Game Theory in Management

“And here comes in the question whether it is better to be loved rather than feared, or feared rather than loved. It might perhaps be answered that we should wish to be both; but since love and fear can hardly exist together, if we must choose between them, it is far safer to be feared than loved.”

                        -- Niccolo Machiavelli, The Prince[i]

I once worked for a medium-sized U.S. Government contractor. This particular company had a bad case of what I like to call the Lifeguard Syndrome, or the notion that, if they could just hire the right Chief Executive Officer, or some other high-level exec, that this person would rescue the organization from drowning in its problems. One of the Presidents helicoptered in was rumored to have said, soon after his arrival, that “Every employee should arrive to work each day a little bit afraid.”

At the time I didn’t know what to make of that quote, assuming that he had actually said it. On the one hand, I could kind of see the underlying thought pattern, that the rank-and-file employee would typically not do their absolute best if they were complacent about their long-term futures with the company. On the other hand, I had first-hand experience with a much larger government contractor, where fear of being laid off was both pervasive and intense, leading to a highly toxic work environment, and I absolutely did not want to relive that nightmare. That having been said, there was really no comparison between the smaller company’s overall performance and the larger one: the larger, more fearful (albeit toxic) organization dramatically outperformed the more chill one.

On a more extreme scale, I’ve observed one common element among truly notable champions across multiple sports: they’ve displayed a strong aversion to, or even fear of, losing. In NFL Films’ production America’s Game, the 1992 Dallas Cowboys, Michael Irvin had the following to say about Head Coach Jimmy Johnson:

See, that’s the thing … you can’t lose here. Jimmy made it so uncomfortable losing.[ii]

Many other champion head coaches, like Bobby Knight or Vince Lombardi, were known to become rather, shall we say, harsh towards their players after losses.

I’ve been on high-performing teams. I’ve also experienced my fair share of teams headed by complete incompetents, and it showed. The thing about the poorly-managed teams was that, though the manager had set a sub-optimal (or even unworkable) technical agenda, they would attribute the team’s downward performance spiral and eventual failure to the team’s lack of ability or commitment, and felt at liberty to invoke more harsh treatment of them on that basis. Rather than inspire better performance, these team members would typically (and predictably) resent such handling. The more talented members would seek employment elsewhere, up to and including competitors’ organizations, while the less confident would stay on, enduring the worsening treatment from management, pursuing the wrong technical agenda, and hoping for the best. Were they fearful? You bet. Did it lead to championship-level, or even improved performance? Of course not.

And let’s not forget that, on the other end of the spectrum, PMs who fail to adequately address consistently poor performers are likely to be judged rather harshly by the other members of the Project Team. At some point and at some level, the others will recognize that the presence of poor performers reduces the chances of bringing the Project in on-time, on-budget, and they will most likely have an (entirely reasonable) expectation that the PM will rid the Team of the non-contributors.

For better context, I want to revisit Hatfield’s Incontrovertible Rules of Management #24, which reads:

24. The three essential qualities of managerial leadership are:
    1. The ability to identify the optimal (or at least workable) technical approach to the problem(s). The Team will quickly know if they are barking up the wrong tree, and will become disillusioned.
    2. To genuinely care about the people on your Team. If you don’t care about them, why should they care about you, or your technical agenda?
    3. To be willing to pursue your technical approach, alone if necessary.

My previous story covered why (a) is essential, but what about (b), and how does a PM dealing with failure harshly fit into that? Here’s where I want to call upon an enhanced level of nuance. Harsh treatment from valid leadership isn’t about inducing fear of the leader, personally. It’s about inducing fear of failure, no matter who is in charge. One litmus test would be to ask the question, does the individual continue to loathe the thought of losing even in the absence of the harsh manager/leader? I invite GTIM Nation to recall a teacher that managed to inspire their best performance. I have absolutely no trouble admitting that I was at least a little bit afraid of these types of instructors, but there’s something else: they left me with a stronger aversion to failure, not for its own sake, but because I came to expect more and more out of myself.

