CHAPTER 2 Direct-Labor Dollars

This chapter provides guidance on the estimating and pricing of direct-labor costs. This guidance addresses: (1) direct-labor hours, (2) direct-labor rates, and (3) other considerations in estimating direct-labor costs.

DIRECT-LABOR HOURS

The first step in estimating direct-labor costs is to estimate direct-labor hours. The source of information for the estimated hours varies according to the nature of the solicitation or request for proposal. For manufacturing and services, a solicitation might provide a statement of work that must be analyzed to estimate the number of direct-labor hours required. A statement of work is more common for the purchase of products, whereas the purchase of services might use the term statement of objectives. Regardless of the name, the concept is to develop direct-labor hour estimates based on the categories of labor required and the number of hours necessary for those labor categories. Sometimes product solicitations provide product specifications, which may require a contractor to design and build a product that meets a particular specification and to estimate the hours required to do so.

A solicitation might specify a level of effort for various labor categories; therefore, the number of direct-labor hours are a given. Another situation where estimating direct-labor hours is not important is for certain time-and-material or labor-hour contracts that may request only direct-labor rates (not hours). Direct-labor hours are provided in the solicitation; proposed rates are applied to these hours to estimate direct-labor dollars, to be used only for proposal evaluation purposes.

The government generally expects a detailed or grassroots buildup of cost estimates. This requires attention to detail to ensure that no cost is omitted from the estimate. One technique to assist in developing direct-labor hours involves the use of a work breakdown structure applied to the statement of work and is discussed in Chapter 1. This technique is also useful for cost elements other than direct labor and is described in Chapter 8.

Cost-estimating relationship techniques use a statistical relationship between historical cost data and other cost or noncost variables. Estimating techniques of cost-to-cost relationships have been used since many years before the development of CERs by Rand. Some examples of this are scrap costs to a priced bill of materials, non-touch direct-labor hours or dollars to touch direct-labor hours or dollars, sustaining engineering direct-labor hours or costs to initial engineering direct-labor hours or costs, and even other direct costs to direct-labor costs.

Previous production history is a good basis for developing an estimate of direct-labor hours but cannot be used blindly. Since the previous production, changes may have occurred in manufacturing processes, engineering, design, physical location of the work, or other significant attributes that would render the history unusable or limit its use. If such history is to be used, it should be reviewed for anomalies and to ensure that it corresponds to the conditions under which the product whose cost is being estimated will be manufactured.

A common problem contractors encounter is when a reviewer only considers historical data in evaluating a price proposal for work to be done in the future. Sometimes a reviewer is reluctant to complete a proposal evaluation unless historical data are made available. Some reviewers are not comfortable with contractor estimates and insist that estimates be based only on actual, historical cost information. The problem with this approach is that the results do not reflect the expected conditions during the future period where the work will be done but rather the historical data, which may or may not be pertinent. A reviewer may also look for the audit trail for a price proposal to support estimates with actual, historical cost data. This audit trail would be relevant only for work that has been completed. The term audit trail refers to a review or audit of costs that have been incurred and recorded in the books and records. When a proposal for work that has not been completed is being evaluated, the proper term is the basis of estimate, which is not found in the books and records but in documentation prepared to support a proposed price.

The basis of estimate may use historical data if they are relevant. For example, efficiency factors could be applied to historical data, or standard hours based on work measurement standards and time and motion studies could be adjusted for downtime due to personal fatigue. When labor hours are developed from a grassroots buildup, allowances must be made for legitimate downtime; employees require restroom breaks, walks to the coffee machine, and so on. These must be considered in estimating direct-labor hours. This factor, often termed personal time, fatigue, and delay, can be as great as 18 percent, depending on the work circumstances. The DCAA Contract Audit Manual (DCAM), Appendix I, Work Sampling, contains a discussion of this topic, but much of the meaningful discussion of appropriate percentages has been removed from the original guidance in a DCAA pamphlet. (The vital missing documents relate to support for the concept of personal time, fatigue, and delay.)

In a product environment, improvement curves are an excellent tool for estimating direct-labor hours. As workers repetitively perform a certain task, the time required for the task decreases. In addition, when an improvement curve can be applied, production is often disrupted by breaks in production. Techniques for improvement curve application, including considerations for breaks in production, are presented in Chapter 10.

Another estimating technique involves complexity factors applied to historical data. For example, a contractor may have information on the historical relationship between engine maintenance costs and horsepower for a specific aircraft engine. A complexity factor could be applied to the history to estimate the engine maintenance hours for the new engine. Complexity factors can be quite sophisticated and can be developed from engine horsepower, fuel consumption, and other technical factors and characteristics.

