'Reinventing government,' 25 years later

We are coming up next year on the 25th anniversary of the country's first and perhaps only governmentwide management reform program organized around a coherent theme: the Clinton administration's "reinventing government" effort.

In 1993, the first year of the Clinton administration, I went on leave from my job at Harvard University's Kennedy School, where I was a professor of public management, to take a Senate-confirmed position in the Office of Management and Budget as administrator of the Office of Federal Procurement Policy. That office of 30-odd civil servants did not buy anything itself but had the lead role in formulating governmentwide procurement policy.

At the beginning of the 1990s, the thinking about how to manage well in government began to turn toward performance. As political scientists William Gormley and Steven Balla have written, "The concept of performance came to rival accountability as a standard for evaluating executive branch agencies."

The concept came into its own early in the Clinton administration. In March 1993, President Bill Clinton launched the National Performance Review, led by Vice President Al Gore and later renamed the National Partnership for Reinventing Government. The resulting report, "From Red Tape to Results: Creating a Government That Works Better and Costs Less," was released that September. One of the first laws Congress passed in 1993 was the Government Performance and Results Act. Suddenly, the words "performance" and "results" were everywhere.

The events of 1993 launched major changes in the procurement system, which has continued to evolve in the past 25 years. In general, that evolution has seen the procurement culture shift its focus from compliance to performance, yet despite that shift, it is hard to say that the system's performance has improved.

That is due in part to the fact that the changes in procurement coincided with an increase in problems for contract management. The system was getting better, but contract management was getting worse — so much so that we haven't noticed net improvements.

Performance vs. compliance: The fundamental question

Every organization has goals and constraints. An organization's goals are what it is trying to accomplish, while its constraints are the ethical and legal restrictions on what the organization may do to accomplish those goals. If you are Microsoft, your goals are to produce products customers want to buy and to make a profit. Your constraints include restrictions such as not destroying the environment while producing your product, paying your taxes and refraining from kidnapping your competitors.

In the procurement system, the government's goal is to obtain best-value products and services for agency missions and taxpayers. Constraints include integrity (e.g., not taking bribes) and impartiality when working with contractors. For contractors, those constraints center on honesty in dealing with agency customers.

A focus on goals directs us to maximizing the occurrence of good things, while a focus on constraints stresses minimizing the occurrence of bad things. In the context of government procurement, a constraints focus translates into an emphasis on avoiding waste, fraud and abuse, while a goals focus tends to emphasize innovative acquisition methods, with the view that new ways of doing things can produce major performance upsides even though they create a downside risk of failure.

To put the idea into language familiar to federal officials and their contractors, constraints direct our attention to compliance, and goals direct our attention to performance. Constraints favor a legal mentality, and goals favor a management mentality.

So which should be more important to an organization — goals or constraints?

Constraints reflect important ethical ideals. Furthermore, when levels of corruption and cronyism in a procurement system are high, companies will be hesitant to bid for government contracts. That reduces the system's performance, especially because high-quality businesses will be the most hesitant to bid. More broadly, rampant corruption and cronyism can be devastating for the overall performance of an economy because potential entrepreneurs might invest energy into getting favors from government rather than running legitimate businesses.

However, two points should be noted about organizations that focus on constraints. The first is that an individual or organization concerned only about constraints will seldom be seen as successful. The second is that if satisfying constraints consumes too much time and energy, individuals are unlikely to do so because they won't have enough resources.

A well-functioning organization is one in which constraints are so baked into culture and practice that people don't need to think about them. Most people would chuckle at my earlier mention that one of Microsoft's constraints is not to kidnap its competitors because such behavior would never be part of the company's business plan.

A good way to think about the role of constraints in a well-functioning organization is to say that the organization should maximize attainment of its goals while respecting its constraints.

1990s: Moving to a performance culture

In the past 25 years, the emphasis on constraints has given way to a performance culture in government, and procurement debates are now typically framed in the language of performance rather than of compliance. Most of that change took place during the 1990s.

This essay is the cover story of FCW's November/December issue. Click here for the print edition.

More from Steve Kelman

The Lectern blog explores acquisition reform, public sector innovation and executive leadership. Click here for the full archives.

