In Theory…

Knisely’s Second Law states that “In theory, there’s no difference between theory and practice, but in practice there is!” My First Law states “Anything that exists is possible.”

Both laws are relevant to how Americans are treated by our legal system in the wake of the Supreme Court’s decision in the Gideon case, and why I’m ecstatic that Lawrence Tribe is joining Eric Holder’s Justice Department to work on improving legal access for the poor. But that’s just my personal opinion.

The post-Gideon history of legal representation for poor defendants is an excellent example of practice obliterating theory, and of America’s inability to establish – or at minimum observe – feedback loops from reality back to policy making. This is the design-of-government issue.

The Wikipedia entry for Gideon v. Wainwright will tell you more than you need to know about the case, starting with: “…the Supreme Court unanimously ruled that state courts are required under the Sixth Amendment of the Constitution to provide counsel in criminal cases for defendants who are unable to afford their own attorneys.” The ruling, in 1963, was unanimous. That was the law.

(BTW, googling “Gideon” will get you more than you want to know; the case is properly called a “landmark case.” There is even a famous book, Gideon’s Trumpet (1964). Now with a study guide!)

Ralph Temple delivered an address in honor of the 40th anniversary of the Gideon decision, in which he brought into clear focus its importance to a free people:

“Walter Van Tilburg Clark, in his book The Ox-Bow Incident, wrote: ‘True law, the code of justice, the essence of our sensations of right and wrong, is the conscience of society. . . . None of man’s temples, none of his religions, none of his weapons, his tools, his arts, his sciences, nothing else he has grown to, is so great a thing as his justice, his sense of justice. The true law . . . is the spirit of the moral nature of man. . . .’

“When we enter the court, we enter the temple of justice. And that’s why the Gideon case is so important. For the right to counsel is the most important of all rights, because without it none of the other rights can be protected. In this sense, the right to counsel is the key to the temple of justice.”

But of course the right to counsel means that either (a) lawyers must be found to work for free, or (b) someone must pay them, since the defendants cannot. It also must mean some minimal standards for attorney competence, or the right is meaningless (lawyers would say moot).

Georgetown University’s Law Center (my alma mater, class of ’72) held a symposium in 2003 marking the fortieth anniversary of Gideon. The website for the National Association of Criminal Defense Lawyers, one of the sponsors of the symposium, tells the story of Gideon in practice:

• The American Bar Association’s Standing Committee on Legal Aid and Indigent Defendants published a report, “Gideon Undone,” in 1982 (Gideon plus 19 years) clearly stating that criminal defense lawyers for the indigent were underpaid and overworked, whether public defenders, assigned counsel, or contract attorneys.

• A New York Times article in 2003 (Gideon plus 40 years) stated: “The recent spate of exonerations based on DNA tests has demonstrated that inadequate representation can, and does, lead to wrongful convictions. A Montana man, speaking at an Open Society Institute panel this month, told of spending 15 years in prison on a sexual assault charge after a trial in which his court-appointed lawyer did no investigation, hired no experts and failed to file an appeal. After 15 years, he was cleared with DNA evidence.”

(It is helpful that in 2001, the Fifth Circuit Court of Appeals finally decided that you couldn’t be put to death in Texas if your lawyer had slept through much of your trial. New York in 1996 and California in 1984 also came to recognize that sleeping lawyers weren’t much help.)

The larger problem, of course, is money. Few state or local governments have been willing, and fewer now are able, to pay for adequate numbers of public defenders or contract lawyers. And fewer private attorneys are willing to be assigned as counsel. This lack of resources is highlighted in Justice Denied, America’s Continuing Neglect of our Constitutional Right to Counsel:

“Because of insufficient funding, in much of the country, training, salaries, supervision, and staffing of public defender programs are unacceptable for a country that values the rule of law. Every day, the caseloads that defenders are asked to carry force lawyers to violate their oaths as members of the bar and their duties to clients as set forth in rules of professional conduct. In addition, private contract lawyers and attorneys assigned to cases for fees receive compensation that is usually not even sufficient to cover their overhead and that discourages their participation in defense systems. Equally disturbing, in most places across the country there is no oversight at all of the representation that these lawyers provide, and the quality of the work they provide suffers as a result.

“In addition, defendants throughout the country, especially in the lower criminal courts, are still convicted and imprisoned each year without any legal representation at all, or are “represented” by lawyers who have hundreds of other cases (thus violating rules of professional conduct), and lack the requisite expertise and sufficient support staff, including persons who can investigate their clients’ cases. Sometimes people who cannot afford an attorney sit in jail for weeks or months before being assigned an attorney; others do not meet or speak with their lawyers until the day of a court appearance. Too often the representation is perfunctory and so deficient as not to amount to representation at all.” (page x)

We’re not talking a lot of money here – not compared with redesigning the health care system, or even with Wall Street bonuses. There really should be enough money available to let our legal system hold its head high. There are so many countries where the legal system is a sham – it’s a shame that mine is also, in so many cases involving poor people.

And what does it say about our “respect for law,” and for our system of government? The Supreme Court can be quite explicit about what the legal system requires, and yet the other two branches ignore the Court’s direction, or just give it lip service?

I used to offer free advice (I still do…) to people who worked for me, and to my kids and their friends: “Just do what you say you’ll do,” I’d say. “Don’t promise the moon, don’t promise anything by COB yesterday, JUST DO WHAT YOU SAY YOU’LL DO. And you’ll go far.”

