Game Theory in Management

I once worked with a Vice President who had a Ph.D. in statistics. On this one occasion we were attending a PM workshop of some kind, and, at the end of the day, were in the hotel/conference center’s lounge, having adult beverages and comparing notes on the paper presentations we had attended. This fellow – I’ll call him “Dave,” because that was his name – told me about his dissertation and its subsequent defense. He had developed it in the 1970s, well before personal computers were available, and mainframe computers were just being introduced to college campuses. Dave had taken a couple of computer courses, but was by no means a competent programmer. He did, however, know how to use the text editor program that the real programmers used, and how to send output to a dot-matrix printer, the kind with a tractor feed and green-and-white shaded paper. Since Dave didn’t have access to a working typewriter, he simply typed out his dissertation on the text editor attached to his college’s mainframe, and printed out the results on the dot-matrix printer.

When Dave presented his text to his faculty sponsor, the professor was awe-struck.

“You wrote your thesis on a computer!?” the professor asked.

“Yeah, I made use of the mainframe in the computer science department.”

Dave confided in me that his review committee never asked for any corrections or comment resolution. The paper sailed right through the review, and the defense was unexpectedly light on questions. It was rather plain that the aura surrounding computers of that time – fueled, no doubt, by the HAL 9000 computer from 2001: A Space Odyssey, known for having “never made a mistake, or distorted information” – had so cowed the review committee that they assumed that the resulting text was flawless.

I was reminded of this story when I saw a paper on an OpenAI project to teach a machine how to play a virtual version of hide-and-seek.[i] Once the “room” and “teams” had been set up, several million game iterations were spent rather chaotically, with iterations 0 – 2.69 million spent on seekers “learning” to chase hiders. Episodes 2.69 to 8.62 million saw the hiders using two virtual cubes to block off the two doors in the play area, and 8.62 million to 14.5 million episodes needed for the seekers to utilize an available ramp. Episodes 14.5 million to 43.4 million saw the hiders learn to stow the ramp prior to blocking the doors, rendering them inaccessible to the seekers.[ii] The AI went on to develop surprising behaviors from both the hiders and the seekers, but the above-referenced progress will do for this blog.

Now consider what would happen if a real-life version of the game environment had been created, and the two hiders and two seekers were fourth-graders (typically around ten years old). Let’s further posit that an average game of hide-and-seek would last around two minutes in such a basic setup. Even if we had fourth graders who could play this game non-stop, it would take them over 165 years to perform this number of “episodes.” I’m fairly confident that four typical fourth-graders could discover the door-blocking and ramp-utilization strategies within one day, or even one hour.

So, why is everyone so in awe of artificial intelligence? Have people in general, like Dave’s thesis review committee, become so over-impressed with the implications of AI’s capabilities that any idea that even has the trappings of being associated with it is given deference with respect to its validity?

In my book Game Theory In Management[iii], one of the “games” I evaluated is known as the Ultimatum Game. In this game, the researcher approaches two random people and informs them that he will give them $100 (USD) if Player A can propose its distribution among the two of them, and have Player B accept those terms on the first iteration. Game Theorists (not me) had “calculated” how to maximize Player A’s payoff: by proposing $99 for Player A, and $1 for Player B. The thinking was that Player B would be presented with the choice of receiving $1, or nothing at all, and would always agree with Player A’s distribution scheme.

A funny thing happened on the way to the actual distribution of the $100, however. This strategy almost never worked when tried in real-life. Perhaps put off by such an unfair distribution of unearned largess, or by other factors, the 99 – 1 strategy was almost always rejected. When confronted with the real-life results indicating a broad-based refutation of the Game Theorists’ calculated optimal strategy, many of them blamed “cultural influences” for the discrepancy. But, by pointing to something as inchoate as “cultural influences,” these Game Theorists were essentially admitting that attempting to calculate a specific strategy, even within the confines of something as basic as the Ultimatum Game, was next to impossible due to the number of contributing factors that couldn’t possibly be recognized, let alone quantified.

And so it is, I believe, with much of what presents itself as artificial intelligence. Sure, it’s fun to see how AI can generate graphic images, or even text (I won’t say literature, at least not yet), but when it comes to creating usable strategies in the Project Management world? Its proper response ought to be “I’m sorry, Dave. I’m afraid I can’t do that.”

