While at The Fletcher School, I must have heard the phrase rational actor a thousand times. It appeared everywhere in economics, international relations, negotiation theory, strategic analysis, political science, and organizational behavior. The assumption was deeply embedded in the frameworks: actors evaluate incentives, assess consequences, and pursue outcomes aligned with their interests.
And honestly, I remember quietly cringing almost every time I heard it.
Not because the concept itself was flawed as a modeling tool, but because I had already spent enough time in the real world to realize how rarely I encountered truly rational actors in practice.
I had worked with companies where departments openly undermined each other despite shared goals. I had watched executives reject objectively beneficial ideas because they threatened internal politics or exposed prior mistakes. I had seen organizations spend more energy protecting territory than improving outcomes. Even in markets and negotiations, people routinely made decisions driven less by long-term optimization and more by fear, ego, identity, short-term pressure, or self-preservation.
The longer I worked inside organizations, the more I began to suspect that much of modern business theory rests on assumptions that break down the moment actual humans enter the equation.
One of the most persistent assumptions in economics, international relations, management theory, and corporate strategy is the belief that people behave rationally. Entire frameworks are built around it. Markets are expected to respond logically to incentives. Governments are assumed to pursue outcomes aligned with national interests. Executives assume employees will support initiatives that improve the company. Consultants assume departments will cooperate once the “right answer” is presented clearly enough.
Yet the real world repeatedly shows us something very different.
Not because people are inherently irrational, reckless, or emotional, but because the definition of “rational” changes depending on whose interests are being protected. What appears irrational at the organizational level may be entirely rational at the individual level. What looks strategically self-destructive from the outside may make perfect sense to the person whose status, authority, political survival, or identity feels threatened.
That gap between system logic and human logic explains far more business failure, political instability, and organizational dysfunction than most frameworks are willing to admit.
The Rational Actor Model Was Never Meant to Be Reality
The rational actor model was always intended to simplify complexity. Economists needed a way to model markets without accounting for every emotional impulse or personal bias. International relations scholars needed a framework for understanding state behavior without having to trace every internal political feud, bureaucratic rivalry, or leadership personality. Management theory needed a way to assume organizations would act in pursuit of efficiency and shareholder value.
As a modeling tool, the concept has value. Without simplification, systems become impossible to analyze. The problem begins when institutions stop treating the rational actor model as a simplification and start treating it as reality. Because people rarely optimize for the system itself. They optimize for their position within the system.
That distinction changes everything.
The Diplomat’s Problem
I was reminded of this while watching a Season 3 episode of The Diplomat. The premise in one scene hinged on the assumption that the British Prime Minister, referred to as a “rational actor,” would cooperate because the proposed outcome created clear mutual benefit. The logic was clean and familiar: cooperation reduced risk, aligned interests, and created the best strategic outcome for all parties involved.
In rational actor theory, this should work.
But human beings are rarely optimizing solely for collective outcomes.
A political leader may instead prioritize domestic political survival, public perception, coalition management, ideological positioning, or the avoidance of humiliation. A decision that appears irrational from the standpoint of long-term national interest may be entirely rational when viewed through the lens of personal political preservation.
This is one of the enduring weaknesses in many international relations frameworks. States are often described as unified strategic actors pursuing coherent national interests, but in reality, they are collections of competing bureaucracies, personalities, factions, incentives, and pressures. The “state” is rarely thinking with one mind.
Corporations operate much the same way.
The Myth of the Unified Corporation
Organizations love to speak as though the company itself is a singular rational entity. “The company decided.” “The business wants transformation.” “Leadership has aligned on the strategy.” These phrases create the illusion of unified intent.
In practice, however, organizations are ecosystems of competing priorities, competing incentives, and self-preservation mechanisms.
There is also an underlying assumption in much of business theory that companies naturally optimize toward shareholder value, as though the organization itself were a single coherent economic brain. In reality, that assumption often breaks down long before decisions reach execution.
As I explored in The KPI Trap, organizations frequently become collections of localized optimization systems where departments pursue their own metrics, incentives, political capital, and operational survival rather than broader enterprise outcomes. Everyone may technically be doing their job while the company itself becomes less effective.
This is something I have quietly wondered about for years. Much of modern business theory assumes organizations are populated by rational actors aligned around shared outcomes, yet many companies function in ways that directly contradict those assumptions.
After working with countless organizations over the years, I am often amazed not by how efficiently companies operate, but by the fact that many operate successfully at all.
Some organizations feel less like coordinated systems and more like barely controlled chaos held together by momentum, institutional inertia, a handful of exceptional individuals, or market position. In some cases, it feels as though a few capable leaders at the top are constantly steering a ship that naturally wants to drift sideways due to internal friction, competing incentives, and organizational gravity.
That raises an uncomfortable question that most management frameworks rarely acknowledge: are many companies successful because of their operating models, or despite them? And perhaps more importantly, how much more effective could organizations become if people actually behaved the way the models assume they do?
If departments genuinely optimized for enterprise outcomes rather than local KPIs, if leaders rewarded long-term contributions over short-term optics, and if employees trusted that change would not automatically threaten their positions, many organizations would likely move faster, innovate more effectively, and waste far less energy fighting themselves internally.
