0.00 928.00 1365.00

JoAnn Willye Lewis 0.00 207.92 0.00 0.00 0.00

Arena Murry 0.00 0.00 0.00 0.00 0.00

Mary Jefferson Williams 0.00 0.00 1200.00 203.79 1242.08

Leo Towner Wilson 270.00 0.00 123.50 211.82 655.44

Because Job Order number 1 began in 1970, only the last four years are relevant to the calculation of average interim earnings. The ten claimants discussed would have their earnings imputed as follows:

Tommie Louis Brown 1970 1971 1972 1973 640.00 640.00 1364.53 4594.79

*1191970 1971 1972 1973

Bernice Burten 2961.69 640.00 640.00 640.00

Mary Lee Causey 1707.36 752.25 640.00 640.00

Linda O’Neal Cherry 671.33 1316.19 640.00 640.00

Sallie Mae Foster 640.00 640.00 640.00 889.70

Dorothy Ann Idleburg 640.00 640.00 928.00 1365.00

JoAnn Willye Lewis 640.00 640.00 640.00 640.00

Arena Murry 640.00 640.00 640.00 640.00

Mary Jefferson Williams 640.00 1200.00 640.00 1242.08

Leo Towner Wilson 640.00 640.00 640.00 655.44

EQUITABLE LIMITATION OF THE BACK PAY AWARD

The court heard evidence on limiting recovery by the plaintiffs on two groups of claims: those arising from referrals to Baxter-Travenol Laboratories and those arising from out-of-code referrals.

Baxter-Travenol Laboratories

The evidence established that during the period November, 1969, to December, 1970, 67 men were hired as material handlers and 159 women were hired as assemblers. A nondiscriminatory rate of employment for females on job orders requiring a tenth grade education was stipulated as 56.136%. The table below compares actual hires with hypothetical nondiscriminatory hires.

Actual N ondiscriminatory

Male Female Male Female

Material Handler 67 0 29 38

Assembler _0 159 70 89

Total Hours 67 159 99 127

The court concludes that MSES’s referral practices resulted in thirty-two more females being employed than would have been employed had nondiscriminatory referrals been made.

The plaintiffs’ expert, Dr. Marc Bendick, Jr., disputes the argument that more referrals to women as material handlers would necessarily mean fewer referrals to women as assemblers. However, plaintiffs own evidence establishes that in 1971 (when discriminatory referrals for the assembler position ended), women received 161 of 279 assemblers positions or 57.706% of all assembler positions. This is extremely close to the nondiscriminatory percentage.

Plaintiff contends that a limitation of the back pay award to consider benefits to the class is not allowable under Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982), and Furnco Construction Corp. v. Waters, 438 U.S. 567, 98 S.Ct. 2943, 57 L.Ed.2d 957 (1978), because the principal focus of the statute “is the protection of the individual employee, rather than the protection of the minority group as a whole.” Teal, 457 U.S. at 453-54, 102 S.Ct. at 2534. This cause is distinguishable from both of the above-cited cases. This court has found that a greater number of women were referred than would otherwise have been the case. As the two types of positions had identical educational requirements and were filled contemporaneously, the women referred to material handlers positions would necessarily have been the same women who were referred to assemblers positions. Here the very individuals who were disadvantaged by referrals to one job were advantaged by referrals to the other.

In Teal and Waters, the individuals who received the benefit and the individuals who suffered the injury were distinct and separate groups. In Waters, the alleged discriminatory practice was refusal to hire *120firebricklayers at the gate and instead relying on hiring of individuals the company had previously hired as firebricklayers or who had been recommended to the company. The Court found that a prima facie case was established even though the exclusion of those blacks who applied at the gate resulted in an increase in working hours for those blacks who had previously worked for the company or had been recommended to the company. Waters, 438 U.S. at 569-72, 98 S.Ct. at 2945-47. Similarly, in Teal, a test was administered which disproportionately excluded blacks from consideration. The Court found that such a test was not permissible even though those blacks who passed the test were promoted at a disproportionate rate to compensate for the tests effects. Teal, 457 U.S. at 452-56, 102 S.Ct. at 2533-35.

The court concludes in the exercise of its equitable discretion that the award of back pay on the Baxter-Travenol Job Orders should be limited to the sum of $190,000.00 plus interest accruing at the stipulated rate from May, 1970. Such amount should serve to adequately compensate the plaintiff class without providing the plaintiff class with an unjustified windfall and should also serve to deter the defendant from engaging in such behavior in the future without being punitive in effect.

