The study investigates whether competed service sector contracts with a single offer or sole source awards indicate a lack of qualified firms or significant barriers to entry.
The study concludes that:
- the prevalence of service sector contracts receiving only a single offer is about half as large as the data appear to suggest; and
- the use of short-term contracts and modifications to fill the gap in services between the end of one contract and the beginning of the next is a significant source of sole source contracts.
This study finds that the prevalence of service sector contracts receiving only a single offer is smaller than the data appear to suggest and that the use of bridge contracts is a significant source of sole source contracts.
The presumption established in the Federal Acquisition Regulation (FAR) is that federal contracts should be awarded on a competitive basis whenever possible and that competed contracts should be available to multiple offerors. This presumption applies to all Department of Defense (DoD) contracts for services. DoD supports vigorous competition for federal government contracts.
In a focus on the services sector, the data on competition raises some questions. In FY 2008, DoD committed approximately $200 billion1 in contracts for services. Over $28 billion of this total consisted of competed contracts that attracted only a single offer. Moreover, nearly $26 billion in DoD service contracts were awarded sole source. Together, these two categories accounted for $54 billion in FY 2008, or over 25 percent of the total volume of DoD spending on service contracts in that year.
In 2009, the DoD Office of Industrial Policy (IP) asked the Institute for Defense Analyses (IDA) to examine DoD contracts for services that are: (1) competed but that receive only a single offer, and (2) sole source. We were asked to determine whether the prevalence of single offer and sole source contracts for DoD services represents an industrial base concern, such as a lack of qualified firms or significant barriers to entry.
This paper focuses on two findings of the IDA study that highlight the necessity of carefully reviewing top level statistics, such as those above, before drawing conclusions regarding the level of competition in DoD services contracts. During the study, we found that: (1) the prevalence of competed DoD services contracts receiving a single offer is only about half as large as the data cited above appear to suggest, and (2) a significant number of sole source DoD services contracts are bridge contracts—short-term contracts that fill the gap in services between the end of one competed contract and the beginning of the next.
Data & Methodology
Our two primary data sources were the FY 2008 Federal Procurement Data System-Next Generation (FPDS-NG) database, supplied by the sponsor, and a database containing Justification and Approvals (J&As) for not-competed contracts, which we created from publicly available data on the FedBizOpps website.
FPDS-NG FY 2008 Data Set
The FPDS-NG database records all contract actions greater than $3,000. Initial contract actions receive a contract number. Contract actions that alter an existing contract are assigned a modification number while retaining the original contract number. Contract award types include purchase orders, task orders, and definitive contracts.
- Purchase orders are used for items purchased using simplified acquisition procedures and are typically for less than $100,000 (FAR, Subpart 13.3, 2010).2
- Task orders are orders written against Single or Multiple Award Indefinite Delivery Vehicles (IDVs) that act as large master contracts. In the database, these task orders automatically inherit the terms of the master contract. Each master contract can have multiple task orders written against it.
- Definitive contract awards are standard individual contracts.
The FPDS-NG database records information about each contract action. Of particular interest to this study were the contract value fields (Dollars Obligated, Current Contract Value, and Base and All Options Value), the Type of Contract field, the competition information (Extent Competed, Reason Not Competed, and Number of Offers Received), as well as the Statutory Exceptions to Fair Opportunity in the case of multiple-award IDVs.
The FPDS-NG database is a live database, meaning that updates and changes to past entries can be made at any time. In fact, changes to contract actions in any fiscal year can be made well past the fiscal year end. To ensure consistent results for queries regarding these contracts, we were provided a version of the FY 2008 FPDS-NG data containing DoD contract actions frozen as of 15 January 2009.3
We used the Extent Competed field of the FPDS-NG database to categorize contract actions as competed if the Extent Competed field was given as:
- Full and Open Competition
- Full and Open Competition after exclusion of sources
- Competitive Delivery Order
- Competed under Simplified Acquisition Threshold
We classified a contract as sole source if the reason not competed was given as:
- Unique Source
- Follow-on Contract
- Unsolicited Research Proposal
- Patent/Data Rights
- Brand Name
- Only One Source
Justifications and Approvals Database
For our research into significant causes of sole source contracts, we collected the J&As posted on the FedBizOpps website between 1 March 2009 and 30 September 2009. The J&As state which FAR exception was applied and provide supporting documents explaining the use of the exception. For example, for some sole source contracts, the FPDS-NG database would specify Only One Source as the FAR exception. The additional documentation might state that a short-term contract to an incumbent was awarded because the replacement contract had not yet been competed. We used this additional detail regarding the FAR exceptions to indicate whether or not the not-competed contracts were bridge contracts.