The change was officially proclaimed in new language under Part 1 of the Federal Acquisition Regulation, specifically the "Statement of guiding principles for the Federal Acquisition System" adopted in 1995. The reinventing government report called for replacing the entire FAR with a series of guiding principles modeled on Australia's revamped procurement regulations, which were only 93 pages long. What happened instead was that the voluminous FAR remained, and the statement of guiding principles was added.

Two elements of the 1995 additions are particularly noteworthy. The first reflected the new view of the role of goals and constraints. "The vision for the federal acquisition system is to deliver on a timely basis the best-value product or service to the customer, while maintaining the public's trust and fulfilling public policy objectives," the principles state. In other words, goals take priority.

The second message was to deemphasize the role of rules. In a system tilted toward constraints, rules play an important role because they are an effective way to communicate a message that a certain behavior is always prohibited or required. In a system oriented toward compliance, one complies with the rules.

However, if the emphasis shifts to performance, the system often requires approaches tailored to situations that cannot typically be commanded by rules. What are sometimes called "rules of thumb" or "standard operating procedures" might communicate the results of past learning about what works.

The message was to reduce the rules' hold on people by counteracting the then-widespread view that if the rules did not specifically authorize a certain behavior, that behavior was prohibited.

"The role of each member of the acquisition team is to exercise personal initiative and sound business judgment in providing the best-value product or service to meet the customer's needs," the updated FAR declares. "In exercising initiative, government members of the acquisition team may assume if a specific strategy, practice, policy or procedure is in the best interests of the government and is not addressed in the FAR, nor prohibited by law (statute or case law), executive order or other regulation, that the strategy, practice, policy or procedure is a permissible exercise of authority."

This important change, perhaps surprisingly, survived strong headwinds that could have sunk it. The first challenge came from scandals in the 1990s involving misuse of government credit cards for buying items for personal use. Unscrupulous individuals took advantage of a reform that allowed purchases under $2,500 to be made directly by a government customer rather than having the procurement office review and process the purchase.

When efforts had been made in the past to loosen constraints on the system, they typically collapsed at the first whiff of scandal. This time, to my astonishment, that didn't happen. New checks were set up to detect misuse, but very few people suggested abandoning the change.

An even greater challenge, as will be noted in a somewhat different context later in this article, was a "perfect storm" during the George W. Bush administration of a procurement leadership more sympathetic to a constraints-oriented legal perspective than to reinventing government, efforts by Democrats in Congress to emphasize scandals and compliance as part of a partisan battle over the Iraq War, and severe contract management understaffing that threatened the system's performance.

Yet the changes of the 1990s largely survived even those challenges. A performance orientation returned during the Obama administration and appears to be continuing under President Donald Trump.

Has the system's performance improved?

If one looks at the results, has the turn toward performance produced an improvement in procurement practice?

The 1990s did produce some improvements, but they mostly involved reducing the irritation and friction the system created for its government customers rather than improving the quality and reducing the cost of what the government was buying. When I came into government, a colleague at the Defense Department reported waiting four months for the purchase of a $40 Dictaphone machine, and medium-sized IT service contracts of around $1 million typically took 18 months to award.

The government credit card (whose use was dramatically expanded starting in 1993) and the development of multiple-award indefinite-delivery/indefinite quantity contracts (authorized by the Federal Acquisition Streamlining Act of 1994, which advocated the use of task orders rather than a fresh procurement for each requirement) notably decreased the delays and frustration of career employees.

However, the bigger question is whether there are large categories where one can say that the results delivered have clearly and visibly improved in terms of quality and/or price. Here there are fewer reasons for optimism.

In important ways, we don't know the answer to this question because there are so many different contracts, and they change over time, making apples-to-apples comparisons difficult. But one would expect that if there had been clear improvements across at least most of the board, they would be visible. As a general matter, I don't believe we can claim that.

As I scour the procurement landscape for proven results, three areas do come to mind. All three date from 2009 or later, and two remain works in progress.

The most important performance improvement is the reduction in cost growth for developing new weapons systems, the largest cost category in the government's procurement budget. The fact that the systems require beyond-state-of-the-art technology creates significant risk in terms of cost growth and performance uncertainty, and it necessitates cost-reimbursement pricing that is not conducive to controlling costs.