[i] Bowen Baker, Ingmar Kanitscheider, Todor Markov, Yi Wu, Glenn Powell, Bob McGrew, Igor Mordatch, “Emergent Tool Use From Multi-Agent Autocurricula,” 17 September 2019, retrieved from https://arxiv.org/abs/1909.07528 on August 12, 2024, 19:45 MDT.

[ii] Ibid.

[iii] Hatfield, Michael, Game Theory In Management, Gower Publishing, 2012.

At a presentation I attended as a newly-minted staff member of a now-defunct major beltway bandit, the Vice President on the stage put up a chart I will never forget. It was a line graph, with the line starting high on the Y axis with a slight downward slope, until it got to about midway on the X axis, where it fell to below dramatically. This line, he explained, was the time-phased budget across all active projects, with the fall-off point at around the six month point on the X axis. What it was showing was a rather unsettling truth: if the corporation were to stop bringing in new projects tomorrow, most of us would be getting laid off in around six months.

With a brand-new wife and first mortgage, I was highly disinclined to be laid off for want of new work, so this VP had my full attention. In order to avoid this fate, he continued, it was incumbent on every single member of the staff to put considerable effort into proposing and winning new projects. Oh, and since I was now salaried, this extra effort was expected to be offered freely, nights and weekends. This last part wasn’t articulated aloud – it didn’t have to be. The corporate culture was so focused on winning new project work that technical or administrative excellence became almost an afterthought. In my view, there were essentially two classes of personnel in this organization: Group A were adept at writing winning proposals, and were golden; Group B were second-class citizens, whose employment could come to an end at any time, depending on how they were perceived by members of Group A.

Predictably enough, the corporation thrived, but its employees (especially those in Group B) were rather anxious, generally speaking. But this dichotomy did present an interesting take on the optimal business model in a heavily Project-driven organization. Consider the following payoff grid:

Let’s deal with Scenarios A2 and B1 right away. If the Project comes in on-time, on-budget (or even early and under-budget), AND attracts more work, either from the existing customer in terms of add-on tasks to the existing contract, or even a new contract, then there’s no question such a PM would be considered a success. Similarly, if the PM crashes and burns in actual contract performance, and the customer doesn’t want anything more to do with them, then that PM would be largely considered a failure. Where this payoff grid gets interesting is in the other two Scenarios.

I have often maintained that the whole point of developing a more robust PM capability is to increase the chances that the Project comes in on-time, on-budget, but the above payoff grid challenges that notion. For if, as in Scenario A1, the Project is actually completed successfully, but there’s no follow-on work, either from that specific customer or another within the same industry, what happens then? Well, the members of the Project Team would have to move to other work within the organization, charge their time to an overhead account for a (presumably) limited amount of time, or … get laid off.

Now consider Scenario B2, where the Project comes in late, overrun (maybe both), but, somehow, the PM secures more work in the same or an adjacent field. If the new work has a similar Budget at Completion (BAC) to the just-failed Project, then nobody has to be let go, the Project Team has a shot at getting it right this time, and the corporate higher-ups are happy. My previous definition of a “successful” PM notwithstanding, how could the PM in Scenario B2 be considered a failure? I had a friend and associate who was an exceptional PM, but rarely brought his Projects in on-time, on-budget. How could such a one be considered exceptional, GTIM Nation may ask? It’s because he would be brought in to Projects that were already in deep trouble in cost/schedule space, and would find a way to complete them with only mild negative variances. Essentially, he had a real talent for turning portfolio-cratering disasters into annoyances, and, for this, I consider him one of the best PMs I’ve ever known.

While performing well in cost/schedule space certainly makes it easier to bring in more work over the long term (particularly from the same customer), the key takeaway from the above payoff grid appears to be that the ultimate metric of PM success is keeping the Project work coming in the company’s front door, essentially handing the Strategic Managers (who chase market share) the tools they need for success. Ah, but if that’s the case, I’m happy to let the Strategic Managers have at it, and the answer to the question in the title would be “no.” We PMs did our part by coming in on-time, on-budget. And, if I’m needed to help the Strategic Managers win more work, well, I don’t work for free, evenings and weekends.

Any more.