We conducted interviews with contracting office personnel and contractors to gain insight into the factors driving the number of bids on government contracts. In addition, we conducted a literature review for a theoretical model governing the number of bids or the use of sole source contracts (Williams et al., 2010).5
We selected contracting offices to interview based on the following criteria: (1) the office was one of the top locations in terms of dollars or contract actions in a given industry segment, or (2) it had a high percentage of single bids or sole source contracts compared with other offices working in the same industry segment. We selected at least one contracting office from each of the industry segments that had been identified by IP.
In addition to the contracting offices, we interviewed contractors from three trade organizations: the Tidewater Government Industry Council, TechAmerica, and the Coalition for Government Procurement. The Tidewater group primarily comprises small businesses, while TechAmerica and the Coalition for Government Procurement primarily comprise large firms.
Our discussions with contracting officers and contractors led us to look at the data more closely. Specifically, they helped us to understand anomalies in the competed contract data (discussed in the next section) and suggested bridge contracts as a cause of sole source contracts.
We were asked to investigate causes that underlie or go beyond the FAR exceptions. There was a consensus among the contracting offices that bridge contracts—written when the competition or the award for the new contract has been delayed—are a significant source of sole source contracts. Delays can come from various sources for various reasons: the requiring agency may have a holdup in preparing the requirements documents or experience a change in requirements; the contracting office may be unable to use an existing contract vehicle or an issue may arise during any of the several review boards that constitute the process; or other sources, such as protests, may be responsible.
Single Offer Results
Scope of the Single Offer Problem
Using the 15 January 2009 snapshot of the FPDS-NG database for FY 2008 contract actions, we find that of the $155.3 billion in competed contracts, $28.4 billion, over 18 percent, appears to have been awarded after receiving a single offer (see Figure 1). Unless otherwise noted, by “competed contract dollars,” we mean the total dollars obligated on the original contract, plus the contract modifications. This is the standard method of reporting contract dollars used by IP.
IP classifies services into various industry segments. Figure 2 shows the relative shares of competed contract dollars by industry segment. Construction-related (CR) contracts receive the largest share of competed contract dollars, with Management Support, Professional and Administrative (MSPA) and Research, Development, Test & Evaluation (RDT&E) receiving significant shares as well.
Figure 3 shows the relative shares of single offers on competed contracts by industry segment. For single offers, MSPA and RDT&E take over the largest shares, with Equipment-related (ER) as the third-largest segment. CR falls to the sixth-largest share of single-offer dollars.
Single Offers in RDT&E
From Table 1, we see that RDT&E accounts for close to 26 percent of FY 2008 competed dollars awarded after receiving a single offer. A closer look at the data shows that RDT&E is an obvious outlier with respect to single offers on definitive contracts, accounting for over 50 percent of all single offers. However, we found upon closer examination that definitive contracts in RDT&E are frequently Broad Agency Announcements (BAA) and Small Business Innovation Research (SBIR) contracts. These contracts often appear as if they were competed contracts receiving a single offer in the FPDS-NG database, regardless of the number of offers received.
BAAs differ from a standard request for proposal in that they: (1) are focused on advancing the state of the art or increasing knowledge and understanding, rather than on a specific system or solution; and (2) are defined by a statement of the problem rather than a statement of work.