Now, however, there seems to be plausible evidence that the cost growth for developing new systems has decelerated since 2009. That trend appears in evidence provided by DOD and confirmed by the Government Accountability Office.

Officials believe the single most important reason for that downward trend was the spread of "should-cost" analyses for programs under development. Those analyses, which are conducted every year, are filled with concrete suggestions for how a program might reduce costs. Acquisition executives in the military services develop overall cost-savings targets for each program's should-cost analysis, and program managers are supposed to track progress the same way they would with other performance metrics.

Those analyses have been crucial to creating counter pressures on acquisition officials. Without that metric, they would focus only on the traditional metric of "getting money out the door," which created incentives against cost control. 

Impact of agile: Too soon to tell?

Two other areas that show promise involve IT, though it is still too early to tell. One is the acquisition of IT systems. The big-picture, 25-year view here shows a system still beset by significant underperformance, with projects regularly showing big cost and schedule increases and performance shortfalls. In 2015, GAO put IT acquisition on its list of agencies and program areas at high risk of waste, fraud, abuse or mismanagement.

It now seems possible that using agile software development methodologies rather than traditional waterfall techniques might be helping, but we don't have a lot of evidence yet. Executives in the office of the U.S. CIO told me they don't have enough data to venture even a tentative conclusion. The accounts I give here are based on reports by participants, who obviously have their own biases, and some media accounts.

The first major use of agile in the federal government was a transition at U.S. Citizenship and Immigration Services from a traditional procurement approach to agile for an organizationwide business process redesign in an agency that had been mostly paper-based. It was led by CIO Mark Schwartz, who came to the agency from the private sector in 2010.

It had been a classic troubled big project. USCIS spent five years automating a single, limited business process, which later had to be discarded because it wasn't effective for agency adjudicators. After arriving in 2010, Schwartz started doing four releases a year. In 2015, he moved to the cloud and full-fledged agile development. Two years later, Schwartz reported, the agency had digitized 30 percent to 40 percent of its volume. By the time he left in 2017, the development team was doing two to three releases a week.

Dave Zvenyach, who until October ran the acquisition team at 18F and managed the General Services Administration's blanket purchase agreement for agile services, shared what he observed during the transition from waterfall to agile for development of a searchable government spending website required under the Digital Accountability and Transparency Act of 2014 and for task orders under the agile BPA.

The Data Act requires taking elements from perhaps 600 federal data sources related to spending on contracts, grants and payments; standardizing the data; and putting it into one online, searchable system. Earlier, the agency had taken four years to develop an analogous capability for accounting data for a similar project using waterfall methodology. Yet when the new spending reports began posting to USASpending.gov in early 2017, legislators and GSA officials praised the team for implementing the changes on schedule and without drama.

"There was no breaking news about the website crashing," Sen. Rob Portman (R-Ohio) said at an event soon after the launch. "It went as planned."

"It worked because of the agile development process," added Rob Cook, commissioner of GSA's Technology Transformation Service.

As for the agile BPA, there have been six task orders to date (the contract is only available for 18F customers), with values ranging from $140,000 to $2.5 million. Zvenyach, who is now acting assistant commissioner of the Office of Systems Management in GSA's Federal Acquisition Service, said every one of them has been successful.

The promise of category management

The second area that shows promise is category management — an initiative launched in 2014 that aims to get lower prices, better terms and better customer service by buying IT commodities "as one government." Category management is a sort of Release 2.0 of strategic sourcing, which dates back to the 1990s and became White House policy in 2005. The mantra, then as now, was leveraging government's buying power by obtaining quantity discounts for large volumes. However, a 2012 GAO report concluded that the government capitalized on only a small portion of its buying power.

The officials working on category management cite all manner of ways that the initiative differs from strategic sourcing. However, by far the most important difference is that OMB has moved to designate specific "best-in-class" contracts in different commodity areas, and agencies are required to use those contracts unless they receive permission to opt out. Strategic sourcing failed because, for whatever reason, there was only modest agency uptake of strategically sourced vehicles. They built it, but people didn't come.