I want to hold my country to that same standard. For both theory and practice.

“The mood and temper of the public in regard to the treatment of crime and criminals is one of the most unfailing tests of any country. A calm, dispassionate recognition of the rights of the accused and even of the convicted criminal, …[and] the treatment of crime and the criminal mark and measure the stored-up strength of a nation, and are the sign and proof of the living virtue within it.”

– Winston Churchill

The Sword, the Clock, and the Mirror

I wrote this in 2001. I’ve gone through and updated some of the numbers and inserted a few hyperlinks. Significant changes are [noted].


The Sword, the Clock, and the Mirror

I. Government doesn’t learn

With the publication of The Fifth Discipline in 1990, Peter Senge of M.I.T. brought “learning organization” into the marketplace of ideas. After a decade, it’s clear that organizations can be taught to learn. In times of rapid change and increasing complexity, whole organizations, not leadership alone, must learn. This insight has escaped the Federal government.

Senge explains why: “As I began my doctoral work, I had little interest in business management. I felt that the solutions to the Big Issues lay in the public sector. But I began to meet business leaders who came to visit our MIT group to learn about systems thinking. These were thoughtful people, deeply aware of the inadequacies of prevailing ways of managing. They were engaged in building new types of organizations — decentralized, nonhierarchical organizations dedicated to the wellbeing and growth of employees as well as to success. Some had crafted radical corporate philosophies based on core values of freedom and responsibility. Others had developed innovative organization designs. All shared a commitment and a capacity to innovate that was lacking in the public sector. Gradually I came to realize why business is the locus of innovation in an open society: business has a freedom to experiment missing in the public sector and, often, in nonprofit organizations. It also has a clear ‘bottom line,’ so that experiments can be evaluated, at least in principle, by objective criteria.”

I read that and smacked myself: “That’s just plain Chuck Darwin: variation and selection. The private sector has variation and selection, while government only has variation!” If you don’t believe that the Feds have variation, go see the Catalog of Federal Domestic Assistance ( It lists 1454 separate programs [up to 2,016 as of 22Feb10] in 57 agencies [now 64]. If you believe that we have selection, go read Herbert Kaufman’s wonderful Are Government Organizations Immortal (Brookings, 1976). Or ask how much effort the National Performance Review invested in 1993 in killing off the helium reserve (needed to be sure we have enough blimps) and the mohair subsidy (for warm uniforms in WWI).

The key insight is that industry must learn in order to survive. Whether it’s called variation and selection, or evolution, or cybernetic feedback, or just learning, industry must do it. Government doesn’t have to. But like it or not, the Federal government is intimately interconnected with America’s future. We need the best thinking America’s got. So government absolutely must learn to learn, at the office level and at the Department level. The center of government – and the White House’s Office of Management and Budget (OMB) is all we have – must focus on teaching the government how to learn. Learning is government’s surrogate for “the bottom line.”

Here’s the challenge. With 1.847 million employees (not counting 861,000 in the Postal Service or 1.42 million in the uniformed military), the Federal government will always be the “Fortune One.” Next in line is General Motors, with only about 500,000 workers. There are 14 cabinet departments, 62 independent establishments and government corporations, 74 boards, commissions, and committees, and 4 quasi-official agencies in the Executive Branch. The Congress has over 58 employees for each Senator and Representative. [Figures not updated] Management “reformers” will never run out of work. Management “improvers” will never get government within sight of America’s private sector either, unless government can learn to learn.

Peter Drucker is credited with saying that leadership is “doing the right thing” while management is “doing things right.” (He doesn’t mention “doing the far right thing.”) Improving management is my concern here. As a civil servant I left improving leadership to the voters. Improving efficiency and effectiveness through better management – learning – was the key to American industry’s successes in the last century. It helped Japan between WWII and the late 1980’s.  It has not been government’s forte.

Consider the automobile industry. According to the Economist (January 13, 2001, p. 57): “During the oil crisis of the 1970s, Detroit was forced to ‘downsize’ its models to compete with more fuel-efficient imports. As the Japanese producers set up car plants in America and Europe, the battleground shifted to building cars that were more reliable and had fewer defects; it then moved on to manufacturing vehicles on “lean” production lines. What counts now is the flexibility to make different sorts of vehicle, especially on the same production line.” That’s mind-boggling. Management has had to learn continually, from Frederick Taylor’s time and motion studies to Toyota’s assembly line innovations to today. Variation and selection have been practiced ruthlessly. More of this story is told in The Machine that Changed the World. [And time has indeed marched on for Toyota.]

Variation and selection have been the hallmarks of our bricks and mortar industries. And they’ve been the bread and butter of the “new economy” companies. I don’t believe I have to make that case. We’ve been in the “selection” phase for over a year now. [Written during the last recession]

The study of how to improve management – of variation and selection, of organizational learning, of the evolution of management theory and practice – has been the business of business schools and their publications for decades. I used to read the Harvard Business Review and list the agencies that could use the insights I got from every issue. Now that the information revolution is in high gear, many of the articles are no longer relevant to government’s activities. The gap is too great. I believe that America’s most serious “digital divide” is between industry’s use of information and information technology and government’s failing attempts to catch up. [This claim may or may not still be true in 2010; opinion varies!]