Each proposal presents a unique solution to a problem and is evaluated on its own absolute merits rather than its merits relative to the other proposals received. A single BAA could receive 100 proposals and make two awards or could receive 20 proposals and make 20 awards. BAAs may remain open for up to a year, and firms may make proposals and receive awards throughout that time (DARPA, 2009). SBIRs are similar to BAAs but have additional requirements: the firm must be a for-profit small business with 500 or fewer employees, the work must be performed in the United States, and the principal investigator must be employed at least half-time by the proposing firm (Department of Defense Office of Small Business Programs, 2012). In addition, SBIR requests for proposals usually have a fixed ending date, after which all proposals are evaluated.
Both BAAs and SBIRs are considered competitive solicitation procedures, and the FPDS-NG Data Dictionary (Version 1.3, 102) directs contracting offices to code these procedures as full and open competition. Almost all the contracting offices we spoke to said that they enter the number of offers for successful BAA proposals as “1,” regardless of the number of proposals actually received.6 The results for SBIRs were more mixed; some offices said they enter the total number of proposals received, and others said they enter a “1,” for all research proposals. As a result, most BAAs and many SBIRs appear in the FPDS-NG database to have received a single offer, whether or not multiple proposals were submitted.
Because BAAs and SBIRs represent a fundamentally different type of competition from the standard definitive contracts, we believe that they should not be counted as competed actions receiving a single offer. Unfortunately, there is currently no method for separating BAA records in FPDS-NG from other definitive contract actions in RDT&E. For SBIRs, some of the contract descriptions state SBIR Phase I or SBIR Phase II, but most do not. Instead, the contract documents must be examined individually.
To determine the portion of the apparent single offer RDT&E contracts that are actually BAAs or SBIRs, we asked nine contracting offices to identify whether their single-offer contracts in FY 2008 were actually BAAs or SBIRs.7 For this effort, we did not perform a scientific sampling; we focused instead on contracting offices with the highest volume and value for RDT&E definitive contracts.
The results were surprising. Based on the responses of the contracting offices surveyed, BAAs and SBIRs accounted for 69 percent of the single-offer RDT&E8 definitive contracts issued in FY 2008 and 62 percent of the contract value attributed to single-offer definitive contracts. Table 2 shows the detailed breakout of BAAs and SBIRs for each of the contracting offices that categorized their contracts.
Our sample represented 11.5 percent of the total value of single-offer definitive contracts in RDT&E. This is a large enough sample that we are comfortable assuming that the proportion of remaining RDT&E single-offer definitive contracts dollars that are actually BAAs or SBIRS is also close to 62 percent (totaling about $3 billion). Based on this estimate for the proportion of BAAs and SBIRs, RDT&E accounts for $4.3 billion (17 percent) of all single-offer contract dollars rather than $7.3 billion (26 percent). Removing BAAs and SBIRs from the total pool of single-offer contracts reduces the total dollar value of single-offer contracts to $25.4 billion, as shown in Figure 4.
Our sample from the contracting offices allows us to estimate the proportion and dollar value of BAAs and SBIRs; however, it does not allow us to identify such contracts in the FPDS-NG data. As a result, we excluded RDT&E definitive contracts from our statistical analyses below. Additional research into this area would be facilitated by the ability to reliably identify BAAs and SBIRs in the FPDS-NG database.
Award Type and Single Offers
Although not raised as a factor in either the literature review or the interviews, award type turns out to be a predominant factor affecting the total value of competed contracts receiving a single offer. As discussed above, the DoD uses three primary contracting vehicles (purchase orders, task orders, and definitive contracts) to contract for services. Each type of award places different requirements on the contracting officers with regard to competition and the parameters of the acquisition:
- Purchase orders are used for items purchased using simplified acquisition procedures and are typically for less than $100,000.9
- Task orders are orders written against established Multiple Award or Single Award IDVs. IDVs allow the government to purchase goods or services on an as-needed basis through the issuance of task orders.
- Definitive contract awards are standard individual contracts.
