Designing Government by Design

Robert A. Knisely, Esq.

September 7th, 2010

“There are those that look at things the way they are, and ask why? I dream of things that never were, and ask why not?”– Robert Kennedy (1)

“I am not an advocate for frequent changes in laws and constitutions. But laws and institutions must go hand in hand with the progress of the human mind. As that becomes more developed, more enlightened, as new discoveries are made, new truths discovered and manners and opinions change, with the change of circumstances, institutions must advance also to keep pace with the times.” — Thomas Jefferson

Executive Summary

In the absence of government, according to Thomas Hobbes, there are: No arts; no letters; no society; and which is worst of all, continual fear, and danger of violent death: and the life of man, solitary, poor, nasty, brutish and short.” Government’s role, then, is to change all that – to alter outcomes.

To do so effectively requires an understanding of society – past and present — and of past efforts to alter outcomes. In an age of increasingly complex, critical, and urgent problems, governments must also anticipate the future.

The resources needed include broad-based perspectives on society’s problems, accessible evaluations of past interventions, an extensive toolkit, a deliberative process, and a thorough vetting of proposed actions.

None of these are available to, or used by those designing government in America today.

A. Broad-based Perspectives

Any attempt to solve a problem must begin with an understanding of the problem in all its complexity – “things the way they are.” California’s restructuring its electricity market, as discussed in Chapter Two of the book “If We Can Put a Man on the Moon” makes this point. In sum, “After promising to reduce electricity costs by 25 percent or more, the new scheme saw electricity prices shoot through the roof, from around $30 to upwards of a staggering $1,000 to $1,500 per megawatt hour.”

We pride ourselves that ours is a government “of laws, not of men.” But as Oliver Wendell Holmes remarked, “In the law, an ounce of history is worth a pound of logic.” Apart from logic , we also need to enlist all of the relevant sciences. As I have said elsewhere, “Of the sciences overlooked, first and foremost are the sciences of complexity and the systems sciences developed around the time of WWII. Norbert Weiner and cybernetics come immediately to mind. There are many others, from game theory to highly reliable organizations.” Perhaps the most significant omission from the design of government is system dynamics. Computer simulation could be used both to sharpen our understanding of society’s interactions and to anticipate problems with our interventions. See Jay Forrester’s 1971 paper, “The Counterintuitive Behavior of Social Systems.”

Recommendation: Task the Congressional Research Service to create a checklist of the analytical methodologies and tools needed to describe societal problems and the systems surrounding them in sufficient detail to support legislative action. Provide for assistance from the General Accountability Office, the National Science Foundation, and the National Academies of Science and Engineering, as well as university centers such as Harvard’s Kennedy School of Government, MIT’s Sloan School of Management, and SUNY/Albany’s Rockefeller College of Public Affairs and Policy.

B. Accessible Evaluations

We need insightful evaluations of what’s working and what’s not, and why. These feedback loops in the Federal government have been evolving over the past half-century — or perhaps styles just change.

In the 1960’s and 1970’s, “program evaluation” was popular. A search for “program evaluation” on the General Accountability Office’s website returns 4,573 hits. In the mid-‘70’s I started an Office of Program Evaluation at the Department of Commerce. In 1989 I became Deputy Assistant Secretary for Budget and Programs at Transportation. The earlier office title, Budget and Program Evaluation, had been changed. No formal evaluations were being done or requested.

The less said about Zero-based budgeting the better. It was a passing fad of the Carter Administration (1977 – 1981).

The [now] IBM Center for the Business of Government published “Using Evaluation to Support Performance Management“ in January 2001. It states “[N]ine of the 13 federal departments contacted for this study had downsized their central evaluation capacity since the 1980s or never had a central evaluation office. The number of staff for program evaluation work has declined steadily from the late 1970s into the 1990s.”

The Government Performance and Results Act was passed in 1993, sounding the death knell for program evaluation. GPRA only looks forward: to plans, goals, and objectives one and five years in the future. Yet GPRA is now fading from view. A search for “GPRA” on GAO’s website returned 202 hits between 1992 and 1999, 91 between 2000 and 2009, and only one for 2010.

Mitch Daniels, President George W. Bush’s Office of Management and Budget Director, introduced the Program Assessment Rating Tool in a memo dated January 14th, 2003. PART’s archive at OMB has guidance only through FY2008. The website promising PART results — — also ends with FY2008 data. I found only two references to PART on the OMB website.

On February 18th, 2010, Jeffrey Zients, President Obama’s Chief Performance Officer, laid out his agenda: “Here are the six performance strategies: eliminate waste, drive top priorities, leverage purchasing scale, close the IT performance gap, open government to get results and finally, attract and motivate top talent. These are the six strategies that represent the biggest opportunity to boost performance and get government working for the American people.”

When “government is working for the American people,” our children get educated, the eggs we buy are free from salmonella, the American economy is creating jobs, medical research is tackling diseases, and so forth. We hold competitions, set safety standards, regulate capital flows, fund medical colleges, and on and on. These are the tools of government.