An obvious conclusion is that improving management in government – teaching government to learn – requires no innovation. We just need to imitate. This is Knisely’s Sixth Law: “There are no new problems.” Government’s incentives only bring forth innovation in time of war.  There the list is long (see The Pursuit of Power, by William H. McNeill, 1982).

Even this cautious approach can backfire. In 1993, when we started Vice President Gore’s National Performance Review, we adopted management concepts from the late 1980’s. We said focus on customers, empower the front line workers, and downsize as much as possible, especially middle management. We set a target for downsizing, since that was measurable: a 12% reduction of civilian personnel in five years.

Even in 1993, however, the private sector was learning that downsizing didn’t lead always lead to increased profitability. “Chainsaw Al” Dunlap proved that at Sunbeam. Not surprisingly, no one at the White House or at NPR ever questioned the need to downsize the government. We had learned what to do from the private sector. We weren’t watching what happened next. They learned; we stopped learning.

The best-known case of over-downsizing is Governor (later Senator) George Allen’s 11% reduction in Virginia’s Department of Transportation (VDOT) in 1995. According to a report in the Washington Post (September 7th, 2000),  “Engineers in the critical division that designs road projects were managing 23 projects at a time – three times the average in the private sector – and some were juggling as many as 60.” Many projects were delayed for up to four years, while Virginia’s growth put 25% more cars on the road in the 1990’s. [This of course has gotten worse, not better, since 2001.]

Similar problems keep surfacing in the Federal government. “[T]he Pentagon never came up with an overarching framework for managing the downsizing. It now lacks adequate staff to get the work done in fields such as linguistics, acquisition, technology and scientific research and development,” according to Senate-House testimony reported in the Washington Post on March 30, 2001 (page B2). Between 1992 and 2000, the IRS workforce fell 17% while the number of tax returns climbed 13%, and the audit rate has dropped by two-thirds (Washington Post, February 16, 2001).  See Government Executive Magazine for more details on the legacy of NPR’s downsizing initiative.

And we still haven’t learned this lesson. The new [Bush 43] Administration’s first budget says “To shrink the distance between citizens and Cabinet members, the Administration will flatten the Federal hierarchy, reduce the number of layers in the upper echelon of Government, and use work force planning to help agencies  redistribute higher-level positions to front-line, service delivery positions that interact with citizens.” Been there, done that. Where were you?

Teaching the Federal government, and eventually the 73,000 smaller units of American government (, does not involve exhorting the workers to learn. It requires creating and maintaining mechanisms that will cause learning. Nerves in the fingers and memory do more to keep a child from a hot stove than all that parents say.

For example, Disney holds “trade fairs” for the people who operate their amusement rides. If you’re running a ride at a theme park, you don’t have much time to learn what the other rides are about. It’s sort of like being in a different agency in the same Department. So Disney gets the workers together to trade ideas about improving service, safety, or profitability. When one Disney World ride invented a machine that allows “guests” to avoid standing in line by picking up tickets for a pre-set time, the idea spread like wildfire. The “guests” avoid standing in line and the ride is busy more of the day. The presenters were rewarded both by Disney and by the imitation of their peers.

We could start with a trade fair of all the Feds who are engaged in compliance and enforcement activities. How inspectors try to achieve compliance with rules and regulations is quite similar. They’re all dealing with people. Such a trade fair should include EPA, OSHA, the IRS, FDA, USDA, and countless others. We could do it in person, and we could follow it up with an ongoing conversation on the Internet. Has this ever been done? Not to my knowledge. Would it be difficult? It might take a few phone calls from OMB.

We also need to look for the ways we discourage or forbid learning. In 1993, the National Performance Review was getting underway. Some of us met with Phil Lader, then the Deputy Director for Management at OMB. We were talking about customer service. Phil asked, “Well, why don’t you just ask them what they want?” I took out a newly minted CD from my home agency, DOT’s Bureau of Transportation Statistics. I showed him the user comment form inside the jewel case. It asked the usual questions: what the user liked about the CD’s data and what else the user wanted us to collect. It was just a postcard. He said, “That’s it, exactly!”

I told him to send me to jail. Any user comment form is a survey according to the Paperwork Reduction Act. I had bypassed Phil’s own Office of Information and Regulatory Affairs. I used the card without their formal review and written permission. Eight years later, nothing has changed there. [Now seventeen??] Where else are we still punishing learning behavior?

From here on I discuss other ways of incorporating learning into the Federal government. Needless to say, I think these are the most important ones. What’s really important is that everyone – workers, managers, executives, political appointees – should be encouraged and rewarded for inventing additional learning tools. Copying ain’t hard. America’s private sector has been showing us the way for 100 years. They’re still ahead.

And are “learning organizations” the last word? Of course not. The private sector is still evolving, and the Federal government must tag along. We need to avoid the fads, but we need to remember that management improvements in the private sector are layered over each other. The best corporations will learn to learn, and then learn something else. For a look at the current possibilities, one source is the Executive Book Summaries (

II. The Sword

Former Commerce Secretary William Daley took aim at an old Washington mega-problem, the structure of the Federal government (in “Reinvented but Still Redundant,” Washington Post, Op-Ed, March 12, 2001). He offered only another commission – a government solution if there ever was one. Yes, the structure of the government is a Gordian Knot. But no commission can forge a sword sharp enough to cut through government’s structure. The Congress and the lobbyists have been tying it up for decades. Our swords must be the innovations created by today’s private sector. Excellent solutions are available for the many mini-problems and inefficiencies that government still imposes on American citizens and businesses. Here’s how we can reinvent the grant, loan, and loan guarantee programs.

