Eric, I think we are talking about two different things here. You give a list of underlying procurement principles, but that is not what I understand this service contract issue to be about. I don't think there is any, or very much, argument about underlying contract principles. The argument I've seen is about how a long term service contract is structured to meet almost certain stresses and changes in service demands without breaking those contracting principles.

How, for example, does on in 2000 build into a computer maintenance service contract running into 2004 a mechanism to meet some predictable, but undescribable technology, operational or threat change. Write too much "wiggle room" in and you have valid criticism that you've given the contractor way too much rope. Write a tightly structured tasking trying to cover every base and suddenly find you simply cannot accomodate some situation without a huge flap because the PO and CO have made a neat box requiring major restructuring to meet a pressing unexpected need. I'm arguing that there are techniques and processes for judicious identification of probable stress points and placement of relief valves in the contract at inception that provide flex without license. It does perhaps require a different sort of planning.

In simplistic terms, I think they can avoid the incident I once heard about where the housekeeping contract was drawn too tightly to allow intensive rest room clean-up during a big conference with many times normal load and longer hours of use. Looked bad for the organization, but they had so tightly bound the thing that getting extra work hours and extended time windows was going to require a change order. Not one of your underlying principles was violated, in fact, they may have been over exercised.Zeal for overly precise specification and controls apparently led to an unsightly mess.

That is one of the places POs and COs can also come into conflict. Beautiful, tightly drawn written contract, lousy result perhaps as a result of being too beautiful and too tight in the abstract without ability to meet the demands of the actual.

Ramon

By Eric Ottinger on Friday, March 3, 2000 - 12:17 pm:

Ramon,

I took an undergraduate course from a professor who had done some interesting statistical analysis of Supreme Court decisions. He had concluded that there were only a half dozen underlying issues that would decide most cases that reach the Supreme Court. His statistical analysis would not tell you what the vote would be for any particular case, but it would predict very accurately how the justices would line up.

I don’t think there are more than a half dozen principles underlying Federal Procurement.

Arms-Length dealing.
Public Accountability.
Don’t spend any money that Congress hasn’t appropriated
Competition is Good.
Prices should be fair and supported by objective facts or data.
Etc. (Pick your own.)

If you understand those principles and organize your knowledge around those principles, it isn’t hard for the experienced professional to find the knowledge he/she needs and make fairly accurate predictions about the way things will play out in any particular scenario. It isn’t rocket science and it doesn’t require a Philadelphia lawyer.

Folks who think otherwise want to trade on their “expertise.” Or, they want to bluff their customers with their unique and esoteric knowledge. Or, they have an agenda. (Personal Opinion.)

Each agency has its own set of bureaucratic hurdles and gates; purportedly, to make sure that the worker bees don’t do anything really stupid or embarrassing. Also, to make sure that the boss doesn’t look stupid when his/her boss asks a question. Most of the working knowledge that we have is knowledge of how to operate within a specific bureaucratic environment. Much of this is an intuitive understanding of what the real agendas are behind the formal agendas.

(There are some very complex and specialized areas of expertise in procurement, e.g. transportation, quality assurance, CAS, Part 25, SBA regulations. We don’t normally have to deal with more than one or a few of these in any particular job and we usually have specialists to guide us.)

(Also, when things go wrong, we wander into the domain of the Lawyers. As several have pointed out, don’t go there unless absolutely necessary.)

(When I say, "Go see a smart lawyer," that means that, in my opinion, we have wandered away from the Water Cooler into the swamp.)

Getting back to the matter at hand. It is a common sense priciple that we should know what we are buying and be able to count, measure or evaluate it in some fashion. The principle is simple. The implementation can be difficult when we are not buying simple goods and services.

Eric

Eric, this touches on something we've discussed before in connection with credentials vs. experience. A quote from J. A. Proude, "Experience teaches slowly, and at the cost of mistakes" applies. Some never learn from mistakes, gaining uninformed and useless "experience." When learning takes place we have valuable and sometimes unquantifiable experience with deep knowledge.

That old saw about the consultant gets that across. The man with the non operative engine who called in the expert who finally tapped a spot with a hammer complained that he could have paid someone $5 to do that instead of $1,000. The expert replied that in that case the hammer tap was $5, the $995 was knowing where to tap.

It is true that this is often not laid out in the usual "rational" form. No drawings, papers or proposals. It is still there. Maybe the other old saying about someone having "forgotten" more than most know is a way of putting this. We all have knowledge we've "forgotten" that influences decisions. It is not necessarily irrelevant. It sometimes is vital. It is difficult to extract and examine and is sometimes explained away as an expert's "gut feeling," but it is not uneducated speculation and guesswork. Inexperienced participants, who may have nothing more than that as foundation for their opinion, will often make such accusations. I expect we've all seen meetings degenerate where foolish courses of action were pressed by a clueless majority over an experienced minority on the basis that nobody's opinion counts more than anyone else's. That has been responsible for some of history's great disasters and uncounted small fiascoes.