So, as to the answer to my original question, about whether or not it’s a good thing for employees to show up to work a little bit afraid, I’ll leave an overall response to individual GTIM Nation members. For my own part, I think that, if I ever lose my aversion to failure, if I’m not at least a little bit afraid of falling short, I’m probably in the wrong job. It’s a fear of failure – any other variety may very well lead to a highly toxic work environment, and I wouldn’t wish that on anybody.

[i] Retrieved from https://www.enotes.com/topics/prince/questions/could-you-explain-what-qoute-means-when-ch-17-269368 on February 25, 2024, 17:38 MST.

[ii] From America’s Game, the 1992 Dallas Cowboys, as shown on https://www.youtube.com/watch?v=C6z-Zb6WcHw, viewed on February 24, 2024, 19:39 MST.

This week I want to expand on one of the main assertions from last week, that comparing time-phased budgets to actual costs (“actuals”) has nothing at all to do with project cost performance, regardless of granularity, as counter-intuitive as that may seem. To be clear, I’m not saying that comparing how much you’ve spent to your budget, time-phased or otherwise, is useless in all cases. If nothing else, it’s an indicator of how much you have left to spend before you go over budget. But “How much do I have left to spend?” is a very different question than “How is my Project (or Control Account, or Work Package) performing with respect to its costs?” Confusing these two questions can lead to ill-advised Project decisions.

Games People Play (Grove Press, 1964) created quite a sensation when it was published. In it, Eric Berne asserted a series of “mind games”[i] that people will engage in, or “play,” that carry an emotional or egotistical payoff, but have little or nothing to do with the actual source of the original interpersonal conflict. It is not my intent to engage in long-range psychoanalysis of people whom I’ve never met, primarily for two reasons: (1) Games People Play, while a best-seller, has subsequently declined in its status as an insightful psychological treatise, and (2) I’m not a psychologist. That having been said, I think it’s fascinating how one “game” in particular, “Now I’ve Got You You Son Of A…”[ii] , umm, female dog, where small errors are blown up to absurdly exaggerated proportions, seems to occur among managers who are against the advancement of PM capability within their organizations, for whatever reasons, but can’t come right out and say so. It’s been my observation that, instead, they will use the comparing-budgets-to-actuals technique to masquerade as legitimate cost performance analysis, and try to influence others to do so, as well.

To demonstrate why comparing budgets to actuals is such a poor technique, consider that the point of any Management Information System (MIS) is to deliver actionable, accurate information. Any information stream that isn’t accurate will not only fail in its primary objective, it will actually work against it, since it often becomes misleading rather than merely decision-neutral. Its implications are shown in the following payoff grid:

Scenarios B and C represent the way Projects’ performance reflected in its cost performance information system ought to function. We only really get into difficulty in Scenarios A and D. I’ve referred to Scenario A as an executive’s worst fear because, in my experience, upper management endures a great deal of angst at the thought that they are sitting on top of a major cost overrun on one (or more) of the projects within their portfolios, and no one is telling them about it. At the other end of the organizational power structure, the members of the Project Team despise Scenario D, where some mis-aligned or erroneously-designed cost/schedule performance system raises a red flag for poor performance when, in fact, everything is doing fine. Ill-advised upper management-types descend upon them like locusts, demanding causal analyses and corrective actions for problems that don’t actually exist, accompanied by the implicit threat of reducing personnel assigned. Consider a scenario where some mis-guided analyst, upon comparing cumulative actual costs to cumulative time-phased budget, discovers that the former figure is significantly higher than the latter, and raises the alarm. Add to this the undiscovered fact that the Project in question has actually accomplished a far larger percentage of the scope than had been planned so that this Project is, in fact, doing really well in cost performance space. No matter – the analyst plays the PM-version of the Bernesian game “Now I’ve Got You…”, the Project Team suffers, and executive attention is turned away from where it ought to go. Equally as damaging is the opposite scenario, where actual costs lag behind the budget, but so does accomplishing scope. The budgets-to-actuals comparison won’t flag this Project as a problem, even though it most certainly is, leading to anxiety-producing Payoff Grid Scenario A.

Nor does cranking up the granularity of budgets-to-actuals (BtA) technique work. Consider a project that, in its original Basis of Estimate (BOE), estimated $75,000 for labor, and $25,000 in equipment, while, in the event, it’s on a pace to spend $70,000 on equipment, and $20,000 in labor. Even though this project is on a pace to come in waayyyy under budget, the BtA comparison indicates real problems, placing the Project Team right back into dreaded Scenario D.

So, my recommendation is to avoid any comparisons of budgets to actuals when the discussion turns to Project cost performance – unless you’re really in to mind games.

[i] Retrieved from https://ericberne.com/games-people-play/now-ive-got-you-you-son-of-a-bitch-nigysob/ on February 11, 2024, 12:26 MST.

[ii] Ibid.

Regular GTIM Nation readers are probably familiar with Hatfield’s Incontrovertible Rules of Management #3, which reads

The 80^th percentile best managers who have access to only 20% of the information needed to obviate a given decision will be consistently out-performed by the 20^th percentile worst managers who have access to 80% of the information so needed.

So, what happens when managers experience a decline in the amount of the information they need to consistently make optimal, or even appropriate, decisions? Sure, they make poorer and poorer choices, but the results that come from a management environment where the information systems are falling short, in my experience, follow a very predictable path: management becomes increasingly reactionary.

And by reactionary, I mean that these actions become more dramatic, intense, and far-reaching, because the problems they address have been allowed to become more dramatic, intense, and far-reaching than they would have been if they had been discovered earlier. For example, the problems behind the Control Account that’s about to significantly overrun next reporting period call for more extreme remedies than when that same CA experienced a mild out-of-threshold negative Cost Variance six months ago. Make no mistake – over-correcting is often just as damaging to the situation than no correction at all. A pilot once told me that they even have an axiom for this phenomenon, I think it’s along the lines of “if you get behind on your controls, you’re lost.” I interpreted this to mean that if the pilot is surprised by the reaction of the aircraft to the changes to the controls, he’s likely in trouble.

So, how does the PM or PMO Director prevent an erosion in the Project Management Information Systems (PMISs) needed to keep the decision-makers informed? Let’s pull out the Game Theorists’ favorite analytical tool, the Payoff Grid, shall we?

Note that, unlike many other Payoff Grids, this one has only one acceptable scenario, (B). In the other scenarios, even if the PM is savvy enough to know the value of Earned Value or Critical Path Methodologies-based systems, if such systems are unavailable, chances are that, sooner or later, they will enter into the downward spiral of purely reactionary management. But here’s the kicker – I have personally witnessed (and I’m sure many other PMs have, as well), time and again, the PMs themselves will interfere with or even defeat attempts to set up the very PMISs that not only should they seek, but the other, more enlightened PMs will need to attain Scenario (B).

It follows, then, that this battle must be fought on two fronts: (1) Any PM who eschews the basic MISs needed to measure the Project’s cost and schedule performance should be given only the most basic, least-technically challenging scope to work, and (2) a valid PMIS must be available to all of the PMs, wanted or not. I’ll admit right now that front #1 is easier to work, because front #2 raises the question “what counts as a valid Project Management Information System?” For this we turn to Hatfield’s Incontrovertible Rule of Management #22, which reads:

All useful management information has the following three characteristics:

• • It is accurate,
  • It is timely,
  • And it is relevant.

If I could add a corollary, it would be that the General Ledger is incapable of providing all of the information needed to assess a Project’s cost performance. That crucial piece of information only comes from a working Earned Value Management System, even if that system is very basic. Indeed, the practice of using only the budget and actual cost data elements, typically from the general ledger, to determine the Project’s cost performance is nicknamed the “watching the actual costs go by” method, and is a clear indicator that the PM doing so is doomed to enter the reactionary scenarios above. Also noteworthy is the fact that comparing budgets to actual costs remains a useless analytic technique even when the level of granularity is increased. Whether performed at the total Cost Account level, or at the Basis of Estimate (BoE) line-item level, it still has nothing to do with cost performance. As far as deriving Project cost performance is concerned, comparing budgets to actuals is not only inaccurate, it has a 50% chance of being flat-out wrong, meaning that decisions based on it have an even chance of also being wrong.

