As part of our initiative around procurement, we’ve been meeting with several experts, inside and outside the government, to get a broad perspective on contracting in the federal government. Many agree that federal procurement suffers from problems, whether it’s over-budget, over-time contracts, the limited pool of government contractors, or the over-reliance of government on contractors to perform core governmental functions. What is the real source of these problems and how can we overcome them? Well the answer is pretty complicated, but over the last several weeks, the picture is coming into focus. Across all of these conversations we’ve had, there are several themes that have consistently stood out. Since we’re making an effort to blog about our work as we go, I wanted to step back and summarize what we’ve learned. While there will be a host of problems specific to one agency or type of procurement, these themes seem to apply across the government, and aren’t necessarily as earth-shattering as you might think.

Writing Good Requirements

Defining the scope and features necessary for a project is an important skill, whether you’re inside government or not. Sometimes this is difficult without having subject matter expertise in the exact product or service you want to buy. And the people you’re buying from may not always have the best incentives to keep your costs and scope manageable. Most people who have worked on projects with a large team have heard the phrase “feature creep” or suffered from having too many cooks in the kitchen. But the size of the federal government and the amount of goods and services it buys brings this issue to the forefront when citizens ask why we spend so much on contracting. This is exacerbated by the government’s tendency to buy all the things, as opposed to hiring new staff or creating products internally, thus retaining institutional knowledge about a field or topic area. Indeed, much of the current efforts in contracting reform are focused on making it easier for contractors to participate in the process. This is good for a lot of reasons, but doesn’t address the knowledge gap that allows the government to both assess whether a contract is needed at all, and also what exactly they need.

Lack of Data

When things go bad (or good!) in government contracting, it can be hard to detect why. Sometimes it’s inertia, sometimes there are specific problems within certain departments or types of contracts. Being able to identify trends in contracting stats can help procurement officers make better decisions. An uptick in solicitations for a single product might indicate that these solicitations should be consolidated across government to get a better deal (also known as strategic sourcing). Being able to see a trend as it’s happening will allow the government to act before contracts are awarded. Right now that data is in FedBizOpps, in mostly prose form. Statistics on the number of bids, the awardee, the contracted price, etc. all show up later in USASpending.gov, after the contract has begun. Being able to answer basic questions about who is procuring what, and when, is important in making evidenced based decisions.

There is also woefully little data on contract performance and management. The General Services Administration maintains a few different databases around contractor performance, including the Excluded Parties List (for debarred or suspended contractors), the Federal Awardee Performance and Integrity Information System, and the Past Performance Information Retrieval System (PPIRS, or hilariously pronounced “peepers”). Even these databases aren’t always used as they should be and there are far fewer systems dedicated to monitoring contracts over their lifetime. One example is the IT Dashboard, which reviews the IT investments of varying agencies, reporting whether they are on budget or on time. The IT Dashboard had some initial success, showing that when data about the performance of a contract is public, steps can be taken to either eliminate the contract or get it back on track.

Lack of Procurement Officers

Procurement is such a big part of the federal government, that we have a workforce dedicated to it. That workforce requires training in the Federal Acquisition Regulation (FAR) and procurement policies in general. That workforce is shrinking. Additionally, the workforce that oversees the suspension and debarment of contractors that don’t fulfill their contract or otherwise behave unethically, is also very small. Simply put, understaffed workers are going to have a harder time overseeing existing contracts and effectively creating new ones. Is this because we have too many contracts or too little staff? It’s probably some combination of both.

So, what steps should be taken to address these problems? Some are no-brainers, like providing more data around the contracting process (at least providing enough to put us on par with Russia). Others are more complicated and require more discussion with a larger community. There are several groups in this space working on this issue from different perspectives that all have a unique take on the problem (including the Project on Government Oversight, Code for America, and the Department of Better Technology). We’ll keep posting whatever we learn and hopefully generate some good discussion along the way.