An interesting story in connection with the search for the USS SCORPION may show a method of capturing this in better form. I believe there is a paper about this somewhere, but I've forgotten where so I'll have to go on memory. The experts were swirling when someone who had studied this tapping of forgotten and gut knowledge applied a means of extracting it. He proposed a betting game (bottles of Scotch I believe) in which the experts would place bets against various possibilities influencing actual location. If I recall, some fruitful advances were made because high bets were placed against some. None of the experts could quite explain in detail, demonstrate with equations, or write a paper on why -- they were tapping deep experience they had "forgotten."

I believe one of the web site cites (Ah! There is a fun one with that grating misuse of "site," "cite," and "sight" we've seen so much of lately.) I gave earlier touched on tapping such expert knowledge of a group to shed light into the shadows of bounded rationality. Techniques exist for doing this in an organized way. Formalized risk assessment uses some and I believe other disciplines have developed others. For multi million dollar contracts it would seem worth applying this to formally delineate areas of the unexpected and shed as much light as possible on probabilities within. Unfortunately, except perhaps in some high tech and R&D work, this does not seem to be common. It could prove a sound investment of effort.

Ramon

By Eric Ottinger on Tuesday, February 29, 2000 - 05:31 pm:

Beverly,

Good input! But.

It isn’t that I question the use of performance measures in service contracts. I question whether there is really a significant impact when you look at the big picture. I honestly don’t know what proportion of the service contract dollars go for food service contracts and such, and what proportion goes for highly technical SETAs and such.

It is an axiom in the education field that all experiments succeed. I am glad that you think that this is a success, but you will need to be more specific to be convincing.

I was about to ask you how you applied performance based principles to general scope advisory and assistance contracts; then you said the secret word: "TASK" (which are not necessarily FASA Task Orders).

The problem with applying merely quantitative measures to highly professional work is that unquantifiable qualitative issues may be much more important than the quantifiable measures.

For instance your HMO may think your doctor is whiz because he processes each patient in less than 10 minutes.

If however, you are the patient and he misses your cancer, which is not diagnosed until six months in the future, you may have a very different attitude toward that doctor.

(Note, that a careful reading of the regulation indicates that you have satisfied the “performance based” when you apply it to the extent which is practicable, which may not be much.)

I am not opposed to quantifiable measures. (That is motherhood and apple pie, and I am sure Vern would agree.) But, the more difficult and more professional the work, the more the emphases has to be on judgmental, qualitative measures of performance.

I think most of the skepticism was directed at the detailed and comprehensive SOW which must describe in detail what the contractor is doing today and describe what he/she will be doing five years in the future, at the same level of detail. (As long as we have some freedom to fill in those details using tasks, there is hope.)

Hoping this clarifies,

Eric

By Eric Ottinger on Tuesday, February 29, 2000 - 05:13 pm:

Ramon,

We all think we are rational. I would advise a young person that arguments in the form “I am more rational than you are,” are usually not very productive.

I was thinking of an “old school” doctor who could make a very accurate diagnosis based on a little bit of poking, smelling and listening. Also I was thinking of businessmen who can eyeball a project and quickly make a fairly accurate estimate of what it will cost. Of course, this kind of intuition reflects years of study and experience.

Infantry officers can usually say, “That is 300 meters,” or, “That is a kilometer,” with adequate precision. If you asked how they know, there is no answer other than practice. The brain does a lot of things very well in the back, where we aren’t conscious of what the brain is doing.

If we are going to distinguish rational from reasonable, rational has to equate to logic.

I am very skeptical about the use of formal logic in practical affairs. Usually, when a person tells you how logical he is, he has started from some shaky assumption and proceeded through some process of rigorous logic to an utterly crank conclusion.

The best thing that we can all do for rationality is to be very careful in testing our assumptions.

(Otherwise, you have made some good points as usual.)

Eric

By Vern Edwards on Tuesday, February 29, 2000 - 12:25 pm:

Beverly:

I'm sorry that you are puzzled about my opinions about long term, complex service contracts. My views are neither new nor controversial outside government policy circles. I am not an original thinker in this regard.

I have stated my reasoning and the sources of my thinking. I have also published several articles in which I have explained my doubts about PBC policy. If you have read what I've written and the sources that I've mentioned and still don't understand how I have reached my conclusions, then I don't know what else to say and I have nothing more to add.

With regard to your personal experiences with PBC, I can only say that I do not know the details, but that I'm glad that you are pleased with the results that you have had.

I am very surprised at the "nay-sayers" regarding Performance Based Service Contracting! It seems everyone can intellectualize why it won't work, but the pilot program has proved it does. As Quality Assurance Program Coordinator, I have actively participated in the re-writes of numerous contracts to the Performance Based concept. This initial re-write is generally more time-consuming than writing the former way, because we have to re-think the approach instead of copying the old way. However, we anticipate that the time spent will be well worth it. The flexibility of the Performance Based concept lends itself to all the services we have used it for, including base operational services such as food service and refuse SOWs; Range Operation and Maintenance SOWs; engineering and technical services; Advisory and Assistance Services; etc. I really am puzzled by Vern Edwards remarks regarding the agencies not being able to specify the results they want before contract award. Surely, they must have a specific result or outcome that can be identified at least in a general manner. We have a general outcome specified on contracts for which we further define specific outcomes for tasks under each task order awarded under that contract.