In last week’s blog, I started with a quote from Eric Hoffer, specifically “Every great cause begins as a movement, becomes a business, and eventually degenerates into a racket.” I went on to discuss how this quote could be paraphrased into something more management-science-y, as in all great businesses start with a grand idea, or discovery, or insight into making a superior business model, etc., then moves into a phase where our friends, the Asset Managers, monetize everything, and then on to a phase where much of the original fire has left the macro organization, and instead it performs like it’s just there to keep existing. Working from that foundation, I want to bring in some Project Management theory, specifically in regards to the question: what’s happening to our subject organization’s business model this whole time? Does its evolution have anything to do with that long-time nemesis of the PMO, the organization’s reluctance to change in general, or accept basic PM precepts in particular?

First, consider how informed decisions are made in medium-to-large organizations. Sometimes you’ll see utter geniuses (or those who think of themselves that way) making virtually all of their choices based on their reflexes, gut feelings, prior experiences, or some combination thereof. Mostly, though, upper-level management will depend on information, whether from direct observation, members of the staff preparing and presenting it, or computer-based information streams. Of that last category, it’s likely that the main component of the diverse Management Information stream is the General Ledger – after all, the point of “all” management is to “maximize shareholder wealth,” right? So, even if the organization’s founder(s) is still passionately pursuing the original vision, by the time the movement turns into a business, and everything’s getting monetized, then the General Ledger, almost definitionally, must be the main source of management information. This being the case, the Chief Financial Officer (CFO) is likely to attain near Oracle-at-Delphi status, being the source and residence of most of the relevant data needed to make informed decisions.

Now, imagine the poor person who’s been assigned the PMO Director position. Reduced to its very core, what is this person’s message? Isn’t it that, if given just a bit of budget and organizational leverage, she can deliver an information stream, outside of the General Ledger, that nevertheless informs decisions on the optimal use of resources in pursuit of accomplishing scope? Just to be clear, this is in stark contrast to the Asset Managers’ message, which can be reduced to “this is how we can make money in the performance of scope realization.” In other words, the central question driving the development of the business model is either “How do we optimize resource allocation to make our customers happy?” or else “How do we squeeze maximum profits from customers, happy or otherwise?”

It's not a trivial distinction. It is, in fact, the determiner of how the business model changes as the organization matures, either consciously or accidentally. For if the pursuit of scope becomes the priority, then those who had been previously approaching Oracle-at-Delphi status when it comes to delivering actionable management information will experience a reduction in their utility and, therefore, prestige and organizational standing. From the Asset Managers’ point of view, those PM-types delivering the occasional insight that helps make a better decision now and again is okay, but the primary basis of the really key choices involving actual currency should be predicated on Generally Accepted Accounting Principles (GAAP). To believe otherwise is to directly challenge their most basic precept, that the point of all management is to “maximize shareholder wealth.” It’s in the very nature of the changing business model to resist significant incursions from newly-established PMOs, particularly in portfolio-level cost and schedule performance analysis – probably the most valuable contribution that a PMO has to offer.

Keep in mind all of this pertains to the organization that’s moving from the “movement” phase into the “business” phase. In the management world, organizations only stay alive as long as they make a profit. By the time the organization is moving from “business” to “racket,” the monetize-everything approach has so dominated the business model that it’s becoming more and more unlikely that any attempt at returning to a customer-focused approach – the very essence of PM – will succeed. The business model has become so ossified as to approach near-granite density. The frustrated PMO Director may come away believing that his efforts have been thwarted by a widespread reluctance to change, but it may easily have been due to a hardened business model, made so by a predictable process, entirely consistent with theories being taught in business schools across the land to this day.

So, to the newly-hired PMO Director, I would say this: Does it feel like bringing changes to your organization is like drilling through granite? There’s a reason for that.

“Every great cause begins as a movement, becomes a business, and eventually degenerates into a racket.”