An IDV contract does not specify the actual quantity or specific service required. Instead, it provides a quantity range or general description of required services. Actual quantities and detailed service requirements are specified in the subsequent task orders. The IDV establishes qualified sources of the goods or services. In a Single Award IDV, all the services will be purchased from a single contractor, with the orders placed as needed. A Multiple Award IDV establishes a pool of qualified contractors. When the government issues a task order for services, those qualified contractors compete to provide the service.
As shown in Figure 5, purchase order contracts account for only 1 percent of the total competed contract dollars. Single Award IDV task orders represent 30 percent, Multiple Award IDV task orders are 23 percent, and definitive contracts account for 46 percent.
Categorizing IDV Task Orders as Single versus Multiple Offer
It is important to distinguish between Single Award and Multiple Award IDV task orders. In a Multiple Award IDV, the master IDV contract competition had multiple offers (by definition), and multiple firms were selected. These firms are exclusively allowed to compete on the task orders that are issued over the course of the contract. If only one of the selected firms bids on a task order, the value of that task order is counted as a single offer contract.
For Single Award IDVs, the master IDV contract competition may have had one or many offers. One firm is selected to perform the work specified in the task orders that are issued over the course of the contract. In the case that the master IDV contract competition received only a single offer, the value of all the task orders on the contract are counted as single offer.
Determining whether a particular task order belongs to a Single or Multiple Award IDV is complicated by the fact that IDV contracts tend to run over several years. The FPDS-NG database snapshot used in this study contains many IDV task orders whose master IDV competitions occurred in previous years. Previous-year master IDV contracts are not included in the data we used (Appendix A of Williams et al., 2010).
The value of single-offer contract dollars shown in Figure 6 indicates that IDV task orders make up a larger than expected proportion, based on their prevalence in the total competed dollars pool. Multiple Award IDV task orders account for 23 percent of all competed contracts, but 43 percent of single offers. By contrast, definitive contracts account for 46 percent of competed contracts, but, after excluding BAAs and SBIRs, just 23 percent of single offers. Single Award IDV task orders fall in between—accounting for 32 percent of single offers and 30 percent of total competed dollars. Purchase orders account for only 2 percent of competed contract dollars receiving a single offer and will not be analyzed further in this study.
Effect of Multiple Award IDV Task Orders on Single Offers
We see that Multiple Award IDV task orders are disproportionately responsible for single offers. We believe, however, that Multiple Award IDV task orders that receive a single offer do receive a level of competition. The competition for the master IDV contract received multiple offers, and multiple qualified contractors were selected to supply the services specified in the task orders. The second stage provides fair opportunity for all multiple award contract holders to compete for the task order unless an exception to fair opportunity is approved. Individual firms know that their competitors are qualified, but do not know if they will bid on the task orders. The threat of other bids prevents an individual firm from bidding as a monopolist. Thus, to the extent that firms believe there will be multiple bidders on the task orders, the benefits of competition are conferred from the IDV competition to the task order competition even if only one offer is received.
Single Offer Conclusions
In FY 2008, RDT&E and services together represented more than 50 percent of DoD contracts. Of the $155.3 billion in competed DoD service contracts, $28.4 billion, or 18 percent, appeared to be awarded after receiving a single offer; however, this appearance is misleading. Approximately $3 billion in RDT&E single-offer contracts actually came from BAAs and SBIRs, which appear as single offers in the FPDS-NG database regardless of the number of proposals received.
An additional $10.9 billion was due to single-offer Multiple Award IDV task orders. The pool of firms eligible to bid on the task orders was selected based on an IDV competition with multiple offers. Although it is difficult to quantify the benefit, to the extent that firms believe there will be bids from the other qualified firms on the Multiple Award, the benefits of the competition for the master IDV contract confer to the task orders. We therefore characterize the single-offer Multiple Award IDV task orders as having received some competition.
The final $14.5 billion, 9 percent of competed DoD service contract dollars, in single-offer definitive contracts ($5.9 billion) and Single Award IDV task orders ($8 billion) was awarded without multiple offers at any stage.10 Figure 7 shows the revised breakout of competed contract dollars based on these findings.
