During the Covid pandemic, there was considerable concern that corners were being unduly cut in a bid to get vital protective equipment to the right places at the right time.  Accusations were labeled that government ministers had “VIP lists” of preferred suppliers, many of whom were friends of either them as individuals or of their party.

Research from Carnegie Mellon finds a similarly dysfunctional picture in the United States, with minimal competition from contracts.  The researchers use a principal-agent model in which the government wants contractors at a particular cost and negotiates with them to try and achieve that.

The study suggests that there is actually of limited benefit for the government agency from attracting more bidders because the agency can already extract informational rent from each bidder.

Minimal competition

The researchers began their study after finding that around 44% of US government procurement spending went on contracts where there was a single bidder in 2015.

To better study why this is, they focused on two particular features of federal procurement.  The first of these is the fact that the buying agency has broad discretion as to whether a project needs to draw competitive bids.  Indeed, the final contract price can often be much higher than the initially agreed price.  The authors believe these factors are important to assess in unison because they directly affect the initial terms and the final contract price.

The researchers modeled the procurement process as a two-stage, noncooperative game whereby the government first selects the level of competition they want among bidders before then negotiating the specific terms of the contract.

Data was gleaned from the Federal Procurement Data System for all IT contracts between 2004 and 2015.  The researchers focused on contracts with fixed schedules and quantities, which narrowed things down to around 7,000 contracts with a total cost of around $2.5 billion.

A select few

The analysis found that in just over two-thirds of contracts there was neither full nor open competition used.  Sometimes this was due to the discretion of the agency, such as the need for speed, while other times it was more a statutory requirement.

The researchers believe that regardless of the number of bidders for any contract, the procurement agency will be able to extract more rent from the winner through negotiating than through open competition.

For instance, if agencies were not permitted to design and negotiate each contract then it would probably mean more bids would emerge, but those winning contractors would scarcely see any increase in their payment.  By contrast, if agencies are allowed to use their expertise of the supply side to reduce procurement costs this can be effective even if it also sees rent-seeking behavior emerge.

“We found that the agencies would increase their search intensity and enlarge the pool of bidders if there was greater heterogeneity in the privately-known seller cost components,” the researchers conclude. “Our framework provides a template for analyzing other procurement auctions that attract only a modest number of bids.”