― Eric Hoffer, The Temper of Our Time

I am going to be so bold as to attempt to convert the above quote from Eric Hoffer to something more applicable to the Management Sciences in general, and PM in particular. I think Hoffer brilliantly laid out the typical life cycle of how ideas gain wide acceptance, are then monetized as they are advanced, and ultimately end up within the domain of an organization that has lost sight of its original underlying vision, and instead directs its energies only to keeping itself in existence. This nominal life cycle structure is rather similar to those laid out by some of the giants of Management Science literature, who have theorized that, absent some form of re-invention of itself, corporations will also tend to follow the template of (1) the invention or discovery of a new technology, business model, manufacturing technique, et.al., (2) the monetization of #1, (3) attempts to overcome competition within a given market, (4) a steady decline in said market share,  until a competitor or newcomer enters Step 1, and either acquires the organization, or drives it out of business.

Next, I want to shine a light on Hatfield’s Incontrovertible Rule of Management #7, that Project, Asset, and Strategic Management are different by type, with different goals, different techniques to achieve those goals, and different Management Information Systems used to execute those techniques. I’m struck by how neatly Rule #7 aligns with Hoffer’s three brackets, so:

• Project Management is all about pursuing Scope, right? Well, what is Scope? Is it not the articulation of a goal, an objective, an idea, or, dare I say, a movement?
• Asset Management’s prime axiom, oft criticized in this blog, is that the point of all management is to maximize shareholder wealth. In my opinion, this is simply an exaggeratedly blinkered way of saying that everything must be monetized, to its maximize capacity, and that this monetization is the primal essence and end-goal of all of the management sciences.
• By the time we’re approaching “racket” status, the organization would have had to have attained significant market share—otherwise, customers wouldn’t stand for a deterioration of the quality, affordability, or availability of its goods and services, and the organization would quickly cease to exist.

Put graphically, this alignment would look like this:

I believe that there’s an organizational behavior and performance component to this, and to illustrate it I want to invoke the Maccoby Archetypes (Craftsman, Company Man, Jungle Fighter, and Gamesman). I think that, if one were to want to measure a given organization’s pace of moving from great cause/movement to business to racket, an excellent way would be to note the number of employees that fall within the Maccoby Archetypes. In the Great Cause/Movement phase, it’s clear that the risk-taking Gamesmen, teamed with the Craftsmen, would be the prime archetypes to advance these ideas. As these ideas become monetized, the organization will likely become less attractive to the Gamesmen, but more attractive to Company Men, who derive their personas from the corporate culture around them. Finally, as the business degenerates into a racket, the Craftsmen will find it more and more unpalatable, but it will become like a magnet to the Jungle Fighters. Put graphically, this alignment would look like this:

Now, not to come across as a total fanboy for PM, but remember the previous reference to organizations following this cycle, absent some form of re-invention of itself? When organizations form and maintain a Project Management Office (PMO), I think this represents an effort to slow or even reverse movement along this structure, as the organization’s energies are re-directed towards its original Great Cause/Movement/ pursuit of Scope. When this happens, I believe that the organization becomes more attractive to Gamesmen and Craftsmen, fueling a sort of symbiotic return of the organization to its original purpose, or cause. I’m reminded of perhaps the ultimate Gamesman, Steve Jobs, whose return to Apple in 1997 was largely seen as the reason Apple returned to prominence after being on the brink of bankruptcy.

So, yeah, the Asset Managers want to monetize everything (I intend to hang that whole “maximize shareholder wealth” stuff on them like a Coleridgean albatross), and the Strategic Managers are looking to maximize market share, and they certainly have their places in the boardroom. You want to return the organization to its original vision?

You’re talking Project Management.

This may be one of my most controversial blogs ever, so strap in. I want to examine, under the rubric of PMO-on-a-budget (ProjectManagement.com’s theme for June), what would happen if the entire PMO staff were to be hired for a fixed rate, for a series of one-year contracts. Providing our jump-off point is my favorite NFL Team, the Dallas Cowboys.

Consider how different the payroll scheme is for the players as compared to the cheerleaders. The players have agents, who attempt to negotiate the highest possible salary, and for the longest period of time, for each of their players. The football clubs themselves have to observe a salary cap, determined by the League office. For the 2024 season, it’s $255.4M (USD)[i]. Dallas’ star quarterback, Dack Prescott, is in the last year of his contract. For this upcoming season, he’s scheduled to earn over 21% of the total allowed for the entire team[ii]. There are 53 people on an NFL team’s roster, which works out to 1.89% of the cap, if it were to be distributed evenly. Dak is not the only star earning significantly more than 1.89% of the cap, of course. Many other exemplary players, whom the Club wishes to retain on the roster for years to come, also easily top that figure. But it has to be said, that, within this business model, every percentage point those stars make over 1.89% simply has to come out the available cap. It’s why, when a particularly talented quarterback makes more than Dak’s 21% haul, it’s a real possibility that other parts of the team will experience a reduction in available talent.