As a final point, we argue that receiving a single offer does not necessarily mean that the competitive process was ineffective. Firms have limited resources with which to prepare bids and proposals, a time-consuming and costly process. Firms are selective, choosing proposals for which they believe they have a competitive advantage. These subjective expectations are, in part, a result of the firms’ beliefs about which other firms will bid the project. For this reason, number of offers may not be a sufficient metric for gauging the level of competition for these contracts.
Sole Source Contract Results
Although a clear preference is found in law and regulation for full and open competition in federal government contracting, in accordance with the FAR, contracts may be awarded without competition under certain circumstances. These exceptions to full and open competition include the existence of only one source, unusual or compelling urgency, international agreements, and public interest, among others. In FY 2008, of the $201.9 billion in DoD contracts for services, $46.5 billion was awarded without competition. Contracts awarded under circumstances other than full and open competition are further categorized into “Sole Source” and “Other Not Competed.” Table 3 shows the FAR exceptions that are considered sole source and those considered other not competed.
The purpose of this study is to determine whether there are underlying causes, beyond the stated FAR exceptions, driving the number of sole source contracts. We focus on sole source contracts, but discuss our results in the context of all not-competed contracts when appropriate.
Scope of Sole Source Contracts
Not-competed contracts account for $46.5 billion (23 percent) of the DoD service sector contracts awarded in FY 2008. Sole source contracts account for $25.9 billion (56 percent) of all not-competed contract dollars.
Sole source contracts can be used for all award types, including task orders issued under Multiple Award IDVs that utilize an exception to fair opportunity. Figure 8 shows the relative percentage of sole source contract dollars by award type. Definitive contracts account for the majority, followed by Single Award IDV task orders. As was explained above, we categorize a Single Award IDV task order as sole source if the original Single Award IDV contract was awarded sole source. For Multiple Award IDVs, the IDV competition selects a set of contractors who compete for the task orders. All contractors selected for a Multiple Award IDV are expected to receive fair opportunity to compete for the task orders. However, just as FAR exceptions to competition are allowed in standard contracts, exceptions to fair opportunity are allowed for Multiple Award IDV task orders. Multiple Award IDV task orders awarded without allowing fair opportunity are categorized as not-competed contracts and included in the total value of not-competed contracts. Those task orders awarded under exceptions to fair opportunity that are not considered sole source are not included in our sole source analysis.
As noted earlier, there was a consensus among contracting officers that the use of bridge contracts was a major contributor. A bridge contract, which may be a new contract or an extension to an existing contract, is typically short term, usually less than six months, but may be written for a year or more.
Bridge contracts are used when a delay in the acquisition process causes the existing contract to expire before the new contract is ready to be put into place. Delays in the acquisition process may come at any point in the process due to such events as:
- the requiring agency changing the requirements or failing to submit requirements in time.
- the contracting office discovering that the planned contract vehicle cannot be used or a problem occurring at any of the several review and approval boards that constitute the process.
- protests of the contract award or other obstacles.
While we were unable to quantify the effect of the possible causes of acquisition delays, we did examine the scope of bridge contracts in DoD services sector contracts.
Bridge Contracts in FPDS-NG
Since we were unable to determine a way to definitively distinguish bridge contracts in the FPDS-NG data, we attempted to use the information about the typical length of bridge contracts as a proxy for identifying them in the database.
Many task orders and definitive contracts have initial terms of one year plus options, so we examined contracts with terms strictly less than twelve months and those with terms of six months or less. Figure 9 shows that 31 percent of new sole source definitive contracts are written for six months or less and 58 percent are written for less than one year. Task orders are even more likely to be short-term: 52 percent are written for six months or less and 73 percent for less than one year.11
As we learned, due to the fact that the majority of all sole source task orders and definitive contracts are written for less than one year, there is no way to determine whether these contracts are bridge contracts, short-term projects, or an artifact of the way contracts are written using a base year plus options format. In addition, this analysis excludes modifications, which are often used for bridge contracts as well. We therefore turned to another source to estimate the effect of bridge contracts.
Justification and Approval Database
A J&A is a document required to justify and obtain appropriate level approvals for contracts awarded without the benefit of full and open competition as required by the FAR. In early 2009, a requirement was implemented for federal agencies to post their J&As, with supporting documents, to the FedBizOpps website within fourteen days of their approval.
We collected all J&As posted on the FedBizOpps website between 1 March 2009 and 30 September 2009. Using the text of the J&A and attached documents, we attempted to determine whether each of the J&As was for a bridge contract. Although “bridge contract” is not an official term—and we could not find an official definition—J&As frequently (but not always) state that the sole source contract is a bridge contract. Determining which contracts are bridge contracts requires some interpretation. For the purposes of this study, we classified a contract as a bridge contract if:
it was a not-competed contract or a sole source extension to an existing contract due to a delay, and the award process or competition was planned or had already been held, or the J&A stated that the contract or extension was a bridge contract.
Results of J&A Analysis
Over the seven months from March through September 2009, we obtained 958 J&As from the FedBizOpps website. Of these, 777 were for DoD contracts and 217 were for DoD services. The J&As showed that bridge contracts occurred in all federal agencies and in both sole source and other not-competed categories. Figure 10 shows the breakdown of the data.
Figure 11 shows the percentage of sole source J&As that we classified as bridge contracts. For sole source J&As, bridge contracts were a higher percentage in the DoD services sector than in DoD as a whole, but not as high as in other federal agencies. Over 23 percent of the DoD services sector sole source J&As were identified as bridge contracts, compared with less than 2 percent of non-services DoD J&As.
To determine the value of the DoD services sector bridge contracts identified through the J&As, we called on the Defense Manpower Data Center (DMDC) to pull the contracts from its database and provide the contract values. The DMDC was able to match 135 (62 percent) of the 217 DoD services sector contracts, of which 25 were bridge contracts.
Of the DoD services sector J&As for which we obtained a contract value, bridge contracts represented over 10 percent of the sole source DoD services sector contract values. Due to the amount of data missing from both the bridge and non-bridge J&As, we hesitate to extrapolate this finding to estimate the total value of bridge contracts in DoD services. Due to the short-term nature of bridge contracts, however, the total contract value likely underestimates the cost of their use.
Sole Source Conclusions
In FY 2008, of the $200 billion in DoD contracts for services, not-competed contracts accounted for $46.5 billion (23 percent). Sole source contracts, a subset of not-competed contracts, accounted for $25.9 billion (56 percent) of all not-competed contract dollars. The purpose of this study was to determine whether there were underlying causes, beyond the stated FAR exceptions, driving the number of sole source contracts.
The one underlying cause that we uncovered was the use of short-term contracts to fill the gap in services between the end of one contract and the beginning of the next. These bridge contracts, as they are called, are due to delays in the acquisition process. Delays can be caused by many factors at either the requiring agency, the contracting office, or from external sources.
To analyze this issue, we collected J&A documents from the FedBizOpps website from March through September 2009. We found that during this period, nearly one in four J&As for not-competed DoD services posted was a bridge contract.
The value of these short-term contracts appears to be small, about 10 percent of the total sole source J&As whose contract values we were able to obtain. However, the use of bridge contracts represents a potentially large cost to the DoD due to process inefficiencies. This cost must include the costs of preparing and administering the bridge contracts at the requiring agency, the contracting office, and the contractor. The use of bridge contracts also places a strain on the limited DoD contracting workforce because the contracts must be justified and awarded in addition to the follow-on competitions for the required services that must be performed. Finally, the DoD does not receive the benefits of competition during the period that the bridge contracts are in force.
The ability to identify bridge contracts in the FPDS-NG database would facilitate further research into this area, allowing the DoD to more definitively quantify the use of these contracts in the services sector and possibly to evaluate the cost to the Department of their use.
Some Policy Implications
The question that prompted the original study was whether single-offer and sole source contracts represent a lack of competition for DoD services. What we found was that the top level data may overstate the lack of competition for DoD services. Of the $28 billion in apparent single-offer contracts, half received some level of competition. Significantly, the BAAs, SBIRs, and Multiple Award task orders are not readily identifiable from the FPDS-NG data.