Yet nowhere, from program evaluation to performance strategies, is there any discussion of whether the Federal government picked up the right tools to ensure that government is working. And there is no collection of program evaluation, or GPRA, or PART, or performance strategy reports. These would need to be extensively indexed by tool, and made available for research into the choice of tools for the next problem.

The tool is usually a “given” from the perspective of the Executive Branch, and all too seldom are the questions asked, “Can this tool solve this problem?” and “What are the possible unintended consequences of wielding this tool?”

Two quick examples come to mind. First, as discussed elsewhere, President Nixon’s Wage and Price Controls could not effectively control prices. At a restaurant on Capitol Hill the price of the souvlaki stayed the same, but the portions got smaller over time. Ashby’s Law of Requisite Variety made nationwide enforcement of Wage and Price Controls impossible. And they failed.

Second, again discussed more fully elsewhere, the introduction of food stamps in rural Louisiana (a) enriched the local storekeepers, (b) permitted both the hiring of white clerks and the firing of Black warehousemen, and (c) was soon driving the poorest of the poor out of state.

Liberals such as Hubert Humphrey and Estes Kefauver had pushed for food stamps for decades. President Kennedy announced a pilot program in February 1961, and President Johnson made it permanent in 1964. The program required that recipients purchase their food stamps. I found that the really poor people of Louisiana were not in the cash economy at all. They could not pay for the stamps. And so they left for the cities of the North.

I’ll never believe that’s what Hubert Humphrey had in mind.

Recommendation: Task either CRS or GAO with creating and maintaining an online database of studies of the effectiveness of government interventions, in both America and abroad, indexed by tool (intervention type), and structured for use in tool choice. Provide sufficient resources to enlist the support of the academic community.

C. An Extensive Toolkit

The “Bible” of these tools is a 669-page book, “The Tools of Government” edited by Lester Salamon. Dr. Salamon directs the Johns Hopkins Center for Civil Society Studies in Baltimore. Amazon lists 91 publications by Dr. Salamon. The Oxford University Press published the book in 2002. An earlier version of the book, “Beyond Privatization: The Tools of Government Action,” was published in 1989.

“Tools of Government” lists fifteen tools of public action, and has chapters on each ranging from about twenty to forty pages. There are an additional five chapters, from introduction to summary. The tools are:

• Direct Government

• Government Corporations and Government-Sponsored Enterprises

• Economic Regulation

• Social Regulation

• Government Insurance

• Public Information

• Corrective Taxes, Charges, and Tradable Permits

• Contracting

• Purchase-of-Service Contracting

• Grants

• Loans and Loan Guarantees

• Tax Expenditures

• Vouchers

• Tort Liability

I would add the criminal law, and there may be other tools yet to be invented. These are more than enough to start with!

Many, many years ago, I used to play a game at Washington parties. Friends would come up with a societal problem, and I would attempt to find at least three “tools” that could be used to fix it. (They were really wild parties.)

Here’s a current example. Massey Energy was cited for 1,342 safety violations over the past five years. In only 25 of these, fines of $10,000 or more were proposed, and Massey has settled only four. In 2009, Massey had gross sales of $2.3 billion on 36.7 million tons of American coal. Massey’s lawyers and the government’s lawyers are arm-wrestling over peanuts.

President Obama’s FY2011 budget for the Mine Safety and Health Administration (MSHA) calls for over $350 million and 2,430 workers. This is for the regulations that have been generating resistance, not compliance.

Since coal mining is inherently dangerous, the Congress should require each mine owner to pay the heirs of every minor killed or injured on their property a sum adequate to encourage the owners to take responsibility for safety. Each fatality should cost at least $10 million, payable immediately. The 25 fatalities from the Upper Big Branch mine would have cost $250 million, or about 25% of Massey’s 2009 net income. This would get even Don Blankenship to focus on safety in his mines. It would also bring some relief to the families.

The coal industry’s response to such legislation would be to buy insurance, and the insurance industry’s response would be to insist on the best and most up-to-date safety procedures from around the world. There would be no lengthy publication and comment period, as required for Federal regulations. The insurance companies would then do the inspections to assure compliance. And the cost of the insurance would be passed on to the consumers of the coal, rather than all taxpayers.

In sum, Tort Liability would be a cheaper and more effective tool to achieve safer coal mining than Economic Regulation, and it would allocate the costs correctly.

We need both principals and staff in the Legislative and Executive Branches to take a similar agnostic and open-ended approach during their initial deliberations about the problems they face. For as the psychologist Abraham Maslow once said, “if the only tool you have is a hammer, you tend to see every problem as a nail”.