By Ramon on Monday, February 28, 2000 - 10:59 am:

Eric, strange, one of my dictionaries defines "rational" as: "Having reason or the faculty of reasoning; endowed with reason; agreeable to reason; pertaining to or acting in conformity to reason; sane; judicious" and another has similar language. I can only conclude your "rational" person reaching "unreasonable" results must suffer a form of multiple personality disorder.

I will not deny that in popular culture strict application of rationality often acquires the label of "unreasonable." That only proves that quite often reason and fact can't overcome desire and prejudice, i.e., "Officer, it is unreasonable to deny me the privilege of speeding down the shoulder for an important meeting."

I certainly do not agree that "Reasonable decisions are often based on the weight of many years of experience and intuitive good judgment without much supporting detail." I'd put that into the sheer luck category. Such decisions are reached by years of experience gained through actual practice and absorbing experience of others (education) and being able to sift the meaningful details from the meaningless wild cards. That detail may not be written or even organized for that "intuitive good judgment," but it is most certainly there. It is probably what we really mean by "experience." Anything else is a crap shoot.

There are plenty of people with vast experience, no interest in learning, no ability or desire to think about what happened and who have no memory of the meaningful detail who reach "reasonable" conclusions (in popular terms) that are disasters. The average casino gambling addict might be an example. Their "experienced" and "reasonable" assumption that it is "time for their number to come up" is poppycock. Yet many, without a clue as to real probability theory, agree with them. The same might be said for some contracting types who, instead of quarters, pull the lever with the same old contracting token under the assumption that one day they will hit.

It is not abstract theory that imposing an overly detailed tasking in the attempt to preordain the outcome of a dynamic service environment is playing against house odds. Experience, from case after case, are that you will encounter some situation that crashes into the tasking detail. Systematic recognition of the fact you don't and can't know every eventuality and application of rational, not blind, experience to the broad areas can allow you to begin fencing the areas you do not know well enough to predefine. Then more experience will allow setting some limits to the range of uncertainty (risk assessment operates much in this way) and could begin driving tasking that would allow mitigation as the events unfold.

This is and it isn't rocket science. They do it, but so does every commuter or traveler setting out with unknown weather and traffic conditions. The recognize that things happen, they know an accident is more likely than a Martian attack, and make rational adjustments to accommodate the probable, but precisely unknown obsticles.

Some similar thought in forming the contract journey can lead to similar escape valves. One example that I think was worth its weight in gold in developmental contracts was a funded task for ad-hoc studies to be executed upon approval and precise definition of the COR/COTR. I heard this was strongly opposed in some circles as a slush fund. I suppose it could be used that way, but used properly it was the means of meeting technical surprises by allowing immediate exploration and development of the issues related to a technical problem, otherwise requiring a change order, that both buys progress and rational support for the potential change. In more than one case it allowed development of an alternate to the immediately assumed solution that could be done cheaper, faster and without contract modification.

I wonder if something similar could not work into service hours to meet unexpected situations. For example, a computer maintenance contract that is being stressed by increased incidence of service loss due to virus attacks. The contract provides for funded ad-hoc additional hours and services within the already tailored broad tasking. Additional service in removal and recovery can be immediately turned on while metrics are gathered on the additional levels required long term. Time has been bought while the need for and a rational level of additional service in the new environment is explored.

By Vern Edwards on Wednesday, February 23, 2000 - 12:49 pm:

Eric:

I didn't say anything about pragmatism. The term was "bureaucratic pragmatist," which is your term. I think it's wonderfully descriptive, and I intend to use it a lot from now on. (I don't know what William James would have thought of it.) In any event, I didn't intend to use it disparagingly, just descriptively.

By Eric Ottinger on Wednesday, February 23, 2000 - 12:32 pm:

Vern,

Last I heard, “Pragmatism” was the philosophy of the distinguished American philosopher and academic, William James. I have no idea why you should consider the word a pejorative.

You are correct in saying that researchers always have preconceptions, if only for the purpose of framing a hypothesis to be tested. I meant that the researcher should not be burdened with either the preconceptions of our community or the preconceptions of our critics.

I think a lot (but not all) of the research that is done in our field is half-baked stuff that is often crippled by the simplistic preconceptions and hypotheses which shape the “research.”

(i.e. Surveys which take the form—

“Do you feel more reformed than you did a year ago? Choose a number between 1 and 5.”)

(There is a well-known story about the young sociologist who wanted to study the Capone gang. He found that these low-life, thug killers sat around boring each other silly telling the same war stories over and over again. All the researcher had to do to make a friend was to sit down next to a gangster on a barstool and be a willing audience for the old war stories. Somebody with that kind of chutzpah and initiative is the person I have in mind.)