Engineering estimates are legitimate estimating techniques but are often rejected by government reviewers as being too judgmental. General criteria for valid engineering estimates include written notes on the thought process, including references to specific prior experience, rather than phrases such as “based on my 25 years in the industry.” However, combining similar historical data with engineering judgments (and perhaps with other historical data) is often acceptable to reviewers. For example, if a similar, previous item required 100 hours to build and the contractor’s engineers estimate that the new item to be priced is 20 percent more complex, an estimate of 100 + 100 × 20 percent = 120 hours might be acceptable. The 20 percent might even be supportable from historical data. In estimating maintenance costs for an aircraft, the complexity factor could be based on comparable performance or other data. Some examples of such data might be number of engines, maximum speed, aircraft weight, and so on.

Direct-labor hour estimates must be segregated into recurring and nonrecurring functions. Recurring functions may be more likely to have historical data for estimating purposes, whereas nonrecurring functions are often unique and no historical data exist. For future estimating, an accounting system should segregate recurring and nonrecurring labor hours as they are incurred.

Another technique for developing direct-labor hour estimates in a manufacturing environment is to analyze the product in terms of routing through various departments within the company. Each manufacturing step, such as grinding, painting, quality control, deburring, assembly, and inspection, is identified and direct-labor hours are estimated based on shop standards. The estimate of direct-labor hours must be detailed as a function of time (i.e., identified to the applicable cost-accounting period). This is necessary because direct-labor rates and indirect-cost rates must be applied by accounting period.

Some solicitations may request labor prices based on full-time equivalents (FTEs). This is a common technique for pricing grants, prices from nonprofit organizations, and prices from educational institutions. Because costs are recorded based on direct-labor hours, estimates based on FTEs are not in compliance with the Cost Accounting Standards (CAS). In addition, there is often no consensus on how many direct-labor hours equate to an FTE. In the situations where this technique is used, often the figure of 2080 hours (i.e., a standard work-year) is used, despite the facts that the direct-labor hours will be fewer than this amount and that these hours include paid time off. Using this approach, an employee could be on vacation for a month and his compensation for that period would be charged directly to a contract. The proper way to accommodate a solicitation that requires FTE estimates is to first estimate the direct-labor rate based on direct-labor hours, with paid time off accounted for as an indirect cost. Then the estimate can be converted to FTEs based on an hour amount in the solicitation or a good-faith estimate of how much direct labor will be realized from a 2080-hour year.

DIRECT-LABOR RATES

Estimated direct-labor rates are applied to estimated direct-labor hours based on individual employee labor rates, labor category rates, or departmental rates to produce direct-labor cost estimates. Whatever method is selected, it is important to apply the approach consistently. DCAA and other government agency reviewers do not permit cherry-picking by using individual rates for high-priced work and category rates for routine work. To develop fair and reasonable direct-labor cost estimates, a company must consistently apply its basis for estimating labor dollars.

It is possible to use a combination of these estimating approaches as long as the combination is applied fairly and consistently. For example, a contractor might price all engineering effort for the research department based on individual rates; at the same time, the company could estimate engineering for all other activities based on labor category rates. These are different circumstances that would allow both methods. However, within each designated grouping there can be no arbitrary selection of which method to use.

A caution on department rates and similar methods is that a raw, arithmetic average of the department rate may not produce equitable results. For example, if the department rates include interns and the department manager, an arithmetic average gives as much weight to the department manager as to an intern. On the other hand, the department manager may have a limited percentage of his time as a direct-labor charge. Thus, the mathematical average overstates the direct-labor rate for the work performed in the department.

Some solicitations require compliance with the Service Contract Act, Fair Labor Standards Act, or Walsh-Healey Public Contracts Act. When these provisions are required by the solicitation, a contractor must determine the rates for the covered categories and then also ensure that proposed rates meet these requirements. Similarly, if direct-labor rates are established by a labor–management agreement, those rates should be used to estimate costs and should be accepted by the government.

Solicitations and requests for proposals may dictate the labor rate categories to be used on a proposal. Generally, these categories will not be identical to the contractor’s rate categories. Therefore, a contractor will need to prepare a cross-walk to track the solicitation categories to the contractor’s categories. For example, the solicitation might use a category of Senior Engineer. The duties for this labor category as described in the solicitation must be analyzed and matched to the contractor’s labor categories. The solicitation duties for a Senior Engineer might match 80 percent to the contractor’s Engineer IV category and 20 percent to the contractor’s Engineer III category. The rate proposed for the solicitation category is then a composite of the contractor’s category rates and percentage of overall effort. This analysis must be performed for each position.

Local labor market conditions can affect labor rates. For example, a particular labor skill, such as welding, might be in great demand in a location that normally does not have a significant amount of local residents with this skill. Thus, a contractor might be required to pay a higher labor rate to attract people to the area or may incur additional costs to bring people with these skills into the local area to work on a contract.