This suggests, first, that the problem is much smaller than the top level statistics indicate and, second, that policies based on the assumed $28 billion in single offer contracts may have little impact if the true number of single offer contracts is much smaller. Policies will need to be tailored to specific award types. Policies to reduce single offers on Multiple Award task orders may do little to increase offers on definitive contracts and vice versa.
Our sole source results are similar in that a significant cause, bridge contracts, cannot be identified from the FPDS-NG data. Sole source contracts account for $25.9 billion (56 percent) of all not-competed contract dollars for DoD service sector contracts awarded in FY 2008. In the data we collected from FedBizOpps, nearly one in four sole source DoD service contracts was a bridge contract. Bridge contracts put in place due to acquisition process delays caused by requirements documents not having been prepared in a timely manner, issues that arise during the contract review and approval process, or award protests, for example, cause services that were meant to be competed to be awarded sole source, at least in the short term.
The contract value for these contracts appears to be small, about 10 percent of the total sole source J&As whose contract values we were able to obtain. However, the cost to the DoD of their use is much larger. The cost includes the costs of preparing and administering the bridge contracts at the requiring agency, the contracting office, and the contractor. The use of bridge contracts also places a strain on the limited DoD contracting workforce because the contracts must be justified and awarded in addition to performing the follow-on competitions for the required services. Addressing process inefficiencies that cause competitive contract award delays could reduce the number of bridge contracts and save the DoD the cost of administering them.
DoD cannot address the costs of bridge contracts if it cannot identify and quantify them. In addition, general policies aimed at reducing the use of sole source are likely to have little effect in reducing bridge contracts unless those policies also address the causes of delays.
Bajari, P. & Tadelis, S. (2001). Incentives versus transaction costs: A theory of procurement contracts. The RAND Journal of Economics, 32(3), 387–407.
Berteau, D., Ben-Ari, G., & Sanders, G. (2009, February). Structure and dynamics of the U.S. federal professional services industrial base. Center for Strategic & International Studies, Defense-Industrial Initiatives Group. Washington, DC: Center for Strategic & International Studies.
Broad Agency Announcements (BAAs): CMO Contracts Management Office. (n.d.). Retrieved from http://www.darpa.mil/cmo/baa.html
Copeland, C. W. (2008, September 30). The federal workforce: characteristics and trends. Washington, DC: Congressional Research Service.
Defense Advanced Research Projects Agency (DARPA) Contract Management Office. (2009). Retrieved from http://www.darpa.mil/cmo/baa.html
Department of Defense Office of Small Business Programs. (2012). Retrieved from http://www.acq.osd.mil/osbp/sbir/sb/eligibility.shtml
Denes, T. A. (1997). Do small business set-asides increase the cost of government contracting? Public Administration Review, 57(5), 441–444.
Federal Acquisition Regulation, Subpart 13.3. (2010). Retrieved from https://www.acquisition.gov/far/html/Subpart%2013_3.html
Global Computer Enterprises. (2008, September 30). GSA federal procurement data system-next generation (FPDS-NG) data element dictionary (Version 1.3).
Government Accountability Office. (2004, July). Contract management: Guidance needed to promote competition for defense task orders (GAO-04-874).
Government Accountability Office. (2003, August). Contract management: Civilian agency compliance with revised task and delivery order regulations (GAO-03-983).
Government Accountability Office. (2009, November). Defense acquisitions: Further actions needed to address weaknesses in DoD’s management of professional and management support contracts (GAO-10-39).
FedBizOpps.Gov Federal Business Opportunities. (2009, September 15). Retrieved from https://www.fbo.gov
Inspector General, United States Department of Defense. (2009, May 6). SeaPort enhanced program (No. D-2009-082).
Krasnotkutskaya, E. & Seim, K. (2008, November). Bid preference programs and participation in highway procurement auctions.
