OpenSource Connections
Category: Uncategorized
It is a well-known marketing maxim that if a vendor can raise the barrier to exit for a customer, then the vendor can charge incrementally higher costs. If buyers are locked into a vendor because of the belief that either the vendor is the only one who can provide the service/product, or, alternatively, that the costs of switching to a new vendor are too high to justify moving, then free markets are, effectively, eliminated for the buyer.
After being excoriated in the 1980s for excessive spending such as the $436 hammer, the U.S. federal government worked to open up its contracting practices to provide more clarity and, more importantly, more competition. This, in turn, should have made the process of government acquisition more effective and raised the return on U.S. taxpayer dollars. In most cases, this has worked, and the process is much better than it was in the 1980s; however, work remains to be done.
In this article, I shall conduct an analysis of government purchasing between February 18, 2004 and February 6, 2008 in three specific NAICS codes: 541511 (Custom Computer Programming Services), 541512 (Computer System Design Services), and 541519 (Other Computer Related Services) with the intent of seeing if the government suffers from excess vendor lock and look at ways that, if it does suffer from vendor lock, it can ameliorate those issues.
A cursory glance at presolicitations, special notices, synopses, and awards posted on Federal Business Opportunities (otherwise known as “FedBizOpps”) shows that 2,266 out of 8,997 postings, or 25.2% of all postings for the aforementioned categories and NAICS contain the term “sole source.” Naturally, one would expect that not all of these hits returns an actual sole source award or justification, so further digging is necessary. I have taken a sample of 76 of the 2,267 postings to look at the actual posting, which will yield a 95% confidence level with a 11.05% confidence interval. In other words, barring other biases, 95% of the time, results shown from this sample will have a +/- 11.05% margin of error. Its not enough for definite certainty, but provides directional correctness, and suffices for this research.
First off, as a consultant, I feel obligated to add in caveats about the research I am doing.
- Compared to the level of analytical rigor necessary to draw unassailable conclusions, this effort falls short. It is meant to be “quick and dirty” and directionally correct, and to draw attention to what may be a potential shortcoming in government procurement policies and procedures.
-
There are certainly times when sole sourcing is justified. These include, but are not limited to:
- When a need can truly only be filled by one vendor
- The level of capital expenditure to fulfill is extremely high, and has already been spent (an example of this might be the use of an existing and monopoly satellite system)
- The need is not readily met by open sourcing (conventional or non-conventional means)
- The government is looking to meet set-aside goals, and one set-aside category filling vendor is identified as capable of fulfilling the need
Whether or not a procurement should have been open sourced is open to interpretation, and I am providing my own interpretation. I tried to use the above for guidelines, and I readily admit that my approach is both normative and likely biased. Again, this is not meant to have unassailable rigor, but, rather, is meant to serve as a baseline and initial approach for trying to answer the question, as I can find no noteworthy research which addresses the problem.Of the 76 I analyzed, 72 were actual sole source justified requests for proposal/announcements. The other four were either intents to sole source after open competition, or in the case of one, a sole source to introduce a new vendor into competition. Of the 72 analyzed, I found that 44, or 61.1% were justified because of specific knowledge necessary to execute, existing wartime continuity, or because of a set-aside justification.
What this means to me is that there are many opportunities for procurement specialists in the federal government to improve their procurement methodology to avoid vendor lock in the first place. Many of these set-asides were maintenance agreements for proprietary systems, some of which were commodity or non-domain specific areas (e.g. travel and relocation services, asset management). If more work is conducted up front to identify long-term impacts of procurement decisions, then many of these situations can be avoided. It is also up to industry to identify areas where vendors can compete. Sources sought announcements go out frequently, and this is the governments method for market research. When only one vendor responds, then its hard for the procurement officers NOT to sole source.
The entire list is available for download here.
I will comment more on this research in subsequent postings. If you have any comments or questions about my methodology, feel free to leave a comment.