Business volume occasionally impacts labor rates. This could happen when personnel in a particular category of labor are substantially increased. The hiring of interns or lower-paid personnel, due to lack of experience, could reduce the average labor rate for that category.

Employee turnover could impact labor rates. If more senior people leave or graduate into a higher labor category, the replacements at beginner level could lower direct-labor rates. On the other hand, downsizing could result in higher labor rates, because fewer senior people might be terminated.

Salary additives include cost-of-living differentials, hazardous duty pay, dislocation allowances, and other additives, particularly for conditions in an overseas environment. Many overseas factors are based on additives provided in State Department regulations for employees and contractors. Hourly additives for shift premiums, holiday pay, and overtime might also apply. These elements should not be overlooked in estimating labor rates.

Direct-labor rates should be escalated for future business. Before 1980, escalation factors for individual contractors were based primarily on the specific past experience of each contractor. Because of significant price inflation in the late 1970s, the Department of Defense (DoD) elected to constrain labor escalation factors in price proposals as a tool to limit escalation inflation. This limited contract prices as well. DoD used price indices from surveys by Data Resources Inc., now known as Global Insight. Over the years these price indices have ranged from 2.5 to 3.8 percent. This survey service is rather expensive and not available to most contractors on an economical basis. However, DCAA findings will often indicate what escalation factors DoD considers to be appropriate at any particular time.

The obvious inequity of using indices as opposed to actual, specific contractor experience lies in the fact that although a contractor that grants only modest pay increases will be allowed escalation in the price proposals, a contractor that must pay more liberal escalation in labor rates due to market conditions will not be able to price the entire amount of its costs in a contract proposal to the government.

Various rules and regulations address the labor rates that must be offered to existing employees or to employees hired from an incumbent contractor. These requirements must be considered in estimating direct-labor rates. Contract clauses may preclude compensation reductions for current employees. In many instances, incumbent contractor employees have the right of first refusal for positions with the successor contractor.

Uncompensated Overtime

The subject of uncompensated overtime is a controversial one but is unique to the government contracting environment. In a commercial business, cost accounting for uncompensated overtime is uncommon. In government contracting, whenever prices are based on costs or estimated costs (rather than on market prices), the government wants to be sure that the cost is right. That means they want full-absorption costing—they want the cost of compensation to the employee allocated equally over all the work that the employee performs. That work could be 40 hours per week or much more than 40 hours per week.

Many companies do not want to record all time (40 hours plus uncompensated time) because of the resulting accounting and pricing complications. The primary complications are that if all hours are recorded, the hourly rate used to apportion employee compensation to work (contract or project work and indirect or nonchargeable work) will vary each pay period. This makes estimating a nightmare, since it is not clear what rate to use.

Most of the companies that record all time do so because they perceive a competitive advantage in that some government personnel rely on the amount of the direct labor rate in determining reasonableness, rather than the cost (rate times hours). The problem of uncompensated overtime stems from the practices of Washington, DC–area “beltway bandits” that priced at lower labor rates than they actually paid their employees for a 40-hour workweek. A contractor might have had a $50-per-hour employee that it wanted to get onto a contract. The contractor would convince the government buyer that its people worked 60 hours a week and that the hourly rate thus was really only $33.34. Because so many government buyers looked only at the rate and did not consider the hours, the government would accept this “bargain.” This escalated as contractors promised more and more hours to the government. They also forced employees to accept what employees thought were excessive, unpaid overtime as normal working conditions.

If a contractor does not want to record all time, it must still satisfy the government that the paid hours are properly assigned to work in a manner that does not result in a bias against the government. For example, if employees were told to charge government cost-type contracts first, before commercial work and before government firm-fixed-price work, the government would have a legitimate right to object. The best way to accomplish the objective of avoiding total time reporting is to have a policy that does not encourage excessive amounts of uncompensated overtime. For example, overtime should be approved if significant and employees may work casual unpaid overtime without approval, but this should officially be discouraged as a long-term proposition.

The DCAA audit guidance in DCAM 6-410.3, Basic Audit Procedures, states:

a. Evaluate the contractor’s policies and procedures relative to work performed by exempt employees in excess of 8 hours per day or 40 hours per week. For service contracts to be awarded on the basis of the number of hours to be provided, FAR 52.237-10 requires an offeror to submit a copy of its policy addressing uncompensated overtime with its proposal. In addition, this FAR requires that an offeror’s accounting practices used to estimate uncompensated overtime be consistent with its cost accounting practices used to accumulate and report uncompensated overtime hours.

This requirement is the reason a written policy is needed. The latter guidance regarding consistency came about because contractors would use all hours in proposing a rate to win a contract and then would not record all hours after the contract was received. This inconsistency violates the CAS and the Federal Acquisition Regulation (FAR):

b. Determine whether the contractor is recording all hours worked by exempt employees. If a review of the employee time records discloses that exempt employees consistently record only 8 hours per day/40 hours per week, conduct floor checks and/or employee interviews to see whether exempt employees work in excess of 8 hours per day or 40 hours per week. If they do, discuss with contractor representatives the need to record all hours worked by exempt employees in order to ensure that salary and applicable indirect costs are being equitably allocated to all effort performed by the employees during the period. If the contractor refuses to record all hours worked by exempt employees, expand the floor checks and employee interviews to determine whether the failure of the contractor to record all time worked results in a material difference in the allocation of costs to final cost objectives. Obtain the assistance of the contracting officer in requiring the contractor to record all hours worked when a material difference in allocation of costs is determined.

The reviewer must establish that a material difference in cost allocations results from any failure to record all hours worked:

c. Determine whether the contractor is allocating salary costs paid to exempt employees to all effort performed in accordance with FAR 31.201-4 and CAS 418.

Note the DCAA regulatory basis for insisting on total time reporting:

d. If it is determined that government contracts are being over charged by a material amount due to an inequitable allocation of costs because the contractor does not record all time worked, the contractor should be cited as being in noncompliance with FAR 31.201-4 and CAS 418. Any material excess allocation of costs to government contracts should be questioned or disapproved as applicable. Materiality is the governing factor when determining whether noncompliances should be cited and whether a contractor should be required to implement a total-hour accounting system.

If a company accumulates the uncompensated hours, it will provide the data for the DCAA analysis.

Other Considerations

Reasonableness of direct-labor rates may become an issue with a government reviewer. For this reason, rates should be based on wage surveys to establish reasonableness. For most contractors, their human resources department performs this task. However, if a contractor does not use surveys to establish employee pay, a government reviewer might apply published surveys on the basis that the contractor has not established the direct-labor rates as reasonable. DCAA reviewers do not accept data from surveys that are based on voluntary participation. Thus, many of the free online surveys or databases are not acceptable to the DCAA.

Overtime premium can be accounted for as either direct labor—as an other direct cost—or as an indirect cost. Recording as direct labor limits visibility of the cost for estimating and analytical purposes and is not favored. Treatment as an other direct cost is acceptable accounting but creates negotiation difficulties, because individual customers may not want to include overtime premium in their prices. The most equitable treatment is to account for overtime premium as an indirect (overhead) cost. The cost should not be included in the fringe benefit cost pool, because it is payment for services rendered (just like a salary or bonus) due to workload and not incurred for the benefit of an employee. In addition, the premium portion of overtime is not allowable under FAR cost-type contracts unless there is a contractual agreement that allows such premium.

Pre-established rates, such as those established by a forward pricing rate agreement (FPRA), may be used for pricing. The principles of estimating and negotiating FPRA labor rates are the same as those presented in this section for estimating for individual proposals.

Estimating for standard cost systems (as opposed to actual cost systems) presents difficulties for a contractor. The total costs will be recorded as a standard and a variance from standard. Thus, to comply with the consistency requirements of CAS 401, Consistency in Estimating, Accumulating, and Reporting Costs, the estimate must be presented in the same manner. The standard estimate is not so difficult to treat in this way; however, the variance estimate can be an issue. Standards are set by different methods—frozen standards (standards that seldom change), ideal standards (assumes no inefficiencies), or realistic standards (assumes normal inefficiencies). In addition, a contractor may employ either engineered or estimated standards. Depending on the type of standard used, the variance might be significant, zero, or relatively insignificant. Convincing a customer to accept a large negative variance in a proposal might be problematic.

Some contractors purchase labor in significant amounts and some have no employees—all personnel are subcontracted from a company that specializes in providing personnel. Minor amounts of such labor are often recorded as other direct costs. However, significant amounts of contracted labor may require that these costs be recorded as labor. Costs must be recorded and estimated consistently. The estimated rates from companies providing subcontract employees should be based on current quotes. Accounting for purchased labor as though it consisted of subcontractor employees is not uncommon and may even be required by DCAA.

A solicitation may require blended direct-labor rates for both the prime contractor and any subcontractors. These costs are not all recorded as labor but as (in-house) labor and subcontracts. The proper estimating technique is to first estimate a direct-labor rate for the prime contractor and any subcontractors; then, to estimate the relative percentages of the labor category that will be used on contract performance. The proposed rate is then a composite of the individual rates and their percentage of the work to be performed.

BASIS OF ESTIMATE

Development of direct-labor hour and rate estimates should be documented as these amounts are developed. A form for direct-labor estimate documentation is provided in Illustration 2-1.

ILLUSTRATION 2-1: Basis of Estimate for Direct Labor