PMs struggling to bring in their projects on-time, on-budget while facing an institutional lack of information are tragic figures. PMs who choose to be misinformed are just irksome. In my opinion.

I think I will add a new entry to Hatfield’s Incontrovertible Rules of Management that reads:

A plurality, if not a majority, of business model pathologies have their roots in deficiencies of character.

For example, consider Chesterton’s Fence. The notion that someone would have to articulate that it’s a bad idea for a new property owner, upon discovery of a fence in the middle of that property that did not fulfill any intuitive function, have that fence torn down has nothing to do with any theory of optimal property sub-division. Rather, it addresses a tendency of human nature to act before having the relevant information associated with those kinds of decisions which, in turn, come from the character deficiencies of impatience and arrogance.

When it comes to creating anew or replacing existing upper management within the Project Management Office, one common business model pathology I have witnessed has the tendency to create barriers to eventual PMO success, and is addressed in Hatfield’s Incontrovertible Rules of Management #12, which reads:

In large-scale re-orgs, particularly one where two (or more) groups are being merged, employees displaying the highest level of loyalty to new management tend to be rewarded more generously than the most talented members.

In other words, loyalty, not talent, is the coin of the realm in new or re-organized Teams in general, and PMOs in particular.  Writing along similar lines, Nicolo Machiavelli stated “The conqueror does not want doubtful friends who will not aid him in the time of difficulty…”[i] Of course, Machiavelli was writing about the highly politically-charged and violent nature of 15^th Century Italian city-states, but I’ve found some of his writings to be applicable to the organizational behavior and performance characteristics of corporations.

When we combine this effect with those instances where the new PMO management is seeking to either introduce a new technical approach to advancing the organization’s capability maturity, or significantly altering the existing one, we have the following payoff grid:

Every new PMO Director hopes for Scenario B, and may even believe themselves there even as evidence of one of the other Scenarios mounts. But, if the PMO Director has selected a poor technical approach to the advancing-PM-capability problem, then the last thing he would want to see is a staff who either doesn’t recognize its deficiencies, or is willing to declare loyalty to an approach they know to be flawed (Scenario A), which is why Scenario C is the next-best one.

The problem being illustrated by the above payoff grid is not that of a disloyal staff needing to be marginalized, but that of the new PMO leadership automatically assuming that their new/modified technical approach is the optimal one. For if the PMO leadership does not arrogate to itself the presumption of being right 100% of the time with respect to the selected technical agenda, they would be open to the benefits of arriving in Scenario C, recognizing that it is vastly superior to Scenario A. Scenario C carries with it the implication that the PMO leadership can course-correct in time to attain some demonstrable level of PM capability maturity before the macro-organization’s executives become disappointed, whereas becoming enmired in Scenario A points to a path that eventually delivers a situation where the PMO director is comfortable and confident, all the way up to the point that their prioritization of loyalty over talent precipitates an erosion (if not collapse) of the very organizational support needed to maintain their under-performing PMO.

And, while all of these pressures stem from a fear of the PMO Director finding herself in Scenario D, I must admit that, even though it looms larger than it should, the possibility of the existing PMO staff refusing to support a new, demonstrably superior technical approach cannot be ignored. But if we assume that the Maccoby[ii] archetypes of the Gamesman and the Craftsman would typically recognize a superior technical approach when they see it, and the Company Man will usually accept the trajectory of the organization around him, then that leaves only…

That’s right, the Jungle Fighter. It’s the presence of Jungle Fighters within the Team that dramatically increase the odds of the new PMO Director finding himself in Scenario D. What we have here, essentially, is the worst type of Maccoby archetype driving an artificial inflation of the perceived value of the oleaginous loyal over the truly advanced-in-PM members of the PMO which, in turn, significantly lower the odds of the discovery or successful implementation of the best technical approach to attaining the PMO’s goals. So, new PMO Director, before you replace the existing Team’s more advanced members with those who strike you as being more trustworthy, think twice.

Unless the existing people are risk managers. Feel free to immediately send them to someone else’s Program Office.

[i] Machiavelli, Niccolo, The Prince, retrieved from https://effectiviology.com/strategy-lessons-from-machiavelli-the-prince/ on January 26, 2024, 17:32 MST.