Out of Code Referrals

MSES presented no evidence directly on this issue. The plaintiffs presented evidence that the total pay from the more numerous black referred out-of-code positions exceeded that of the white referred out-of-code positions by 7.51%. The evidence established that there was minimal advantageous effect to the plaintiff class arising from the referral practices. The award on these job orders should not be limited.

Other Matters

In the opinion issued on February 11, 1988, [available on WESTLAW, 1988 WL 72795], the court held that the stipulated method of calculating gross back pay was reasonable and would be utilized in the calculation of gross back pay. Opinion at 21-29. The court thereby implicitly rejected several arguments made by MSES’s counsel. Because counsel urges these alternatives upon the court in briefs submitted, the court will now explicitly reject these contentions and explain the reasons behind this rejection.

REFERRAL/HIRE RATIOS

MSES continues to insist upon an adjustment for a “referral hire” ratio. No such adjustment will be made. MSES’s evidence shows that a given individual has a 32% chance of being employed on a given referral. This means that a single referral of a class member to a given position is only 32% likely to fill the position. However, two referrals to the position would have a 54% chance of filling the position.2 Three referrals would give a 69% chance of filling the position; five referrals would give an 85% chance of filling the position; and ten referrals would give a 98% chance of filling the position. Given the large ratio of potential referrals to potential positions, it is clear that the probability of filling a position would approximate 100% in most cases.

BACK PAY PERIOD

MSES continues to argue that the back pay period should be limited to a very short period of time. In its response to the previous motion for summary judgment, MSES argued for a limitation to one year on the grounds that any longer period would be exceedingly speculative. Now MSES argues that it should be limited to the average duration of unemployment in Mississippi: eight to ten weeks. In a previous stipulation, MSES agreed with plaintiffs that the expected duration of employment would not be estimated and that the gross back pay would instead be reduced by 4.38% per quarter to arrive at a turnover reflected earnings rate. Opinion of February 11, 1988, at 27-28. The court found such a method to be a reasonable *121one. The court reaffirms that finding and now finds that such a method is not unduly speculative.

MSES alternatively contends that the back pay period should end when MSES stopped discriminating in referrals, citing Ford Motor Co. v. EEOC, 458 U.S. 219, 102 S.Ct. 3057, 73 L.Ed.2d 721 (1982). Ford held that an employer could toll the back pay liability by unconditionally offering the claimant the job he had originally sought. Id. at 232, 102 S.Ct. at 3066. Ford is readily distinguishable from this cause. MSES has not offered the individual plaintiffs or class members specific referrals to positions comparable to those for which defendants had earlier denied referrals. Rather, defendants merely ceased discriminating against future applicants. This contention is also rejected. Accordingly, the court reaffirms its approval of the stipulated method and adopts as the gross back pay attributable to each job order the figures reflected in the gross back pay column of the calculations contained in Plaintiffs Calculation of Back Pay and Interest for Each Job Order. (Red Book) dated January 31. 1986.

APPLICATION OF AVERAGE INTERIM EARNINGS

In its alternative methods of calculating back pay, MSES has repeatedly compared unadjusted average interim earnings against adjusted earnings from the job orders. This is clearly incorrect and contrary to the prior order of this court. See Opinion of February 11, 1988, at 43-45. The method stated in the prior opinion will be used.

NO ECONOMIC INJURY

In the prior opinion, the court stated that “no claimant should be included in the recovery pool on any year in which his interim earnings (as modified per the court’s conclusions, infra) exceed the gross back pay award for that job order.” This applies to individual claimants. MSES’s contentions that this eliminates the claims of the class on most job orders are rejected.

SUMMARY

In summary, the court concludes:

(1) that each individual claimant who earned less than $640.00 before adjustment in three or more of the five years for which interim earnings data is available will have income of $640 imputed in each of those years;

(2) that the award of back pay on the Baxter-Travenol Job Orders will be limited to the sum of $190,000.00 plus interest accruing at the stipulated rate from May, 1970;

(3) that plaintiffs shall calculate the award of back pay under the method approved by the Opinion of February 11, 1988, as modified by the above conclusions and submit such calculation to MSES within thirty (30) days of this order;

(4) that defendants shall verify the accuracy of the calculation within thirty (30) days of submission of the calculation and shall then present the calculation to the court for entry of judgment.

An appropriate order shall issue.

Notes

. The earliest period for which data is available.

. Such probability may be computed as 100 (1 — (1 — .32)n) where n is the number of referrals and 1 — .32 is the probability that any given referral will not fill the position.