Recommendation: Task OMB with implementing Recommendation DES01: Activate Program Design as a Formal Discipline from Vice President Gore’s National Performance Review (1993). (3)

D. A Deliberative Process and Thorough Vetting

The design of a particular intervention should be in three parts. First, an initial scan by an apolitical entity. Second, a wide-ranging team approach that develops several proposals for further review. Third, the widest possible review (i.e., “crowdsourcing”) of the chosen option and the decision-making process itself.

1. The Initial Scan. The initial scan should be performed by an apolitical organization or organizations. The Congressional Research Service, aided perhaps by the National Academy of Sciences, would be an obvious choice. The CRS should be asked to respond to the following questions, among others:

a. Describe the problem being addressed, from an inclusive set of perspectives: historical, economic, systemic, and others.

b. Describe past and present attempts by American institutions (Federal, state, local, private, not-for-profit) to alleviate the problem.

c. Describe past and present attempts around the world to alleviate the problem.

d. Describe past and present attempts around the world to alleviate similar problems.

e. Compare the results obtained by the various tools used in addressing this and similar problems, in terms of cost, effectiveness, and unintended consequences, among other impacts.

f. Annotate all responses fully, including hyperlinks in an electronic version of the resulting report.

g. Accompany the full report with two summaries: one of two pages or fewer, one of ten pages or fewer.

2. The In-House Proposals. The in-house proposal should be the work product of a small team, including at least one attorney and preferably a design professional.(2) Ideally, the team would have representatives of both the Executive and Legislative Branches.

a. The group should contain, and also have access to, expertise in the problem area at every level from conceptual to service delivery. Members of the group should be encouraged to seek additional advice.

b. The group should engage in scenario planning or other future envisioning in the problem area as well as in the larger society.

c. The group should develop at minimum three proposals to address the problem under study using different tools (see above).

d. The proposals should be “red-teamed” (i.e., strongly critiqued) by a separate team of experts within the same organizational context.

e. The proposal chosen should be developed more fully, with particular emphasis on the likely unintended consequences.

f. The final proposal should be incorporated into a draft Legislative Impact Statement, which must include the initial scan verbatim. The draft LIS must be prepared in accordance with a set structure and format.

3. The Public Proposal.

a. The draft Legislative Impact Statement must be made available in electronic form to the public for comment in a timely manner. Comment must be encouraged.

b. Resources must be made available for the review and consideration of all comments.

c. A final Legislative Impact Statement must be again made available to the public in electronic form prior to any implementation.

Recommendation: Task OMB with implementing Recommendation DES04: Commission Program Design Courses from Vice President Gore’s National Performance Review (1993).


(1) Robert Kennedy was quoting a line delivered by the Serpent in George Bernard Shaw’s play Back To Methuselah : “You see things; and you say, ‘Why?’ But I dream things that never were; and I say, ‘Why not?’”. (Source)

(2) Design as a discipline.

In 1969, in The Sciences of the Artificial, Herbert Simon wrote: “…The professional schools will reassume their professional responsibilities just to the degree that they can discover a science of design, a body of intellectually tough, analytic, partly formalizable, partly empirical, teachable doctrine about the design process.
“…[S]uch a science of design [is] not only possible but is actually emerging at the present time. It has already begun to penetrate the engineering schools, particularly through programs in computer science and ’systems engineering,’ and business schools through management science.”

This has come to pass. Check out the Stanford University School of Design, read this introduction to the world of design, and watch this video of one of the leaders in the field.

Design professionals are deeply involved in the most innovative work of the private sector around the world. It’s long past time to bring them into the design of government.

(3) Program design at the National Performance Review.

Early in the spring of 1993, I discussed program design with Vice President Gore during my tenure as a Deputy Director of his National Performance Review. He said that revisiting the design of Federal programs was too difficult, and that we didn’t have the time. I did manage to get a small team assigned to write Rethinking Program Design, one of the fourteen NPR Systems Reports.

The report’s four recommendations were:

1. Activate Program Design as a Formal Discipline

2. Establish Pilot Program Design Capabilities in One or Two Agencies

3. Encourage the Strengthening of Program Design in the Legislative Branch

4. Commission Program Design Courses

Addendum: The problem in a nutshell

The following quotation from The New Yorker (October 7, 1985, p. 30) on the demise of Manhattan’s Westway captures well the frustration that most Americans feel when confronted with the institutions of government that now exist:

“Deeply embedded somewhere in our thoughts or wishes — no doubt in some corner of the brain that is immune to factual evidence — is the belief that public institutions and programs exist to serve public needs. In this perhaps naive view, the links of the chain of public action begin with the identification of a need of society, by, say, a civic-minded group; next, measures to meet the need are debated and decided upon; and, finally, some institutional arrangement is made, and people set to work. It’s hard to help noticing, though, that in actuality things proceed, as often as not, in quite a different sequence. In this other sequence, the chain begins with the institutional arrangement, and only then is a need sought out — a need that will justify the continuing functioning of the institution. In other words, instead of starting with a problem and looking for an answer one starts with an answer and looks for the problem.”