The researcher should be young because the project that I have in mind requires a lot of stamina. Also, it is easier sometimes for a young person to ask really dumb, obvious questions and get honest answers.

Eric

By Mike Love on Wednesday, February 23, 2000 - 11:29 am:

Research is needed - given. Little or none is now being funded - given. Whether adequate funding will ever materialize is unknown. (I hope some lurker out there has research funding and will call.) The Federal Management Association is proposing a blue ribbon panel on the optimum size and shape of the federal government. Until then, life goes on.

To the person trying to rationally contract for services that cannot be defined over the long term of the contract, what alternatives does the system provide?

· Performance based contracting with fixed pricing preferred.
· Service contracts describing what should be done.
· Cost reimbursement, level of effort.
· Time and materials from the GSA Schedule or other vehicle.
· IDIQ with each increment of work defined in the task order. (If the TO does not describe the work, it is effectively a level of effort or T&M contract.)
· Employment contracts.

Assuming that this person does not just want to label a contract "performance based" and then implement it as circumstances dictate, are the existing criteria adequate to choose between the above? IMHO the answer is no. Can criteria be identified that would guide the choice better based on what we know today? My guess is yes and the criteria might already have been identified in some of the works discussed already. But what changes if any are necessary to allow better service contracting in the federal government? I suspect that policy, both contracting and personnel, needs substantial revision, both acquisition and program managers will need training and stakeholders must buy in.

By Vern Edwards on Tuesday, February 22, 2000 - 08:39 pm:

Eric:

You ask: "What I would like to know is-- Why do we, as a tribe, stick to a theory that is so frequently contradicted by our experience."

Answer: Because the members of the tribe think of themselves as "bureaucratic pragmatists," or because they haven't thought about their experience and won't read books by academics.

When the Government first promulgated the PBSC policy that now appears in FAR Part 37, a very high level policy official called me at home and asked me not to be too critical of the new rules in my GWU classes and in my writings. This person said that she/he was familiar with the writings of Macneil, Macauley, Williamson and others, and understood them, but that those ideas were too complicated for the working level people to understand, much less take into account in their contracting practices. She/he felt that the new PBSC policy was better than nothing, and in any event it was needed to demonstrate the adminstration's commitment to acquisition reform. See, Eric, bureaucratic pragmatism. Oh, well, everyone to his own taste.

To the extent that people go along with questionable new policies without question, it is either because they don't know any better (maybe they haven't read Simon and Macneil, or maybe they disdain academics) or because they're too busy being reasonable to be rational. Or maybe they're just too bored or tired to care.

P.S. There is no such thing as a researcher with no preconceptions. Researchers do research to find out whether their preconceptions -- read, hypotheses -- are valid or not.

P.P.S. Why does the researcher have to be young?

By Eric Ottinger on Tuesday, February 22, 2000 - 05:16 pm:

Chacun a son gout, Vern.

I don’t disagree. I believe I made the same argument in an earlier thread. I mentioned a considerable amount of “private side commentary” to that effect.

At that time, I had specifically in mind a popular book of advice to businessmen and a casual conversation with a lawyer acquaintance.

If “bounded rationality” will get some of our thinkers off the dime, more power to it.

The low rent formulation is “Stuff Happens.”

What I would like to know is-- Why do we, as a tribe, stick to a theory that is so frequently contradicted by our experience.

Regarding academia, I think it would be a very good thing if some young researcher, with no preconceptions, would take a look at our service contracts. I think he/she would find a lot of interesting disconnects between theory and on-the-ground reality.

If most 1102s were inclined to speculative thought, this chat room would be so busy that you and I would have a hard time getting a word in edgewise. Most of our colleagues seem to be bureaucratic pragmatists; I try to frame my arguments accordingly.

If I say, “It is difficult to do weather forecasts for the Washington DC area; most of the time, when we have a foot of snow the forecasters are caught by surprise,” there is general agreement. If I start talking about the “butterfly effect” and “chaos theory,” eyes glaze over and my colleagues start edging away.

I think “bounded rationality” is relevant to our service contracts in the same way that the “butterfly effect” is relevant to weather forecasting. It is an interesting topic for a bull session. For most of my colleagues it is sufficient to know that the weather is unpredictable.

Ramon,

There is a distinction to be made between “rational” and “reasonable.” Rational people frequently reach unreasonable results because they don’t check their assumptions against reality.

Reasonable decisions are often based on the weight of many years of experience and intuitive good judgment without much supporting detail.

(I think Bertrand Russell always made a point of demonstrating to his wives that he was more rational than they were. He went through a lot of wives that way. I try not to make that kind of mistake.)

If we are going to be purveyors of sound business judgement in this new acquisition reformed era, we need to think more like practical businessmen and less like theorists.

Anyway, you are right. I don’t really disagree. The perfect and perfectly detailed SOW is frequently a chimera.