[ii] Maccoby, Michael, The Gamesman, The New Corporate Leaders, Simon & Schuster, 1976.

Without getting into specifics, if one were to place too much of a specific isotope of Uranium in the same place and at the same time, it would initiate a fission reaction, also known as “going critical.” If this were to happen unintentionally, many, many people would have a very, very bad day. This being the case, how on Earth, I can hear GTIM Nation say, is Michael going to draw this analogy, between a nuclear fission event and your run-of-the-mill Project Management Office, or PMO? I’ll explain.

Recall my oft-cited Hatfield’s Rule Of Management #23: Affordability, Availability, Quality, pick any two. When PMOs first start off, they are typically made up of small-to mid-sized teams, even within large macro-organizations. Many will fail, but I’m convinced that its odds of success are significantly enhanced if their overarching strategy takes into account the previously-stated rule, and aligns itself with the owning macro-organization. For example, a contractor that does not deal with new or cutting-edge technology, known for its ability to deliver more basic, usable output on-time and on-budget, would probably not benefit from a PMO that insisted on a rigorous, highly-detailed audit-survivable Cost/Schedule Control System. On the other hand, an organization that did work with high-profile, cutting-edge technology, intended for mission-critical applications is likely to have a customer highly desirous of accurate and timely performance information, and would likely fail without such a Cost/Schedule Control System.

But here’s the rub. Many (if not most) medium-to-large contracting organizations will not be totally one or the other. Their portfolios are likely to be mixed, with a combination of high- and low-risk projects, rendering a one-size-fits-all PMO approach more of a liability than an asset. And yet, it has been my experience that the small/initial PMO, as it moves towards mid-size and maturity, will prioritize the writing and publishing of procedures and guides, standardizing the techniques and outputs for PM in general, essentially mandating just such an approach.

Now consider what is likely to happen when all of the elements of this PMO are brought together in close intellectual proximity, throwing out their ideas on how PM should be advanced within the macro-organization but needing the rest of the PMO to agree with them. Some will, no doubt, stress the need for an enforced solution (basically, writing down everything they believe needs to be done in a procedure, which is then approved by upper management), while others will point to more recent capability maturity theory as the only way to go. Still others, wanting to avoid organizational conflict, will suggest a cafeteria approach, where the technical managers decide the level of baseline and performance measurement system robustness, and there will always be the ones who believe that more PM training will automatically move the macro-organization closer to an advanced PM capability. As the debate continues, temperatures rise, and those who are least prepared to defend their assertions rationally may resort to pointing to their education, including their certifications, or experience in some gee-whiz project, all of which tends to produce more heat than light. It all reminds me of Hatfield’s Rule Of Management #26, that you can put fifty PMs in a room together, and they will not be able to agree on the color of an orange.

As the free neutrons competing ideas slam into other radionuclides self-designated experts, the entire core technical agenda can go critical degrade, with only the introduction of graphite tamp a compromised solution (which nobody really wants, but offends the fewest PMO Team members) preventing a complete meltdown technical agenda collapse. So, how do we avoid even approaching this scenario in the first place?

Let’s start by circling back to a couple of assertions made earlier in this blog, specifically (a) that some parts of the portfolio are going to need only basic PM capabilities in order to increase their chances of successfully bringing in their projects on-time, on-budget, while others will need significantly more robustness in those systems, and (b) a one-size-fits-all approach ignores those needs. This being the case, the savvy PMO director will bifurcate the resources into two Teams: Team 1 will leave the level of cost/schedule performance measurement systems robustness to the Technical PMs, and simply provide the talent to make it happen. This Team will be affordable and available. Team 2 will bring with them the expertise to make advanced PM Information Systems a reality. They will be available and quality-oriented. These two Teams will only play by the same set of rules if those rules are general in nature, as their specific missions will be quite different. By splitting the core PMO’s deployed resources along these lines, you are decreasing the odds that it will go critical arrive at a dysfunctional business model, which would lead to, well, a lot of people having a bad day.

For those members of GTIM Nation who disagree with me, I’m looking forward to seeing y’all nuke my arguments in the comments section.