Lamm, D. V. (1987). Why firms refuse DoD business: An analysis of rationale. National Contract Management Journal , 21 (Winter), 45–55.
MacManus, S. A. (1991, Jul–Aug). Why businesses are reluctant to sell to governments. Public Administration Review , 51(4), 228–244.
Manuel, M. S. (2009, February 2). GAO bid protests: Trends, analysis, and options for Congress, 7-5700. Washington, DC: Congressional Research Service.
Office of the Deputy Under Secretary of Defense, Industrial Policy. (2003, February). Transforming the defense industrial base: A roadmap. Washington, DC.
Office of the Deputy Under Secretary of Defense, Industrial Policy. (2008, July). Competitive services industry, services body of knowledge – FY08 updates.
Office of the Inspector General, Department of Defense. (1999, April 2). DoD use of multiple award task order contracts (Report No. 99-116).
Office of the Inspector General, Department of Defense. (2001, September 30). Multiple award contracts for services (Report No. D-2001-189).
Office of the Inspector General, National Aeronautics and Space Administration. (2001, September 21). Multiple-award contracts (IG-01-040).
Randall, S. O. (1997). An analysis of reasons commercial entities prefer not to participate in defense business. Washington, DC: Storming Media.
Stott, B. & Zlomislic, O. (2004). Single-bid awards under the GSA service schedules. (Thesis). Cambridge, MA: Harvard University.
U.S. Small Business Administration. (n.d.). (2009, December 1). SBA Programs 8(a) Business Development Programs FAQs. Retrieved from http://www.sba.gov/aboutsba/sbaprograms/8abd/faqs/index.html
Williams, Laura M., Rose, Susan L., and Rehwinkel, Andrew S. (2010, March). Competitiveness in the services sector (IDA Paper P-4542). Alexandria, VA: Institute for Defense Analyses.
Wong, M. M. (2006, September). Current problems with multiple award indefinite delivery/indefinite quantity contracts: A primer. The Army Lawyer, 17–30.
1A $13.9 billion data error was discovered after the FY 2008 data set was frozen. Although the error results in an overstatement of services contracts and competed contracts with multiple offers, it does not qualitatively or substantively change our conclusions.
2However, the limit in FY 2008 was typically $5 million for purchases of commercially available supplies and services, but can be up to $11 million in certain national security areas. The limit for commercially available supplies and services has increased to $5.5 million.
3A $13.9B data error was discovered after the FY 2008 data set was frozen. Although the error results in an overstatement of services contracts and competed contracts with multiple offers, it does not qualitatively or substantively change our conclusions.
4Standardization programs seek to standardize equipment and services to reduce training and maintenance costs.
5The literature review and interviews suggested the following areas for exploration: set-asides, contract size and structure, geographical constraints, specialized requirements, time allotted for bid preparation, and the presence of an incumbent.
6One office said they entered “999” as the number of offers received for BAAs, but the rest said “1.”
7We provided the offices with lists of the initial contracts. We excluded modifications to reduce the size of the lists.
8This is the total value of the contracts we provided, including modifications.
9A $13.9B data error was discovered after the FY 2008 data set was frozen. Although the error results in an overstatement of services contracts and competed contracts with multiple offers, it does not qualitatively or substantively change our conclusions.
10The total does not add to $14.5 billion due to single offer purchase order contracts, which are not analyzed here.
11Purchase order contracts, which we have excluded from this analysis, are typically used for small purchases, which are also likely to be short-term.
Susan Rose is a research staff member at the Institute for Defense Analyses (IDA). During her five years at IDA, Dr. Rose has worked on a variety of competition-related tasks for OSD. She has a Ph.D. in Economics from The Ohio State University.
At the time this article was published, Laura M. Wiliams was a project leader at IDA.
Andrew S. Rehwinkle is a senior consultant at FTI Consulting. At the time of publication, he was a research associate at IDA. Mr. Rehwinkle holds a B.A. in Psychology and History from Georgetown University, an M.A. in Applied Economics from The John Hopkins University, and an MBA in Finance and Strategy/Entrepreneurship from Southern Methodist University.