I think we will have more luck persuading our colleagues if we put more emphasis on concrete experience and less emphasis on abstract theory. But this is just my person opinion. If the academic approach works, that’s fine with me.

Eric

By Vern Edwards on Tuesday, February 22, 2000 - 02:53 pm:

Mike:

On February 22, you wrote:

"On the substance of this thread, given our current state of knowledge is the policy mandating performance based contracting appropriate? My reading of the discussion so far is no. The need to define and redefine the work as it progresses will often practically preclude even defining desired end states. Am I correct?"

You correctly understand my view. PBC is based on the idea that agencies can fully specify the results they want ex ante, i.e., before contract award. While this may be possible for simple (one or two task), short-term service requirements, it is difficult if not impossible for complex (multi-function), long-term requirements.

As O.E. Williamson put it, "Comprehensive contracting is not a realistic organizational alternative when provision for bounded rationality is made."

Contracts for complex, long-term requirements must allow the parties to define requirements on an ad hoc basis, as they emerge, without having to resort to the often cumbersome and time-consuming change order/supplemental agreement process.

Vern

By Ramon on Tuesday, February 22, 2000 - 09:37 am:

The necessity to take rational action when all the facts cannot be assembled and subjected to methodical analysis for fully rational action is at least part of what bounded rationality seems to be about. Particularly in the human relations field the necessary raw material may simply not exist as it might in engineering. However, this should not be an excuse for random jumping to conclusions and rushing into thoughtless action simply because it seems to advance toward some assumed goal.

There are ways to assess those ill defined areas without being driven by baseless assumptions. I suspect, from experience, we too often take a vote rather than do an assessment.

I doubt the comment about academics reflects Eric's basic views and to some extent it has some basis in the publish or perish world. Unfortunately, it can be a view held widely enough to shout down well thought out and documented observations. Good research and facts too often get submerged in "I (or we) want this."

By Mike Love on Tuesday, February 22, 2000 - 08:06 am:

Let me chime in. Rigor of analysis and in gathering and examining facts is always welcome. It could even be equated with being professional.

But it is often necessary and even more often advisable to take action before a complete analysis is done. At least one attribute of a good business or other leader is to be able to act with some wisdom on incomplete information.

In the public acquisition field we will often be faced with issues that need resolution, or at least action, before the analysis is complete. No one would dispute that. Likewise no one should dispute that one should take advantage of existing scholarly analysis when it exists. Thus in practice, I would hope that those who tend to favor either scholarly approaches or going off half cocked would appreciate it when someone reminds them with some humor (even if it is less funny than one would hope) that there is another view.

On the substance of this thread, given our current state of knowledge is the policy mandating performance based contracting appropriate? My reading of the discussion so far is no. The need to define and redefine the work as it progresses will often practically preclude even defining desired end states. Am I correct?

Mike

By Ramon on Sunday, February 20, 2000 - 11:34 am:

I'm suspect Eric meant no harm in his observation and sometimes it does seem the less clever researchers pick safe and well plowed fields, but there seems too much easy dismissal of real study and research into such things as contracting. It is so "commonsense" that only training, not study is seen as needed by many.

A quote attributed to one Henry Shaw seems to apply: "The trouble with people is not that they don't know but that they know so much that ain't so." That applies to all of us at various times, but it can apply too long to whole communities.

One of my pet examples was the formal design review that was probably formalized as a logical working mechanism for monster programs and then became an article of faith. It was common knowledge among some that gathering a hundred or so people in dark suits to view slides and hear presentations was absolutely necessary for milestone decisions. Then a study identified the near religious ceremonial gathering as being well up on the ladder of wasteful practices. Still, many "knew" it had to be.

When we tried to change with a more practical and realistic review down in the plant in jeans looking at actual work it was heresy. Presentation of the study result, official that it was, was discounted as being theory. Powerful groups outside the program management "knew" the full dress ceremony was absolutely necessary. Unfortunately, facts rarely change views held as sacred.

We need more of what Vern mentions. Not less. Then we need some accountability to make sure it is recognized in practice.

By Vern Edwards on Saturday, February 19, 2000 - 08:23 pm:

I'm sure that it's fun to ridicule sociologists studying what seems obvious to the common man or woman, but academics and independent scholars do good service when they confirm or disprove intuitions, commonsense notions, and popular beliefs. First, they make the experiences of a few accessible to all by reporting what others have observed and by explaining its relevance and significance to those in diverse fields of endeavour. Second, by studying and thinking about phenomena in systematic ways, classifying those phenomena, giving them names like "bounded rationality," and publishing their findings, they organize and disseminate knowledge and open the door to further study that deepens our understanding and increases our ability to put what we know to effective use.