The financing programs at the Ex-IM Bank, Overseas Private Investment Corporation, Agriculture, and the Small Business Administration (and many others) share a common business model. Applicants who meet certain criteria may receive money – grants, loans, or loan guarantees. They may or may not have to repay it, but must be used for defined purposes. If access to these programs is improved, and if the programs are run efficiently, Secretary Daley’s goals of less confusing government and getting citizens what they need will be met. We don’t have to wait for a new Alexander.

The Front End. GSA’s FirstGov website [now] is the Federal government’s Internet “portal.” All government financing programs are listed there, or should be. Providing potential applicants with comparable information about their criteria and deadlines requires only formatting standards. Thirty years ago OMB started the Catalog of Federal Domestic Assistance with this simple idea. FirstGov can also link the citizen to an email address in each program. The next step, providing online application forms, isn’t rocket science either. NSF’s Fastlane program is a good model. Americans are rushing to pay their taxes and apply for student aid electronically. Used wisely, the Internet can inform, connect, and engage citizens with their governments.  We will have made a giant step forward when a citizen can go to one web portal, sort through all the possibilities, get questions answered promptly, and apply on line. We know how to do this now.

The Back End. In recent years the service units (print shops, personnel, IT support) in many agencies have started “bidding” for work from other agencies. Many such services have been contracted out. In the 1990’s the procurement shop in the Federal Railroad Agency was bidding to do contracting work for other DOT agencies, and we used them in the Bureau of Transportation Statistics. We need to go from bidding for printing and procurement to bidding for all “fulfillment” and “boiler room” activities (in private sector terms). We need to contract out as much “back end” work as possible. If SBA or a private company can demonstrate better, cheaper processing and collection shops, they should get the work. This just extends a process now underway.

The Middle. Government programs (and many agencies) have core business processes that need “enterprise solutions.” Few such business processes are unique, and most have private sector analogs. The Peace Corps recruits, trains, and supports volunteers around the world. So does my Marine Corps. The Consumer Product Safety Commission reviews products, as does the Food and Drug Administration. Every government “business process” has defined steps and identified decision points. In every case, the “enterprise solution” will be a large computer database tracking steps and decision points. All staff will use it interactively as they work. Not for reporting, but to do the work. Everything about a grant application, from regulations and background materials to applications, reviews and approvals and their deadlines, and eventually notification and award can be made instantly available on screen.

I was Deputy Chairman for Management at the National Endowment for the Arts in the early 1980s.  We gave money to a revolving fund at the New York Arts Council to give bridge loans to arts organizations between NEA notification of award and NEA funding. I never saw a clearer example of inefficiency in a government agency. I wrote up a plan for an early “enterprise solution.” I was determined to get the money out the door with the award. We cut over 40% of the time between decision and award and made it happen.

Very little of the elapsed time in any administrative process (whether getting a grant or buying a house) is spent doing the work; most is spent waiting, while paper goes from inbox to inbox. This idea is central to Re-Engineering the Corporation by Michael Hammer and James Champ (1993). In 1993 each disability claim processed by Social Security took 26 workers 155 days, with the actual work taking only 13 hours. If there was an appeal, 43 workers were involved over 739 days (NPR 1994 Status Report, Chapter 1).

If the application and the knowledge about processing steps are in the database system, and people work online, much of this waiting time will simply disappear. Re-engineering the work by assigning grant follow-up to “caseworkers” will eliminate more. Backlogs will be visible to managers and resources will be shifted there. The value added by each worker will – or won’t – show up in the system’s internal record keeping.

The agency can learn how to do its job better by studying these records. Agency managers can easily benchmark similar core business processes in other agencies and the private sector (see Political appointees can request reports on the system’s operations and benchmarking studies. And staffing – up or down – can be done based on the objective needs of the system in operation, rather than political goals (well, I can hope).

The “policy issues” around the purposes of the grant programs and which applicant gets how much money won’t be affected by focusing the front end, standardizing the back end, and creating enterprise solutions. And so the constituencies and Congressmen won’t be offended. None of this will be easy, and it won’t happen by late tomorrow. But we can start any day.

The new Administration’s first budget proposes a down payment of $10m on a $100m program for e-government (budget). This would be a great place to start. And we can save Secretary Daley’s doomsday base-closing commission business model for problems we can’t solve with a little money and a little muscle from the White House and OMB.

III. The Clock

Project Management. Some years ago the staff attorneys at Justice were outraged when management suggested that they track their hours by case. Why, it would take all of the fun out of doing the Lord’s Work in government if they had to account for their time in 12-minute increments. Even thought this is what they do when they leave to go for a Washington law firm.

In March of 2001 the Defense Department discovered it employed more contractors than civil servants. If you ask a contractor on either side of the Potomac if their firm uses project tracking software (Microsoft Project is only one example), they’ll be surprised and say, “Of course – how else can we keep track of who’s working on what, how they’re doing, when products are due, and how much each really cost us? How on earth would we know how much to bid on the next contract if we didn’t review how we did on this one?” If you ask a civil servant, they too will be surprised: “Why no, what good would that do?” Same problem as the lawyers at Justice.

But if you’re not tracking how you’re doing and trying to improve, you’re living in the present, with no yesterday and no tomorrow in mind. Dogs are like that.  People and organizations need to learn.

Project tracking came into its own during the space race, when time and complexity were combining to frustrate NASA’s efforts. Time was crucial: President Kennedy had promised a man on the moon in a decade.  In 1972 I had a friend at HUD, Bert Greenglass. Bert had been in charge of a NASA tracking chart that was two stories high and 100 yards long, in a hanger at Cape Kennedy. He kept a staff busy moving 3×5 cards around showing who was doing what, and highlighting what was behind schedule. Computers make it easier.

One of the Big Things in the Federal government now is “performance measurement.” Executives are encouraged to commit their staffs to performance goals and “hold their people accountable.” Most have almost no idea whether they have too few or too many staff and other resources to meet these goals. It’s no wonder, since few have been running projects with firm deadlines and limited resources. So if they can find additional resources, they throw them at the projects to meet the deadline.

Of course, some have been assigned projects with far too few resources. They can’t object, since there’s no track record of what can be accomplished by which people. And of course, a lot of these projects are done on time. But could they have been done cheaper, better, or sooner? The contractor community has a saying: “You want this done cheap, fast, and good? Choose any two.” They’ve learned.

This is not new stuff, of course. See or In fact, there’s a Project Management Institute ( and George Washington University offers a master’s degree. No need to invent it: just copy and apply.

To be explicit, I’m saying that every product or project requiring resources above a very low threshold should be formally monitored using project tracking software. The reports should be available to all, preferably posted. Senior managers should use the reports as agendas for staff meetings.

I tried this, and found it so contrary to the prevailing culture that I gave up. It won’t happen until political appointees arrive with instructions from the White House to demand it, because the staff in Cabinet Affairs will be reviewing the reports and taking names. And pigs will fly…

Master Calendars. There are many processes in the Federal government that don’t have real endpoints, so no one will think of tracking them as projects. Budget building, for example, is now continual. Learning is still needed here.  It’s still very possible. It’s not done.

When I was Deputy Assistant Secretary for Budget and Programs at Transportation, I did manage a meeting between our leadership and the leadership of the agency budget shops (FAA, FHWA, etc.) to discuss how that year’s budget cycle went, but I never was able to establish a Master Calendar in the Departmental budget shop. Here’s what I mean by Master Calendar and why it’s a good learning tool.

The budget process is like a ping-pong game. First we propose, then you respond, then we respond to your response, and so on. It’s played at every level, until finally the President proposes and the Congress disposes. The more cycles the better, since each one brings both sides closer to agreement. Each internal cycle costs staff time and runs the clock. We need to strike a balance: how many resources vs. how much better information and agreement.

The controlling budget office (OMB over the departments, the departments over the agencies) should announce in advance when they expect to deliver their products and when they expect responses. That would alert their client offices to the year’s workload. They never do. The same controlling offices should track planned vs. actual delivery dates for both what they send out and what they receive, and then hold a meeting to discuss how to improve the process for the next year. That’s unimaginable. But the learning opportunities are obvious.

The OMB Director should demand a Master Calendar (done in project tracking software) and hold monthly internal meetings to discuss planned vs. actual events. Then annually, the Director and his senior staff should meet with Departmental Assistant Secretaries and discuss how the process should be changed for the following cycle. The Assistant Secretaries might even take the idea home and hold meetings with the agency budget heads. Everyone would learn a lot. Time will magically appear on the calendar. It’s not brain surgery. It’s just a focus on learning.

IV. The Mirror

There’s no substitute for looking at yourself critically. Look in the mirror; if you don’t like what you see, change something. In government, if you don’t look, you’ll go from pretty to ugly very quickly. You won’t even know it. This is the most important lesson about being a “learning organization.”

Program Evaluation. Of course, the best way for the government to learn about itself is to find out what effects its actions are having “out there.” This used to be called “program evaluation.” When [then-] Treasury Secretary Paul O’Neill was Deputy Director of OMB in the Johnson Administration, it was very popular. I worked for the Office of Economic Opportunity about this time. I was lucky enough to work in OEO’s Community Action Program Monitoring Office. We went around the country trying to figure out what difference the CAP program made. It was an inexact science at best, but I did come back with a few hard facts, some anecdotal evidence, and lots of humility about the effects of our programs.

It wasn’t perfect. I visited the CAP in Wirt County, WV, in 1968.  The director had half a dozen “neighborhood centers” strewn around several counties. He was spending a lot of time filling out forms from headquarters. So much time, in fact, that the marks he made on paper obviously came from a moving pencil: dash, dash, dash. I put in my report an observation that we were keeping our directors too busy with reporting. The next thing I knew, headquarters had sent out a form asking how much time CAP directors were spending filling out forms! I was struck dumb!

In the mid-1970s, program evaluation was still valued. I started the first departmental Office of Program Evaluation at Commerce, and after the first year OMB recommended doubling my professional staff. Those were the days.

Program evaluation has become a lot less popular since then. In part it was killed off by the creation of the Inspectors General, or “IGs.” IGs are charged with ferreting out “fraud, waste, and abuse” and evaluating programs. When the IGs were created, they simply took all the program auditors into the IG shops. Ferreting out “fraud, waste, and abuse” gets them headlines and more staff.

Program evaluation just gets them reports about the design and operation of programs that neither the Congress nor the Administration wants to hear. Guess where their priorities are. Guess where the program auditors went. Guess how many program heads ask the IG for help with program evaluation. Since the IGs now have the responsibility for it, guess how many program heads are pushing program evaluation.

This is not news. Paul Light, in Monitoring Government, said that the Inspectors General since 1978 have tended to focus on compliance, especially because of the short term rewards from the Congress and the press for finding “the guilty.” As he said, “The IGs would be more valuable to their agencies and Congress if they focused less on short-term statistical accomplishments, particularly those involving investigations, and more on program design emerging from outcome-oriented evaluations and inspections.” (p. 194) And “[compliance accountability] also may distract Congress and the president from addressing the hard questions about how to design institutions and programs to work better from the beginning.” (p. 230)

In 1996 DOT’s IG sent me a draft report blaming my Office of Airline Information (OAI) for taking at face value the airlines’ reports on late arrivals and lost baggage, saying the office should be sending out program auditors. On the merits, I agreed. Accepting unaudited compliance reports is like sending lettuce by rabbit.  But when OAI came to DOT from the Civil Aeronautics Board (CAB), the IG had grabbed the program audit team of 17 and promised to continue the audits. He had ‘forgotten’ the promise. Now I was being blamed for not doing them!

I sent the IG a memo on November 1st, 1996 including the following: “Should the priorities of the Office of the Inspector General prevent it from providing audit support with regard to air carrier submissions it is incumbent upon the IG to honor the commitment of the prior IG to continue this audit effort by restoring to the BTS aviation program the resources of the Bureau Carrier Accounts and Audits (BCAA) that it received from the Civil Aeronautics Board based on that commitment.  These resources included two GM15, five GM14, and a number of lesser graded support staff.”

Translated, that says “If you try to zap me for not auditing the airlines, I’ll zap you for (a) stealing the staff and (b) refusing to honor your commitment to do the auditing yourself.” The final IG report didn’t mention the unaudited reports, so I won. I didn’t get the audit staff back, so the IG won. The only losers were the Secretary of Transportation and the FAA Administrator, who continued to make decisions based on unverified compliance data from the airlines. Oh yes, and the public lost, too.

When I was the Director of Analysis for the Office of Federal Student Aid in the Department of Education, I went to the San Francisco Regional Office to mosey around, kick the tires, and ask dumb questions. That’s how it worked best 30 years earlier at OEO.  I saw (and copied) an Associated Press story about how poor students at the Los Angeles community colleges weren’t being told about student grants and loans. The story included some numbers, and a quote from a dean to the effect that the schools weren’t there to baby the students. If they didn’t take the time to find out about the financial resources available to them, it wasn’t the schools’ fault.

Finding and funding poor students seems like FSA’s’s core business. I farmed the copies around the senior staff. I wanted to take a “tiger team” representing all parts of FSA to Los Angeles. I wanted to find out what was happening and whether it might be part of a larger pattern. Not one  person was interested in the issue. No one had anyone to spare to go and ask questions. Everyone was immersed in their own crises du jour. Several people said the story was impossible, since informing students was indeed the responsibility of the financial aid offices. This was a violation of Knisely’s First Law, which states, “Anything that exists is possible.” I learned that lesson in 1968, doing monitoring visits in Louisiana’s “river rat country” right after Martin Luther King was shot.

FSA was doing an award-winning job of measuring customer satisfaction among schools, the banks that fund student loans, and the students that receive them. They all loved us. Of course, we weren’t asking the students that hadn’t applied since they didn’t know we existed. And we didn’t know they existed, either.

For a while FAA was asking airline pilots to grade FAA’s performance. FAA considered the pilots to be their customers. It would have been too hard to ask the traveling public, and besides the answers might not have been polite. Talk about sending lettuce by rabbit!

Surely all this must have changed by now, since the whole government is focused on the Government Performance and Results Act and “performance measurement?” Not really. According to a report published by the [now] IBM Center for the Business of Government in January, 2001, “Using Evaluation to Support Performance Management,” “[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. Few interviewees felt comfortable about their agency’s capacity for evaluation.”

In the Conclusions and Recommendations, the authors state “However, in general, evaluation capacity within these agencies is not adequate to contribute effectively to performance measurement and management. Most agencies do not allocate funding to support an adequate number of evaluation staff, nor provide the training needed for either the analytical staff already there or the line managers who are being asked to cope with the new analytical challenges.”

There’s still room for improvement.

“We’re Unique!” If there’s one common theme I heard across seven cabinet departments and seven agencies and task forces, it was “We’re unique. No one else has problems like ours.” It drove me crazy. Of course, most federal employees don’t change agencies much during their careers, and like the Disney amusement ride staff, don’t get around to visit other programs and agencies very often. For years, I’d arrive at a new job with a bag of tricks I’d collected at other agencies, amaze the populace with those tricks, and add a few more to the bag before moving on.

In June of 2000 I met with about two hundred other pooh-bahs at the Kennedy Center to discuss the civil service in the new century. The imminent departure of half the Senior Executive Service by 2005 troubled us. To no one’s surprise, we agonized over where the government would get senior executives as good as we were. We worried about recruiting and training generally. I stood up and said that the private sector also had a retiring Baby Boom problem, and that with a hot economy I bet everyone is having trouble recruiting. There’s even a catch phrase for it: “The War for Talent.” I said we should go and see what they were doing about both problems. I said it several times. Everyone said that the government was unique, and the private sector experience wasn’t relevant. Looking for solutions in the private sector wasn’t in the final report. Another missed opportunity.

Being unique keeps federal employees from trying benchmarking. Benchmarking how others do the same work has led to increased efficiencies all across the private sector.

Being unique has led to 177 different loan guarantee programs, and 158 workplace training programs, and a lot of the redundancy in the Catalog of Federal Domestic Assistance.

Being unique has led to about 400 different job series as coded by the Office of Personnel Management (OPM). Hard to cut down on personnel staff when there are 400 different job titles and a “skill library” containing 318 skills and their definitions [updated].

Being unique led FSA to create its own bank to show that below market student loans would hold down interest rates charged by the Citibanks of this world. I guess they never heard of Federal Credit Unions, first chartered in 1909.

V. Summary

My late father told me early and often that some folks learn from the experience of others. Some folks learn from their own experience. Some folks never learn. I started telling that to my two kids when they were very young.

• Believing that you’re unique is a mindset that guarantees you won’t learn from the experience of others.

• Not using Master Calendars and not doing program evaluation and project tracking means not learning from your own experience.

• Not being threatened with going out of business means never learning.

The Federal government can and must do better. The private sector has shown us the way.  No “rocket surgery” is involved. Just do it!


Robert A. Knisely retired in 2000 after 30 years as a civil servant. During that time he worked at Navy, the Office of Economic Opportunity, the Departments of Health, Education, and Welfare and Housing and Urban Development, various agencies as part of the response to the first Oil Embargo, and President Ford’s Clemency Board. He became a senior manager in 1975. He was in the Senior Executive Service at the Departments of Commerce and Energy, the Consumer Product Safety Commission, ACTION, the National Endowment for the Arts, the Departments of Transportation and Education, and Vice President Gore’s National Performance Review. He also spent two years in a Navy “think tank” and five years in the semi-private sector in Washington.

The Din of Inequity

“The Din of Inequity” – a presentation at my Harvard 40th Reunion.

About a week or so before my Harvard 40th Reunion in September, 2002, I got a request from Harvard to participate in a symposium on the first day. The title was to be “The Influence of Money — Personal and Corporate — on Washington Politics,” to be held in Sever 113.

I agreed to do the seminar, but I thought that this was an “old millennium” topic, given the past year’s headlines about Enron, WorldCom, Arthur Andersen, and the like [little did I know!]. I emailed Bob Bennett, the classmate who was to moderate the panel, and he said that the session would be more open-ended and I could bring up whatever I liked. All four panelists met for breakfast before the event, and we each had different favorite subjects.

When it was my turn, I said that corruption in Washington was “so last millennium,” and I wanted to talk about a topic of more concern to Harvard graduates of our generation. I reminded the group that many of us had taken Raphael Demos’ course, Philosophy 1 (A & B), and had of course fancied ourselves philosopher kings, or at least good candidates, ever since.

The two problems worthy of philosopher kings, I said, were global warming and the increasing inequality in American society. These were the two mentioned by a classmate, Peter Barnes, in our 40th Anniversary Report, and I agreed with him. I didn’t want to talk about global warming, so I’d address the growth of inequality. I wanted to make several points (here somewhat expanded):

  1. Inequality has been growing in America for some years. In 2001, for the first time, households in the top quintile (20%) earned more than half the nation’s income before taxes. The top 5%, with incomes above $150,000, earned 22.4% of the national income, up from 22.1 in 2000. (1) This is no longer news. From the New York Times: “For at least the past 15 years it has been hard to deny the evidence for growing inequality in the United States. Census data clearly show a rising share of income going to the top 20 percent of families, and within that top 20 percent to the top 5 percent, with a declining share going to families in the middle…. And other evidence makes it clear not only that inequality is increasing but that the action gets bigger the closer you get to the top. That is, it’s not simply that the top 20 percent of families have had bigger percentage gains than families near the middle: the top 5 percent have done better than the next 15, the top 1 percent better than the next 4, and so on up to Bill Gates.” (2)
  1. History shows that greater  economic inequality is correlated with less political stability. Good examples here are Sweden and Switzerland on the one hand, and Argentina and Zimbabwe at the other. Does the French Revolution ring a bell?