Now, compare and contrast this talent attraction/retention model with that of the Dallas Cowboys’ Cheerleaders, or DCC. Each of them has a one-year contract. If they wish to be a DCC next year, they must compete against all of the other young women who also wish to belong to this elite group each and every time tryouts are held. Veterans can earn more, and are usually (but not always) given deference during the tryouts, but even they can (and do) fail to make the cut. Since both the players and the cheerleaders belong to the same organization, I can’t help but to wonder what would happen if the players’ talent attraction/retention model was identical to the cheerleaders? What would be the implications?

First, the Club would need to publish the payroll structure by position. Since quarterback talent tends to be rarer than, say, linebacker talent, adjustments would need to be made, but always based on percentage points of the total available salary cap. There are 24 “starters” (11 for offense, 11 for defense, with one punter and one kicker), who would receive a higher percentage for each game that they actually started (more incentive to practice well!). Backups would receive a lower percentage, for example:

• The starting QB would be paid 15% of the per-game cap ($2,253,529 USD), but his backup(s) would be back at the 1.89% ($283,945) figure.
• Defensive linemen (typically, 4 starters) would receive 4%, but backups would be 1%.

…and so forth, until 95% of the salary cap had been distributed, leaving 5% for in-season roster adjustments, or even end-of-season bonuses. One-year contracts only. If the veterans want to play next season, they will need to try out, along with all of the other vets, vets from other teams, rookies, undrafted free agents, and walk-ons.

Outrageous? Yeah, probably. But consider some of the implications:

• No more having a cap-busting QB’s salary draining the talent from the rest of the roster,
• Or having 25% (or more) of your payroll sitting on the bench due to injury.
• Agents – don’t bother. The payroll structure is what it is, and Jerry McGuire doesn’t need to shout “show me the money!”
• Think about how incentivized the players would be to achieve the coveted starter position. The energy the coaching staff is expending trying to motivate a bunch of 20-something-year-olds can be better spent in other victory-attaining strategies.
• I would speculate that this business model’s impact on the overall talent level of the team would be to lose super-star players; but, wouldn’t it also eliminate the overpaid ones, as well? And what would be a better predictor of team success, a talent profile that looked like a bell curve, or a roster of all B+ players, across each squad?

Meanwhile, Back In The Project Management World…

Sooooo, what would this kind of model look like at a mid- to large-scale organization’s PMO? To throw out a few numbers, the Director would receive 15% of the indirect budget for the PMO, with three mid-level managers getting 10% each. The Project Controls specialists typically charge their time directly to their projects, but, for the PMO’s “core” team, they would split the remaining indirect budget. Again, one-year contracts only. I believe that this model would bring with it the following advantages:

• No more (Maccoby archetype) Jungle Fighters! PM-types who don’t get ahead based on actual output won’t last long in this business model, and getting rid of the scheming and political maneuvering this type brings may be worth the price of conversion all by itself!
• Elimination of the training budget. Candidates are selected on their current capabilities, not on subjective perceptions of their potential. Note that this would also eliminate the harmful trend of an employee getting trained-up at your expense, only to leave and take this new capability to a competitor.
• No more pushing of superfluous or trendy, untried PM techniques. A given approach to advancing the PM capability has exactly one year to prove itself. If it doesn’t work, its advocates get replaced (risk managers [no initial caps] hardest hit).

For those members of GTIM Nation who would argue that this approach would all but destroy organizational unity and cohesion, I would counter with (a) natural turnover rates may not be that different using the traditional approach, and (b) with the removal of the Jungle Fighters, team comity has a much better chance of developing.

Outrageous? Probably. But that’s not the real question here, is it? The real question is, would this approach work?

[i] Retrieved from https://www.spotrac.com/nfl/cap/_/year/2024/sort/cap_maximum_space on June 25, 2024, 21:04 MDT.

[ii] Retrieved from https://overthecap.com/player/dak-prescott/4848 on June 25, 2024, 21:06 MDT.