This essay is the cover story of FCW's November/December issue. Click here for the print edition.

More from Steve Kelman

The Lectern blog explores acquisition reform, public sector innovation and executive leadership. Click here for the full archives.

This changed dramatically when OMB issued a memo in 2015 announcing new policies for buying workstations. The memo directed agencies to move aggressively to replace open-market purchases and single-agency vehicles with one of three "best in class" laptop and desktop contracts — GSA's Schedule 70, NASA's Solutions for Enterprise-Wide Procurement and the National Institutes of Health IT Acquisition and Assessment Center's CIO-CS. The Army's Computer Hardware, Enterprise Software and Solutions contract was added later as a fourth option.

There was a temporary exception through 2017 for agencywide contracts with mandatory use, and there was also an opt-out procedure. In addition, OMB's memo sought to save purchase and administrative costs by announcing that there would be only five standard workstation configurations (with functionality dimensions and performance specifications such as memory), and the content of the configurations would be reviewed and revised annually. The memo stated that all workstations would need to use one of those configurations unless an opt-out was approved, with a goal that 80 percent of purchases would adhere to one of those five configurations.

Lastly, the memo said the government would organize "buying events" twice a year where agencies would commit to specific purchase volumes, with the aim of gaining further discounts from published prices.

The workstation memo was seen as only an opening shot in introducing category management into other spending categories beyond IT. OMB issued a list of 10 "supercategories" in 2016.

Category management seems to have a decent chance of succeeding, especially compared with the last time the government tried to consolidate purchasing in the 1980s, using the GSA schedules. At that time, GSA had a monopoly, which created poorer pricing, and the governance procedure was GSA diktat. But its control collapsed in the early 1990s when other agencies revolted (and reinvention's decentralization took hold). Today, there are several approved consolidated vehicles, and governance is much more participative.

On at least one dimension, early returns are promising. At the time of the 2015 OMB memo, about one-third of workstation spending went through the three "best in class" vehicles; today, it is about 60 percent — not bad considering ongoing agency unease about losing autonomy. A single-award mandatory-use contract for small package delivery yielded savings between 3 percent and 16 percent compared to the most-used previous non-mandatory contract, depending on the type of service. For desktops, officials have provided anecdotal examples of better pricing.

However, the government still has only sporadic data comparing what it pays for commodities to what large commercial customers do. And for IT hardware, any comparison would be muddied because of the impact of the government's Buy American constraints.

One key question mark is the impact of consolidation on the government's endemic problem with the deterioration of prices over time on IT that is refreshed with new products. Great prices at the time of a contract's award are gradually replaced by prices that aren't as good. On theoretical grounds, one could make predictions that consolidation into fewer contracts could speed or retard such price deterioration. That is an empirical question worth watching.

Why didn't performance improve more?

The obvious question is this: If the system's culture had changed in a positive direction during this period, why didn't performance improve more?

My answer is that improvements in the procurement culture coincided with a major decline in resources available for contract management. The culture was getting better, but contract management was getting worse so that we have ended up not noticing net improvements.

Problems with contract management occurred with the downsizing of the procurement workforce that began during the 1990s, when the number of contracting officials declined by 21 percent. The number of employees at the Defense Contract Management Agency, which mostly works on post-award contract management, declined by half.

In the interests of honesty and completeness, I should note that the reinvention that produced a procurement system more oriented toward performance was also the one that produced cutbacks in the procurement workforce, which prevented improvements in the system from being translated into better performance. If one reads the reinventing government report, one of the biggest sources of the $108 billion in savings it promised was a 272,000-employee cutback in the federal workforce, a number that often became the headline recommendation of the entire report.

Although reducing the federal workforce is always popular, the reinventing government report had a broader rationale. It states that "most of the personnel reductions will be concentrated on the structures of over-control and micromanagement that now bind the federal government," and it includes procurement specialists and auditors on that list. Procurement employees were associated with a focus on compliance rather than performance, hence the link between getting rid of procurement employees and focusing on performance.

