The federal government purchases nearly $500 billion in goods and services each year, and policymakers and administrators are looking for ways to improve procurement to generate value for taxpayers. By establishing more effective sourcing strategies, agencies can save money and utilize their budgets to further fund desperately needed maintenance and modernization projects.
To create this lasting efficacy in public sector acquisition, the federal government must first understand what it is buying. Truly understanding what the government buys is the foundation for all important purchasing decisions: what to buy in the future, how to buy it and whether to buy it at all.
The federal government maintains a database of every contract action above the micro-purchase threshold ($3,500), which totals more than 4 million line items every year. Despite this expansive data set, the government does not capture meaningful visibility into what agencies are actually buying.
This database contains a procurement classification field, called the Product Service Code (PSC). Unfortunately, PSCs are often inadequate to understand what is purchased. The federal government must find ways to use PSCs more effectively today, and, in the long term, must rethink the classification schema to get the most out of procurement data.
In the 1970s, the federal government developed the PSC classification, recognizing the importance of understanding and tracking the billions of dollars spent on goods and services each year. Reported within a centralized database called the Federal Procurement Data System (FPDS), PSCs categorize purchases using a series of more than 5,500 four-digit codes.
About every five years, the federal government revisits these codes, but the codes have not undergone a full restructuring since their introduction. This lack of a full overhaul has resulted in inadequacies, limiting the government’s visibility into public procurement and masking insights that could save billions of dollars.
Drawbacks to the PSC structure
PSCs are often outdated
PSCs were designed to categorize the goods and services that were being purchased at the time it was created. As such, some outdated PSCs that were important for the government’s business activities in the 1970s, but are less relevant today, remain prominently in the system, while many products and services that have now become more prevalent are not classified at a level to make actionable business decisions. For example, there is a PSC specifically for riveting machines (3448), a frequently used code in the 1970s. In fiscal 2015, though, this code amounted to a minuscule $41,000 in governmentwide obligations. In contrast, IT software (7030), amounting to $5.3 billion in obligations, is a single code. Within this single, massive code, the types of software are so numerous that it provides little insight for making effective business decisions.
Many highly-utilized PSCs are non-descriptive
The PSCs contain many “catch all” miscellaneous codes such as R499 (other professional services) that provide very little information on what they contain. Ideally, these codes would be used sparingly because the categories in the rest of the system would be comprehensive and descriptive enough to capture most purchases. However, these non-descriptive categories are extensively used. Almost $50 billion in 2015 was classified as “other” products and services. The heavy use of these categories prevents the government from truly knowing what it has bought.
PSCs are inconsistent with any system used outside of government
Another significant drawback to PSCs is that only the federal government uses them, limiting a wealth of comparative analysis between private sector, foreign governments, and international institutions.
There are a number of commonly-used international classification schemas that would enable comparisons. For example, by April 2017, the United Nations Standard Products and Services Code (UNSPSC) will replace the Government of Canada’s current system of commodity classification, and the United Kingdom Office of Government Commerce has stated that “UNSPSC is the recommended commodity coding structure for central civil government.”
The UNSPSC has also been adopted by several states in the U.S. Being able to compare data using UNSPSC or another broadly-used classification structures could provide a wealth of information not available with PSCs.
With these drawbacks, it is clear the federal government needs to PSC classification, and the Office of Management and Budget’s (OMB) category management initiative provides a great opportunity to do so.
OMB’s category management initiative establishes cross-agency teams to address improvement opportunities in acquisition categories common across departments and agencies. These opportunities may include building better acquisition packages, identifying best contracts, or improving management post award. As such, understanding what is being bought is key to making informed category management decisions.
Analyses performed for category management initiatives are often based on a mapping of PSCs, which limits the accuracy, usefulness and effectiveness of these efforts. Additionally, because federal category scopes do not consistently align with conventional markets, the insights and business decisions that are drawn from the analysis may not reflect the true scope.
But what can we do now?
Making a wholesale change to how government classifies its acquisitions will take time, despite the effort and substantial support for category management initiatives. Knowing the time investment needed to update PSCs, agencies can take steps to use PSCs today to identify and support valuable category management insights.
Use PSCs in combination with other data attributes in order to identify specific and targeted acquisitions. While PSCs alone are not always enough to identify procured goods and services, they can serve as an indicator when used with other attributes such as NAICS or Description of Requirement. For example, while there is no specific PSC to identify graphic design services, one could assume that graphic design services are classified as a professional services-related PSC. By filtering for these PSCs and then searching for graphic design-related NAICS and text descriptions, one could find relevant graphic design contracts.
Use PSC combinations to define markets. Codes can often be combined to define a market and gain valuable insights. Revisiting the example of the highly-specific riveting machine code (3448), this PSC could be combined with other PSCs such as punching and shearing machines (3446) and others to understand acquisitions on metal working machinery.
Use XX99 codes only when necessary. The singular value of PSCs is to provide data on public sector purchases of goods and services. XX99 codes, which are vague and do not supply adequate information on purchase types and categories, are too often selected as the default rather than the exception. Contracting Officers should take the extra time to identify the accurate PSC code when available. The future data integrity will be worth it in the end.
With the momentum currently behind category management, there is a unique opportunity to rethink the federal government’s acquisition classification system and take actions to improve it.
Aram Mazmanian is a principal at Censeo Consulting Group. His areas of expertise include business process reengineering, supply chain management, lean principles, organizational design, and workforce assessment. Most recently, Aram has worked with federal agencies to improve mission delivery through acquisition strategy, and organizational effectiveness initiatives.