By Ramon on Friday, February 18, 2000 - 10:51 am:

"Robbed of the ability to communicate once action was joined, they sought to overcome the obstacles and accidents that would inevitably arise in the unfolding of battle by ever more elaborate anticipation and predisposition. Plans were drawn which laid down minute-by-minute manoeuvre by the infantry and almost yard-by-yard concentration of artillery fire, in an attempt not so much to determine as to predestine the outcome. The attempt was, of course, vain." [John Keegan, "The First World War"]

Where that "article of faith" Eric mentions is in force we have the equivalent of military "experts" who never heard of von Clausewitz and the "fog of war." They send their troops into a meat grinder and stalemate. At least in contracting it is just ulcers and frustration.

The First World War generals were in an iron technology bind where the weapons were available to force stalemate and horrible casualties, but not for breaking the lock. I don't think law and FAR put contract people in quite such an iron box. Unwillingness to think things through and put in the work too often results in a fall back to "you can't do that" without exploration of a solution within the legal framework.

Immediate recognition of the fog of contractual execution at the inception stage can lead to much more rational approaches of risk assessment and mitigation, concentration on doing well those things that can be nailed down and building in accommodating mechanisms for those that cannot. Perhaps among the most is communication. It does expand the opportunity for uncoordinated and even illegal contract modifications, but systematic and controlled use within contractual structures is vital. Non-recognition of the need to create mechanisms to rapidly communicate, assess the dynamic conditions and make course corrections in a contract seem to me to lead to a similar, if less fatal, stalemate and carnage.

The things we've been mentioning are, by definition, unknowns. Still, it is known they can and do happen. Likely "species" can be identified that might lurk in any particular contract jungle by tapping corporate memory (where it can be found) on both first hand and second hand experience. From both disaster and success in those experiences coping strategies can be developed and worked into the contractual framework.

It is probably more difficult than polishing the documents endlessly in search of perfection, it is messier and probably gives less confidence as it admits there will be unknown problems. It also probably works better than the false confidence gained by presumed near perfection. It probably significantly lessens "casualties."

By Eric Ottinger on Thursday, February 17, 2000 - 12:52 pm:

All,

It isn't that I disagree with Vern, but I find this somewhat reminiscent of an old joke about academics to the effect that a sociologist is a person who will spend $100,000 (probably a million in current dollars) to find out something that he could determine much more easily if he just asked.

Our friends in construction know all about bounded rationality. Every time that we hand the contractor the perfect set of blueprints, the contractor knows from long experience that he is going to get well on changes.

We never learn from this experience. It is an article of faith that the perfect blueprint (specification, SOW, etc.) is out there in the ozone somewhere, and readily available if we could just motivate our lazy experts to work harder.

Requirements change due to actions by third parties and higher levels. My experience, (perhaps unique to my agency, but I doubt it) is that all plans get perturbated and redirected frequently with the redirection coming from numerous levels, up to and including, Congress.

In this light, the perfect SOW requires both a comprehensive listing of everything we know about the requirement today, and 20-20 foresight, several years into the future.

If we are honest and realistic, we will find a way to write our service contracts with enough flexibility to accommodate such additional direction and redirection without too much friction.

Eric

By Gene Simmons on Wednesday, February 16, 2000 - 04:48 pm:

Looking for a good sample of a solicitation using PBSC for a FFP services/material contract. Need a draft of QA and surveillance plan, deduction / incentive schedule. Need asap for upcoming requirement

Thanks in advance for your time and assistance

By Ramon on Wednesday, February 16, 2000 - 01:49 pm:

Excluding the bad faith aspects of "cheat and lie" is there any interest in opening a thread dealing with contractual and management mechanisms for dealing with this very real world situation?

We don't want to get into what we'd like to do with cheaters and liars, though various types of termination come to mind.

I expect Vern and others have some approaches that will help in planning for the uncertain, building in coping mechanisms and then managing the result.

By Vern Edwards on Wednesday, February 16, 2000 - 01:08 pm:

The Economic Institutions of Capitalism: Firms, Markets, and Relational Contracting, by Oliver E. Williamson (New York: The Free Press, 1985) is a very good (but not easy) book that ties together the ideas of bounded rationality, incomplete contracts, and transaction costs. The book is available in paperback and you can generally find it in the economics section of a Borders or Barnes and Nobles bookstore. Of special interest to Open Forum readers are the following chapters: 1. Transaction Cost Economics, 2. Contractual Man, and 3. The Governance of Contractual Relations.

Here is a tantalizing and representative quote from Chapter 3, page 70:

"Not every transaction fits comfortably into the classical contracting scheme. In particular, for long-term contracts executed under conditions of uncertainty complete presentiation is apt to be prohibitively costly if not impossible. Problems of several kinds arise. First, not all future contingencies for which adaptations are required can be anticipated at the outset. Second, the appropriate adaptations will not be evident for many contingencies until the circumstances materialize. Third, except as changes in states of the world are unambiguous, hard contracting between autonomous parties may well give rise to veridical disputes when state-contingent claims are made. In a world where (at least some) parties are inclined to be opportunistic, whose representations are to be believed?"

In plain English: Detailed, complete planning and specification of required results for long-term contracts in environments subject to frequent and/or rapid change are likely to be impossible or excessively costly for the following reasons: First, the contracting parties can't always identify all requirements very far in advance. Second, even when they can identify a contingent (might not happen) requirement, they can't always decide in advance what the desired result (performance) should be. Third, unless the facts are absolutely clear, if the parties end up in conflict about their obligations one or both may be inclined to cheat or lie.

This is a very good book, but as you can see it is not an easy read. Even so, the effort will be worthwhile for those professionals who want to develop a deeper understanding of the nature and difficulties of service contracting.

By Ramon on Wednesday, February 16, 2000 - 12:00 pm:

The web is a wonderful resource, but no substitute for real libraries and books now as it tend to be illustrative of that "mile wide and an inch deep" saying. Still, a quick search gave a number of references for bounded rationality. Not surprisingly, many are in the financial world where such uncertainty abounds and exact science can't quite reach.

A very basic definition is at http://risk.ifci.ch/00010586.htm where the parent site is devoted to risk in the financial world and might contain items of interest to contracting people.

The first bullet of a slide at http://www.cba.uh.edu/~jaana/class/c336907/sld009.htm struck me as I was doing this quick search. I wonder how many times in contract formulation that first available choice moving toward the perceived goal has led down the path of disaster.

A more detailed paper, giving many citations that include other works by Simon, is at http://www.di.dk/di/fokus/laeringsprocesser/Towards_a_model.htm where there is interesting discussion of individual knowledge and leverage of group knowledge that might well apply in contracting.

My feeling is that too much in contracting, where real money and future performance by the Government itself is at stake, is driven by politics, ego, rote, and expediency. In the old words, "This ain't rocket science," but neither is it something where scientific methods need be banned. I often thought it ironic that the whole process leading up to contracting for and demanding high performance in "engineering" was not itself engineered.

Perhaps, even more than in engineering contracts, uncertainty and the unexpected can arise in other types of services. Experience should provide some foreknowledge of where those gray areas may lie and mitigation strategies from R&D and engineering could apply to routine maintenance and other "mundane" contracts.

I have serious doubts about any effort to contract in "making the customer happy" as that can change rapidly for short term reasons having nothing to do with actual performance. I've seen very unhappy customers come around to realize they got great service and benefit and become "happy." I've also seen the reverse and for no other reason than some external situation changed mood. I do believe we can build in contractual mechanisms that can prevent real that situation from being the result of contractual inability to perform due to undefinable, but not entirely unexpected, changes in the environment and built in contractual constraints.

By Michael Love on Wednesday, February 16, 2000 - 10:15 am:

Thanks for all the feedback. This discussion is extremely helpful to me to help clarify my thinking and hope to others. Here are some of my further thoughts. I hope to get more of yours.

Looking at this from a purely contractual aspect, several points emerge:

- Performance based service contracting should only be used where appropriate. "Duh!" you say. But as I understand current policy it is to be used in all (virtually) service contracts. What is most often involved in long term, complex and frequently changing relationships? People working together to best serve other peoples' changing needs. In some cases I suspect that this is nothing more than outsourced employees.
- Vern as usual you've added immeasurably to the discussion and have made me schedule a trip to the library not knowing exactly what "bounded rationality" is. But ignorance is no excuse for silence.
- Complete planning is much more difficult if the task is to describe how to do something or what to do in the future. Describing end results once the service is delivered is easier, for example neat well-trimmed lawns or a current, internet connected desktop computer. So far that logic suggests performance based contracting is the better and the even easier way to go. Current thinking is the problems will disappear if the contractor is told what the result should be, not how to get there. This thinking is flawed in some significant respects. If making the customer happy with whatever service the contractor ultimately provides is specified, the contracting officer or the CO's client who is responsible for the program has no good way to manage it or, more precisely, to monitor contract performance to ensure the end result is not disaster. Thus, particularly for contracts of long duration, the real need is to specify some intermediate performance measures at least. These can be difficult to define contractually and most often become directions on how to perform.
- Another point Vern and Ramon nailed is planning. There may be some agencies that find it difficult to define what they have received from a contractor, much less describe what they will need. This is particularly difficult when the efforts of both Government employees and contractors' employees have contributed to the end results. How are these agencies to realistically do anything but level of effort contracts? Only with difficulty and through redefining some government employees' work.
- From a logical viewpoint, all of these issues can be addressed but the question is whether current policies help or a hinder doing so. Also whether these policies serve or hinder delivering best value for tax dollars.

A slightly different aspect of the issue involves the inability to make any progress in the pre-contract planning stage and subsequent funding emergency with dying money. How many of us have been in contract development sessions where everyone is wrapped around that axle?

I've seen, and I'm sure most others have seen, such meetings bogged down and our own labor dollars going down the rathole of trying to pin down some tiny detail that cannot be pinned down at this stage. Then the issue gets pushed off to another session or a study or a committee with no further progress. There are cases where I'm sure ten times the amount of money was burned in such sessions than on worst case event in the resulting contract.

Parameters may have to be defined, but trying to define the undefinable is a waste. It is much better to "cut bait" on that effort and instead define how we can build in a defining mechanism that allows us to work toward satisfactory solutions within the contract.

In my view part of the problem Vern outlines so well is due to not planning for change with frequent overconfidence that planning is already adequate. The problem will not go away, but a contract predicated on recognition of these facts, immediate planning for change and structuring the contract itself to accomodate quick reaction to those day-to-day surprises at the earliest stages of drafting will help.

For example, a work statement that actually places an obligation to accomodate the unexpected within certain parameters is a possible method. It can even be backed by some funding that can be used by mutual agreement to cover the unexpected while more extensive adjustments are perhaps made for long term solution.

I think many of these problems come from trying to over plan, over define, and being over confident that we have covered the bases -- then we over constrain management. Early analysis and recognition of the degrees of uncertainty, then planning for mechanisms to accomodate them can go a long way in solving problems. It applies to high tech, but also keeping bathrooms clean during a snow emergency.

A correction to the fifth paragraph of my previous post: technical direction clauses say that technical direction does not include making changes. I left out the word "not."

By Vern Edwards on Monday, February 14, 2000 - 04:53 pm:

Mike and Stan:

One reason that agency personnel want to engage in hands-on supervision of contractor personnel is that contracts for long-term, complex services are always incomplete. Better planning can never solve the problem entirely because of bounded rationality, a phenomenon that was first described by Herbert A. Simon and which won him the Nobel Prize in Economics in 1978. He describes the phenomenon in his book, Administrative Behavior, 4th ed. (New York: The Free Press, 1997), which is required reading in virtually every college course in public administration.

The problem of incomplete contracts has been analyzed in depth by the University of California economist Oliver E. Williamson and the University of Virginia law professor Ian R. MacNeil. There is a large academic and business literature about this problem. It comes up frequently in commercial publications such as the Outsourcing Journal.

Performance-based service contracting (PBSC) policy requires agency personnel to plan extensively, something that most of them have not been trained to do; it also assumes that they can plan completely, something that no one can do for complex requirements. No matter how long and hard a group of people think about the future, they will always leave something out. If the future is dynamic and will be determined by forces and events that cannot be controlled or anticipated, and if it is complex, involving many, many "micro-requirements," which is almost always the case, then then planning will be even more inadequate.

PBSC policy does not recognize that parties in long-term, complex relationships will have to engage in ad hoc, sequential decision making, i.e., they will have to respond to requirements as they emerge on an almost daily basis. It envisions static contracts that we can award and then let the contractors perform as they choose in order to achieve the required results. The problem is that the required results are not static; they change frequently.

That's why agency personnel want to supervise contractors. They need to tell them what to do as new requirements emerge, which happens daily in complex service functions. The change order process is too cumbersome; NASA and DoD recognized this long ago and developed "technical direction" clauses to cope with the problem, but they tried to stay within the limits of appropriations law, policy, and orthodoxy by claiming that technical direction did include making changes to the statement of work, a stance that is logically indefensible.

Long-term, complex service contracts are a relatively new phenomenon in the world of business and government. They snuck up on us during the late 70s and early 80s. We need to review our service acquisition policies and bring them into line with the realities of the present.

Vern

By Stan Livingstone on Monday, February 14, 2000 - 02:18 pm:

The current issue (February) of the Nash & Cibinic Report has a timely and excellent article on administering contracts with performance based SOWs. It's well worth reading.

By Stan Livingstone on Monday, February 14, 2000 - 11:24 am:

From what I see, absolutely not. I bet we can count on one hand the number of major contracts fitting into the category you describe.

Writing a true performance based service SOW is hard enough. Providing "hands off" administration makes it almost an impossible task for program managers.

I hope my perception is wrong, but I haven't seen any instances thus far and that's too bad.

By Michael Love on Monday, February 14, 2000 - 10:11 am:

OMB is directing that agencies use performance based service contracting, that is, tell the contractor what goals to achieve, not how to achieve them. Recently OMB also reported to Congress that most agencies were doing a good job with this.

To do this effectively, IMHO, the contractor must be given freedom to manage the contract effort. Something akin to a scope of work that could be done for a fixed price. Note OMB guidance for performance based service contracting specifies that it is not limited to fixed price contracting.

But in order to give the government an ability to ensure satisfactory performance, goals or metrics of some kind must be available that indicate progress fairly frequently (at least every 60 days?) otherwise months of effort can be wasted.

The GSA Seat management contracts are good examples of performance-based systems with detailed metrics but no direction on how the goals are to be achieved.

Now the point, I suspect that such performance based contracts are still the exception. That most contracts are still a version of personal services with performance based language in the contract that is ignored during performance. Instead for services that are more than lawn mowing or cleaning, the government manages the work and the work force.

Recognizing that your responses will be anecdotal at best, I am interested in your perception of whether agencies are really delegating sufficient discretion to contractors on how to perform that performance based service contracts are being widely used.