There are many misconceptions relating to government contracting, especially amongst technologists, so I thought I would take this opportunity to clear some of those confusions up.
I once bid several cents per e-mail to send out thousands of e-mails a month for a government organization. Ridiculous, right? Anybody can run sendmail in a colo for $100 a month. What the actual term of the agreement doesn't say, is that the e-mails were to be sent from an application we were to develop with features unique to the organization, and the e-mail addresses were to be collected using a marketing website and software package that we were to construct, maintain, and promote. We also had to provide two dedicated T1s, four dedicated servers, a load balancer, as well as design and produce all the print marketing materials to promote the new service. All of these things were included in the contract, but we were only paid per e-mail sent. Things aren't always as they appear at first glance.
In the case of the bolt, it's not an ordinary bolt. Normal bolts are never individually tested, a single bolt from the lot is taken and destructively tested. In the case of the expensive Department of Defense bolts, they are generally one of a kind, limited production, bolts designed for one purpose. In addition, they are generally non-destructively tested, which means that they are each individually subjected to the forces that they are rated for, and then examined. This is expensive.
For the hammer, the situation is similar. The price was inflated because of the equalallocation formula in use at the time. "The equal allocation method calculates prices for large numbers of items in a contract by assigning "support' costs such as indirect labor and overhead equally to each item. Take a contract to provide spare parts for a set of radar tracking monitors. Suppose a monitor has 100 parts and support costs amount to a total of $100,000. Using the equal allocation method each part is assigned $1,000 in such costs, even though one item may be a sophisticated circuit card assembly, which requires the attention of high-salaried engineers and managers, and another item may be a plastic knob. Add $1,000 to the direct cost of the part and you get a billing price. This is what the government is billed, though not what the part is really worth--the circuit card being undervalued, the knob being overvalued. The need for billing prices arises because contractors want to be paid up front for items that are shipped earlier than others." (from The case for the $435 hammer.)
As far as "cost plus" goes, there really isn't a better way to do what it does. Whenever I bid a contract I estimate cost, then add profit and that is the price. In the case of the e-mail contract I described above, I calculated the cost then decided on a fair profit. After that, I made best case, worst case, and average case estimates for e-mail volume. I ended up basing the per e-mail bid on the worst case number of messages sent. In other words, the bid price was ((cost + [slightly less than fair] profit) / worst case estimate). Had we won the contract and not gotten close to the worst case, the profit would be substantial. Had it been a "cost plus" contract, it probably would have been less expensive for the government overall, however, the risk would have been theirs, not mine (if our software was ineffective or underused, we could have potentially lost money).
Cost plus is most often used when something has to be built that is either difficult or impossible to estimate. If I were to ask you to build something that nobody has ever built before, and intended to have you sign a contract saying that you would construct it for that price, you'd probably greatly overestimate the actual cost, because you would have to make sure you don't end up too far in the red. The costs are evaluated and approved by an oversight group (like government engineers), so they can make sure project costs are really necessary. In addition, the records are audited and unnecessary cost is often disallowed. Cost plus isn't perfect, but it's less expensive in the long run then having the contractors make guesstimates then inflate them to deal with the risk and uncertainty.
In the long run, the single most disingenuous thing I've seen in government contracting is the blatant racism and sexism. Females and minorities are given preferential treatment because of their race or gender. Depending on the contract, their price proposals are also evaluated differently as well, often getting a 5% discount. In other words, a $100k bid placed by a Minority-owned business will be read as $95k when compared to other bids. The process is not only unfair and discriminatory, but can result in less qualified firms winning contracts on the basis of quotas.
In the end, the ultimate check and balance is the openness of the process. Anyone can put their money where their mouth is, start a company, and win some contracts. All you've got to do is demonstrate that you can do the work, and bid low.
Use the following link to trackback from your own site: