Walter McClure, Chair of the Center for Policy Design

Change incentives in major social systems to ensure that self-interest serves the public interest

A Civic Caucus Review of Minnesota's Public Policy Process Interview

April 26, 2016


This is a report of a Civic Caucus interview between Dan Loritz, chair, Civic Caucus; Paul Gilje, executive director, and Walter McClure, chair of the Center for Policy Design. McClure advocates major restructuring of the nation's big social systems, including education, health care, criminal justice, and welfare. An essential element in restructuring, McClure contends, is to develop built-in incentives for individuals and organizations to serve the larger public interest while at the same time those individuals and organizations pursue their respective self-interests.


The Civic Caucus is conducting several interviews the past months on how Minnesota public policy issues are raised, shaped, discussed, and resolved. McClure recently authored a paper titled: Architecting Large Social Systems in which he advances what he calls "a more systematic approach to policy design." It elaborates on the points made in this interview and can be found on the Center's website .


Walter McClure received a BA in philosophy and physics from Yale in 1959 and a PhD in theoretical physics from Florida State in 1967. His dissertation research, on nuclear cluster theory, was performed at the University of Tübingen in Germany.

In 1969 he switched from physics to health care reform policy for reasons, he says, having to do with "relevance". He worked at InterStudy under Paul Ellwood's leadership from 1969 to 1981, at which time he left to start the Center for Policy Design. He directed the Center until his retirement for medical reasons in 1990. He still serves as board chair for the Center of Policy Design. At InterStudy he worked with colleagues on the HMO strategy for health care reform, among other tasks drafting much of the federal legislation.

At the Center for Policy Design he has developed a general theory of why organizations do what they do, and a set of methods to strategically redirect their behavior toward the goals society desires of them. With these methods he and his colleagues at the Center developed a health care system reform strategy to get better care for less, and developed a National Health Insurance proposal consonant with this strategy. He assisted Medicare, Pennsylvania and Cleveland to implement the first step of the strategy, severity-adjusted outcomes assessment of providers, before his reluctant retirement for medical reasons. He has recently become active again.


The Civic Caucus is reviewing how well community organizations are preparing specific, innovative proposals to address critical public policy problems. Among questions we've been raising:

* Do too many proposals seem to be addressing symptoms instead of underlying causes?

* Are proposals actionable, not just expressions of a problem that needs to be solved?

* Are the most critical questions being addressed?

Our interview today with Walter McClure highlights additional dimensions:

* The need for self-correcting change, in which responsible individuals and organizations have built-in incentives to follow self-interest to produce results in the public interest.

* Concentrate on the "big stuff", changing large social systems, not just achieving incremental improvement.

Need to restructure big social systems . At the heart of Walt McClure's thinking is a conviction that intentional major restructuring is needed in the nation's large social systems, including education, health care, criminal justice, and welfare. He says we must "architect", or redesign, our large systems that are chronically under-performing. States and the nation certainly are devoting significant attention and money to these systems, but, he contends, performance falls far below public expectations.

Some benefit results from improvements here and there. Most current efforts at improving these large systems' performance fall into what McClure refers to as "continuous system improvement", although at times he refers to some of these less charitably as "omnibus tinkering". Nevertheless, he doesn't want to diminish the significance of looking everywhere to raise system performance by any possible improvement, big or small.

But organizations can be trapped from making substantial change by the larger system in which they operate. The common deficiency in the improvement approach is that individual organizations operate within a larger system. For example, providers operate within the larger health care system, schools within the larger education system, etc., and these larger systems ("macrosystems" as he terms them) are not neutral or passive, as too many people, including policymakers, think and as a result simply overlook them.

In fact, the structure of these larger systems places powerful incentives on the organizations within them, which incentives that actually largely determine their behavior, and the organizations have little or no choice about it, McClure says. Those that follow the incentives prosper, those that act against them suffer, and will fail if they persist. If the incentives of a macrosystem align with the goals that society desires of that system, all is well and good: organizations will prosper by chasing those goals. But if a macrosystem's incentives for its organizations are anti-aligned with the goals society desires, then big trouble results: organizations who pursue the goals desired by society suffer and perish, while organizations that perform against society's goals, as the anti-aligned incentives reward, prosper.

For good or ill, since an organization cannot change the larger macrosystem by itself, it must follow the incentives placed on it by that macrosystem or it isn't around long. The moral of this story for policy: If the incentives of a macrosystem align with goals, great: make sure that that macrosystem's structure is well-maintained. But if the incentives are anti-aligned, bad news: policy must fundamentally redesign, or "architect" that macrosystem so the incentives become correctly aligned.

So don't heap blame on the people and organizations in a malperforming macrosystem themselves . "There are many brilliant, competent and highly motivated people in all macrosystems whether they perform well or poorly - in the education system, in the health care system, etc. The difference is not the people, it is the incentives placed on them by the macrosystem structure they operate in, that is, what performance it rewards and what it punishes.

"You cannot change human nature, but you can change the incentives of a macrosystem and then the same people and their organizations will act differently," McClure contends.

Therefore, change the incentives . If a macrosystem rewards its organizations for performing as society wishes and punishes them when they stray, "then we have a well-performing system," McClure says.

No brow-beating is needed if the incentives are aligned correctly. McClure contends that organizations acting on their own volition under proper incentives will function far better and with greater innovation than policy outsiders could ever accomplish with orders.

Needed: system architects to design new structures . McClure would like to see policy designers become versed in "Large System Architecture". Large system architecture has two essential components: (1) analyzing why organizations do what they do, and (2) if they are not performing as society wishes, designing and executing policy strategies to alter their behavior.

To help others understand what he's talking about, McClure cites the auto industry and the computer industry. Those industries, he said, are able to keep giving people products that are better for less cost. "Does anyone think this is due to the virtue, altruism and purity of auto executives? Or computer executives? ...If a car company can't make a good car for the money, they are not around very long...The same is true of the computer industry: no matter your motives, if you can't make a better computer for less, you aren't around long," McClure says.

But if you have a malperforming macrosystem, Large System Architecture is simply a disciplined, systematic way to identify its misaligned incentives and redesign that macrosystem's faulty structure to replace them with sound, well-aligned incentives.

Four principal tasks for the large system architect. Under McClure's approach, the large system architect has four tasks to arrive at a redesign strategy for a poorly performing macrosystem:

  1. Identifying societal performance goals for the system,
  2. Comparing actual performance against goals,
  3. Determining the underlying causes of system problems: the incentives that drive actual performance and the structural elements that generate these incentives, and
  4. Designing a system structure that aligns incentives with goals.

Implementation, not just coming up with a sound design, is the other large part of the large system architect's job.
McClure's paper , Architecting Large Social Systems , has a more detailed discussion of both redesign and implementation methods but the following are some points on implementation he mentioned in the interview.

The preliminary step to implementation is to have the redesign in hand. If you want to get there, you must first know where you want to go. A redesign is simply a plan or future model that details the proposed new structure for a problem macrosystem that will align its incentives with goals. Also as implementation action proceeds, the redesign plan helps show whether things are moving toward or away from the desired new structure, and allows course corrections where possible.

Unbalancing the status quo is the first big step. McClure stresses that once the redesign is in hand, it does no good to work on "backlogs" to unstick the "log jam" that is the status quo. "We must find the 'front log' and move that one...the step or action most likely to unbalance the status quo holding the macrosystem in its present form, and produce the most response in the direction of the redesign."

Persuasion is important. To move the "front log", the architect must identify individuals and groups with the power and motive to take the needed action, and then persuade them to do so. A lot of educational work, diplomacy and consulting assistance is needed to illustrate the advantages of action and the disadvantages of inaction, he said. If the action is taken, the situation shifts and a new "front log" comes to the fore, which must subsequently be identified along with the parties who now have the power and possible interest to take the action to move this next "front log". And so it goes, one "front log" after the next. After action is taken, shifts in the macrosystem structure can be unpredictable and some move the redesign forward, some back. The architect runs after the fluid situation trying to choose "logs" that herd the macrosystem toward the redesign model.

Leadership might be more important than legislation. Often a coalition of public and private leadership can produce better, faster, more agile progress, he said, than trying to change the law. But, he cautioned, "in some cases legislation, if done well, can help or be crucial."

Successful change also requires keeping ahead on the battle of words. Architects and supporters of change must be prepared to counter opposition with "frequent and skillful informative rhetoric", McClure said, because special interests will spend considerable effort and money on propaganda and disinformation to muddy the waters and twist the strategy to their own advantage not the public's.

Start small and build larger . "Wherever possible, establish limited test beds first," McClure advised, "to assess and refine a proposed redesign model, and then gradually implement it more widely, always monitoring and refining it along the way for emerging problems as you scale up."

For those who wish to promote it, a Lesson on Redesign from Machiavelli. All policy analysts, all private and public policymakers, and especially funders, whether foundations or public agencies wishing to promote redesign, McClure suggests, may want to reflect long and hard on Machiavelli's observation:
"There is nothing more difficult to take in hand, more perilous to conduct, or more uncertain in its success, than to take the lead in the introduction of a new order of things. For the reformer has enemies in all those who profit by the old order, and only lukewarm defenders in all those who would profit by the new order, this lukewarmness arising partly from fear of their adversaries, who have the laws in their favour, and partly from the incredulity of mankind, who do not truly believe in anything new until they have had actual experience of it."
The Prince
, 1532 AD

To those who wish to promote good redesign: the key step is to encourage professional large system architect groups who know how to do it . Large system redesign has never been done successfully by legislators or citizen groups. It requires too much specialized expertise and far more time in research and development than legislators and citizens can devote. One needs a profession of policy analysts and researchers, he says, who specialize in study and development of redesign for each problem macrosystem. These experts do not replace continuous system improvement efforts; they are a separate parallel effort working on a longer-term, permanent redesign solution.

Note these experts only propose, he says; they do not decide, policymakers decide. Just like a house, he says; the architect proposes, the client decides. Such an approach was used on the moonshot, he said; Congress didn't try to design a rocket in committee, but instead created specialists, NASA, and gave them the time and steady funding to do the design and implementation under its oversight. In the same way one needs to create large system architect groups.

The locally based Center for Policy Design represents one prototype for large system redesign. The Center for Policy Design , a Minnesota-based non-partisan policy analysis and design group, of which McClure himself is chair, is a prototype large system architecture group, he said. The Center for Policy Design has two main projects, public education system redesign, led principally by Ted Kolderie, and health care system redesign, led principally by McClure. These are two problem macrosystems, the biggest items in state budgets, with faulty structure and incentives which desperately need good redesign. The Center's redesign strategies illustrate how redesigning macrosystem structure can engender new incentives well aligned with goals:

  • Public education redesign: An important part of the proposed education strategy, he said, has been to broaden the public school system to include chartered public schools as well as district public schools, to open wide the public system to new innovative schools and the incentive of competition for students impossible under the old district monopoly structure. The Center has many tools to assist charterers and organizers of such schools, he said. The object is not charter schools, it is good schools both district and charter, and letting poor public schools, district or charter, close for lack of students.
  • Health care redesign: The proposed health care strategy, he said, emphasizes informed consumer choice. "The perverse incentives on providers in the present system reward costliness independent of quality. By providing consumers new objective information and incentives, unavailable and unobtainable in the present system, the redesign enables and rewards them to choose providers who offer better care for less cost (the desired goal) at the expense of providers who are costly or poor quality, creating powerful incentives on providers to be better for less or lose patients."

A successful large system architect group capable of redesign requires: (a) the right kind of professional team, and (b) the right kind of funding. Let's start with the right kind of team, McClure says. The key ingredient is an experienced large system architect; they are rare because they do not think just incrementally; they think "outside the box". "Without this kind of lead thinker, you just have just another conventional study group thinking useful incremental system improvement but not knowing how to think fundamental redesign. If you have a chronically poorly performing system whose underlying structure and incentives are anti-aligned with goals, incremental thinking won't do the job."

Larding incremental improvements - like new technology, better training, command regulation, token carrots and sticks, etc. - on top of strong, underlying faulty incentives, which are actually the drivers of the bad performance, will not alter those incentives. It may give you some worthwhile short-term gains with great effort, but you won't solve the problem until you correct those incentives by fundamental redesign, and for that you need a system architect thinker.

This architect needs to find and bring on an experienced veteran of the present system who knows how it really works from the inside, invaluable knowledge for developing the new design. He also needs to bring on a pro with political savvy and tact, invaluable traits for implementation. And these leaders will need support staff.

But it starts with the right large system architect. Without that thinking, redesign doesn't occur. You find such persons by their track records: their ideas (do they seem both innovative and practical?), their commitment to staying the course (redesign and its implementation take at least a decade or two; are they still at it?), and above all, their results (have they made worthy innovative change happen?).

A successful large system architect group needs steady core support . Successful system change can take years, "with not always much to show for it for some time," McClure said "Kolderie and I have been at it more than two decades." As Machiavelli observed, redesign has no private constituency willing to pay for it. Consulting clients have immediate needs for their organization, but not for their macrosystem which they can't change by themselves; none will pay for the long term study and analysis required to come up with a redesign for a problem macrosystem.

Therefore support must come, if it is to come, from public-interest funders: foundations, or government agencies, or endowment of the architect group by private donors committed to the public interest generally or to improving the particular large system that the architect group is working on. Financial support for redesign must recognize the time requirements of redesign (recall Machiavelli): tangible design study and progress may take some years, full implementation some decades. Short term grants, even large ones, by foundations, agencies or concerned individuals are just not helpful if the group does not have adequate core support: meaning adequate assured, steady long-term funding that will stay the course. A large grant uncertain of renewal in a year or two is useless, McClure says; you can't bring on staff and train them up and then have to let them go because your funder is off to the next big thing; that is an utter and costly waste of time and effort. (The alternative is off-mission consulting to support your staff, which utterly destroys the mission with irrelevant marketing and project work, he says.) The group needs steady annual funding of a size the funder can sustain and guarantee in five to ten year increments.

This kind of core support will support the initial study and design work that no one else will pay for, and additionally the unpaid educational work of implementation (seeking parties who can be interested to help move each "front log" as it comes to the fore). If parties can be interested in a particular front log, then core grants can often be significantly augmented with on-mission consulting and project contracts for assistance to those parties at least for that front log. But then funding is needed to interest parties for the next front log, who may be the same or different, so there is usually a hiatus in project funding between one front log and the next. In short, without core support, the group simply cannot stay on-mission.

Foundations could be the secret weapon and key to better policy design, not only for continuous system improvement but particularly for system redesign. Support from government agencies tends to be slow, inflexible, and unreliable. Private foundations offer the most promising source because they can be agile, imaginative and in it for the long haul...if they choose. McClure suggests the Twin Cities and Minnesota have a particularly able, active foundation community. He would like to see them work together more, think beyond the conventional, think more broadly and strategically, aware not only of their own activities but each other's and aware particularly of which of the community's and state's macrosystems need what kind of help. He would like them to be aware of who is targeting gifting for the various kinds of help needed and suggests that someone in our foundation community ought specifically take on this task of routinely gathering and publicizing this strategic update. He would like foundations to be aware of the distinction between public consumption gifting and public investment gifting, making sure there is plenty of the latter, and aware of the difference between short-term investment gifting and riskier long-term investment gifting.

The most assured agile core support would be endowments. But the whole concept of large system redesign is not well-known, and therefore is unfamiliar to most potential private or public donors. As Machiavelli observed, it takes a donor of uncommon vision to see the potential — great pay-off but uncertain success — of an innovative approach with which most people have not had experience.

Redesign is long-term and risky, as Machiavelli observed, impossible on short-term grants, not guaranteed of success in any given length of time. However, McClure believes foundations and donors should take that risk because redesign offers the only way to achieve substantial lasting success with problem macrosystems: it is the only approach that addresses correcting the underlying anti-aligned incentives that actually drive these systems' malperformance. He believes the more architect groups at work on a problem system and the longer they work on it, the more likely a successful redesign can emerge and be implemented.

"Redesign is the one approach that addresses the diagnosis rather than the symptoms", he says. If it eventually succeeds, redesign offers the biggest bang for the foundation or donor buck, returning much more to public well-being and prosperity than it ever cost in investment gifting. A couple of foundations or donors might want to team up to provide core support to a promising redesign group working on a large system that is one of their gifting targets.

Redesign is a long-term investment by the foundation community that should proceed in parallel with other shorter-term efforts. Redesign should proceed in parallel (what Kolderie calls the split-screen approach) with efforts to ameliorate symptoms because there is nothing else in the meantime. But symptom curing alone can't address the diagnosis and end the symptoms. Foundations should see that both efforts are proceeding apace.

Foundations and donors wanting to support redesign: bet on the architect, not the proposal . McClure says the key to successful redesign is an excellent large system architect. Funders should seek out such individuals by their track record until they find one the funder feels confidence in. "If the track record is there, then that individual is worth betting on for the long haul", McClure advises. Proposals mean little because with experience about what is working and what isn't, they all need modified and more new proposal writing, a great waste of scarce skilled time; the good architect will have plenty of ideas how to best alter course if something isn't working. Funders shouldn't wait for an architect to come to them, but go out and look for them and ask how can they help, McClure suggests. Then steadily work with them to supply what they need to build and maintain a proper group. Simply keep posted on effort and progress; stop funding only if and when effort and progress no longer appear effective or promising.

Comment here on this interview with Walter McClure