Without wishing to deflect blame from myself, I should note that the report was written before I arrived in Washington. I was not involved in the downsizing decision, and it was not something I either defended or criticized publicly or privately at the time. Yet I raised no internal objections, accepting it as an important part of the vice president's reinvention talking points. In 1993, I did not think about procurement people being important for successfully managing contracts. However, by my last two years in Washington, I had begun to argue that one purpose of streamlining contract awards was to free up resources for post-award contract management.

During the downsizing of the 1990s, procurement spending rose fairly modestly, by 18 percent from 1992 to 2000, and defense procurement spending decreased by 26 percent. A decline in the need for contracting resources during the 1990s — mostly caused by the introduction of government credit cards and IDIQ contracts — probably meant that the resources available for contract management did not decline in those years.

However, a big increase occurred after the 2001 terrorist attacks, with procurement spending more than doubling from 2001 to 2008 in tandem with a similar increase in defense spending. During that time, despite the vast increases in workload, the size of the contracting workforce was almost stagnant; it increased just 8 percent from 2001 to 2008, with no increases at all until 2005.

That situation left the system reeling due to insufficient management resources. Employment at the Defense Contract Management Agency was actually about 20 percent lower in 2008 than 2001, though the workload had doubled. Also during those years, the trend continued toward a larger percentage of contracting dollars being spent on services — from 23 percent in 1985 to 63 percent in 2014 — and those contracts generally require more resources to manage than contracts for products.

Added to all this was the new procurement leadership's orientation toward constraints rather than goals under President George W. Bush (the first head of Bush's Office of Federal Procurement Policy was a procurement lawyer) and partisanship between the White House and Democrats in Congress over the Iraq War, which led the Democrats to use procurement scandals to oppose the war. Those political battles also fed an orientation toward constraints.

A continuing focus on innovation and improvement

Again, it is remarkable that the combination of insufficient staff and political headwinds did not derail the new goals orientation of the 1990s. It did go into hibernation, though, and no innovations are associated with those years.

Since 2009, the decline in the procurement workforce has been modestly reversed, with the number of contracting officials governmentwide increasing 22 percent by 2015. The Defense Contract Management Agency also increased its headcount during the same period by about 20 percent. And after the dramatic post-2001 increase, spending fell by 19 percent, with the steepest declines in 2014 and 2015.

Together, those changes gave the contract management system a bit of breathing room. In DOD's effort to slow cost growth, the increased staffing levels provided resources for developing and managing should-cost analyses, determining negotiating positions and spending more time in negotiation rather than agreeing to a deal too quickly.

Those changes are creating an opportunity for an increase in the resources and attention to post-award contract management that could yet produce the kind of noticeable improvement in performance not seen in the past quarter-century. But that improvement will not happen automatically.

The increase in resources has not yet given birth to an overall performance turnaround. And it's not as if the system's performance was so wonderful at the beginning of this 25-year period, even though at that point resources devoted to contracting were far greater than in the years to follow.

This essay is the cover story of FCW's November/December issue. Click here for the print edition.

More from Steve Kelman

The Lectern blog explores acquisition reform, public sector innovation and executive leadership. Click here for the full archives.

However, the improvement in DOD's weapons cost performance since 2009 suggests that, with appropriate leadership attention, a recovery in resource availability could produce improved results. Executing what I have called a "pivot to post-award" in contract management should, in my view, be the highest priority for those leading the procurement system.

Perhaps the most significant overall positive surprise coming from the Trump administration has been its approach to improving government management. Although one might have expected that the issue would never make it onto the administration's agenda — indeed, I expressed that concern in a number of FCW blog posts after the elections — that has not been the case.

The Office of American Innovation (with leadership from the top in the person of Trump's son-in-law, Jared Kushner); the president's tweets about getting better procurement deals; high-level meetings with technology leaders interested in digital government; and the pleasant surprise of continued support for the Obama-launched Presidential Innovation Fellows, the U.S. Digital Service and 18F initiatives reflect an interest in government management that is extraordinary.

Furthermore, the Trump administration has embraced an innovation and improvement approach that is very different from the focus on waste, fraud and abuse that has historically been associated with Republicans. This administration seems to want to continue to focus on performance rather than compliance.

It is still early, of course, but there is at least a chance that the years to come will produce the kind of noticeable improvement in the procurement system's performance that many of us had been hoping for in the past 25 years.