  1. Societies with less economic inequality are healthier. “Health in the Americas” (2002 Edition) from the Pan American Health Organization (3) makes this point. As quoted in the Washington Post (4), “The epidemiologists at PAHO found something that had been noticed elsewhere – namely that wide disparity in income in a population is a hazard to health. High-income, wide-gap Brazil has a lower life expectancy (68 years) and higher infant mortality (38 per 1000) than low-income, narrow-gap Peru (70 years and 37 per 1000).” And according to the New York Times article cited earlier: “Canadians can expect to live about two years longer than Americans. In fact, life expectancy in the U.S. is well below that in Canada, Japan and every major nation in Western Europe. On average, we can expect lives a bit shorter than those of Greeks, a bit longer than those of Portuguese. Male life expectancy is lower in the U.S. than it is in Costa Rica… A few months ago the conservative cyberpundit Glenn Reynolds made a splash when he pointed out that Sweden’s G.D.P. per capita is roughly comparable with that of Mississippi — see, those foolish believers in the welfare state have impoverished themselves! Presumably he assumed that this means that the typical Swede is as poor as the typical resident of Mississippi, and therefore much worse off than the typical American. But life expectancy in Sweden is about three years higher than that of the U.S. Infant mortality is half the U.S. level, and less than a third the rate in Mississippi. Functional illiteracy is much less common than in the U.S.”
  1. Rising inequality in America should be of concern to Liberals and Conservatives alike. Liberals will be concerned that median family income declined 2.2% to $42,228 in one year, and that Black household income is down to $29,500, while the percentage of Americans living below the poverty line rose to 11.7 in 2001. But conservatives should be concerned as well, because a stable American society is good for everyone, and (at least) equally good for the grandchildren and great grandchildren of the conservatives. And we would expect true conservatives to take this longer view.
  1. We know how our economy creates inequality. The workings of America’s economic system are no longer a mystery to us; there is precious little “terra incognita.” I discussed briefly James K. Galbraith’s book “Created Unequal.”  In the book, Galbraith argues (convincingly, to me) that we now know the “levers” on American capitalism that determine our levels of domestic inequality.
  1. What are the “levers” that create inequality? Galbraith argues that governmental decisions are the main forces affecting inequality. He includes the reliance on monetary policy to battle inflation, the redistribution of the tax burden, hostility to trade unions, and an indifference to preserving the real value of the minimum wage (see page 20).
  1. What is our intellectual obligation? If indeed increasing economic inequality challenges America’s political stability and physical health, and if we know what government actions are the “levers” controlling economic inequality, then we have an intellectual obligation to set those levers consciously, as a matter of public policy. And where we should set them should be a matter of public debate.
  1. What is our moral obligation? Clearly, we have a moral obligation to set the “levers” so as to minimize the suffering of our fellow citizens, consistent with other goals. For example, if Galbraith is right in asserting that the 1990’s proved that the NAIRU, (the natural, or non-accelerating inflation rate of unemployment) either doesn’t exist or is several percentage points below the level that results from the Fed’s interest rate manipulations, then a LOT of unemployed workers (and their families) have been suffering needlessly (see (5), pp 172 – 182). I myself would argue that many of us could spare the cash to alleviate a great deal more suffering. That’s part of the public debate, of course.
  1. The Role of the Harvard Class of 1962. In the stable society we all want, a public debate about inequality and any resulting changes in the positions of the “levers,” will only come about because the “haves” decide that their longterm costs outweigh their short term benefits from societal inequality. Individuals in groups such the Harvard Class of 1962, — in other words, the “haves” – who favor long term societal stability as well as fundamental “fairness” for all American citizens, should take the lead in moving the debate — and in moving the levers.

That’s what philosopher kings do.



(1) See also Money Income in the United States: 2001, Census Report P60-218, available at This report was summarized in the Washington Post on September 25, 2002: “U.S. Poverty Rate Rises, Income Drops; Increase in Ranks of Poor is First in 8 Years” (Steven Pearlstein; page A-3).

(2) For the full article, see Paul Krugman’s “For Richer” in the New York Times Magazine, October 20, 2002.

(3) Available at

(4) “Americas’ Life Expectancy Rises, but Health Mixed,” (David Brown) September 23rd, p A-2

November 8th, 2002

Plus ça change…

Every five years, Harvard offers its alumni an opportunity to write “a little something” addressed to the other members of his class and see it published. (And yes, by now alumnae and “her” class are included, but I graduated in 1962.) In 1992 the Thirtieth Anniversary Report from our class included the following two paragraphs from me:

“I was invited to join the [Presidential] transition team in 1988, and went to work for Sam Skinner at DOT, first as Deputy Assistant Secretary for Budget and now as the drug czar. We require 25,000 employers to test five million transportation employees for illegal drugs. Happy to discuss that with anyone who calls. Last year I cobbled together an 800 number, voice mail, fax service, and modem access to a drug regulations data base housed in Oklahoma. To our knowledge, this is now the best high-speed, broadband communications tool linking government with a regulated community. We did the systems work in seven months, a minor miracle in the government ADP circles. Call 1-800-CAL-DRUG and give it a go.

“Last year I also completed work on a new homeless support program, collecting $1.75 million from federal departments, writing a request for applications, and awarding grants to New York City, Baltimore, and San Francisco. This was sort of a ‘hat trick;’ there are advantages to becoming one of the ‘old boys.’ One program goal is exploration of ‘services integration,’ attempting to get the various parts of government to treat the various aspects of the homeless person as one problem set. The absence of ‘services integration’ accounts for much of the wasted effort in social service delivery today.”

During the past eighteen years we’ve seen quite an improvement in “communications tools linking government with regulated communities,” as well as with everyone else!

We have NOT seen as much improvement in what we then called “services integration,” and now travels under many other names. We talk about “stovepipes” on both coasts, and “silos” in the fly-over zone. We’re so stuck in our own stovepipes/silos that we can’t even agree on a single term of art!

And the design community talks of “jumping the fence” rather than “thinking outside the box.” I rather like “jumping the fence,” myself.

But will I become persona au gratin (that is, kinda cheesy) if I don’t keep saying “thinking outside the box?” That’s got me worried…