Instead, a significant portion of organizational energy is spent navigating politics, protecting territory, minimizing perceived threats, and managing misalignment between competing definitions of success.
Legal departments optimize for risk reduction. Finance teams optimize for cost control. Marketing optimizes for attribution and visibility. IT optimizes for stability and operational continuity. Product teams protect roadmap ownership. Executives manage investor perception and internal politics. Middle managers protect relevance, authority, and headcount. Employees focus on workload, compensation, recognition, and career security.
The result is that many organizations are not operating as rational unified actors at all. They are coalitions of semi-aligned actors, each responding rationally to different incentives and different definitions of success.
Why “Obviously Good” Initiatives Fail
This helps explain why so many transformation projects collapse despite appearing strategically sound.
Leadership teams frequently announce reorganizations, digital transformations, AI initiatives, operational efficiency programs, or centralized governance structures, assuming employees will support them because the initiatives are “good for the company.”
But employees are not evaluating the initiative through the lens of shareholder value or enterprise optimization. They are evaluating it through the lens of personal consequence.
Will this increase my workload? Will this expose weaknesses in my team? Will automation reduce my importance? Will another department gain influence at my expense? Will this make my role redundant six months from now?
These are not irrational questions. They are deeply rational questions for someone trying to preserve their career, maintain relevance, and reduce uncertainty.
In many organizations, leaders mistake resistance for ignorance when it is actually self-preservation. That distinction matters because it fundamentally changes how transformation must be managed. You cannot overcome fear with another PowerPoint slide explaining efficiency gains. You cannot assume alignment simply because the business case is strong.
The Economic Version of the Same Problem
This disconnect becomes even more visible during periods of economic stress. Policymakers often explain inflation, tariffs, energy restructuring, or industrial policy changes through macroeconomic logic. Citizens are told that temporary pain will create long-term resilience, national competitiveness, or future economic advantage.
Economists may be correct in the aggregate, but individuals do not experience aggregate theory. They experience rent increases, grocery bills, fuel costs, layoffs, and uncertainty.
Telling people to “stay calm” and endure hardship for the greater good assumes they trust the institutions asking for sacrifice and believe the burden is being distributed fairly. When trust is weak or the pain feels unequal, the rational actor model begins to collapse because individuals naturally prioritize immediate survival over abstract future benefit.
Again, the issue is not irrationality. The system assumes everyone shares the same incentives and time horizons. They do not.
The AI Adoption Contradiction
The current AI transformation cycle is exposing this problem in real time.
Many executives assume AI adoption is inevitable because the productivity improvements appear obvious. From a system perspective, the argument makes sense. Faster output, lower costs, increased efficiency, and scalable automation all appear objectively beneficial.
But workers are not evaluating AI from an enterprise optimization perspective. They are evaluating it through the lens of job security, transparency, leverage, and future relevance.
Employees who resist AI adoption are often portrayed as resistant to progress, yet their concerns may be entirely rational given the incentives and risks they personally face.
What makes the current AI moment especially fascinating is the degree to which even the companies promoting AI transformation are revealing the contradictions inside the rational actor model.
Several major consulting firms and technology companies are now openly pushing employees to become heavy AI users, framing adoption as essential to future relevance and employability. In some cases, the messaging is only subtly disguised: embrace AI, integrate it into your workflow, increase productivity, or risk becoming redundant.
There is a deep irony in this. Many of the same firms selling AI-driven efficiency and workforce optimization to clients are simultaneously warning their own employees that failure to adopt AI may reduce their value to the organization. The very systems being sold to eliminate inefficiency elsewhere are now being used internally to pressure workers to continuously justify their own existence.
From a leadership perspective, the logic appears rational. If AI increases productivity, reduces repetitive work, improves speed, and enhances output quality, then encouraging adoption should strengthen both the company and the employee. In theory, everyone benefits: the company becomes more efficient, clients receive faster delivery, employees elevate their capabilities, and the organization becomes more competitive.
But that assumption depends heavily on trust.
Employees are being asked to embrace systems that may eventually reduce headcount requirements, compress staffing models, or redefine the value of institutional experience. Even when leaders frame AI as an augmentation tool rather than a replacement mechanism, workers are still evaluating the long-term implications for their own relevance and bargaining power.
That creates a fascinating question: is wholehearted adoption always rational from the employee’s perspective?
An employee may fully understand that AI improves organizational efficiency while simultaneously recognizing that improved efficiency often leads organizations to ask difficult questions about staffing levels, utilization, and cost structures. From the company’s perspective, adoption is rational because it strengthens competitiveness. From the employee’s perspective, cautious adoption may also be rational because the same efficiency gains could eventually reduce the need for their role altogether.
This is where many AI discussions become overly simplistic. Organizations tend to frame resistance as fear of technology or unwillingness to evolve. In reality, much of the hesitation may stem from a very rational assessment of shifting power dynamics, economic incentives, and long-term career risk.
The irony is that both sides may be behaving rationally at the same time — just according to different definitions of survival.
Humans Are Not Spreadsheets
Ironically, this is also why soft power matters so much in both international relations and organizational leadership.
Hard power assumes compliance follows from incentives and consequences. Soft power recognizes that legitimacy, trust, emotional alignment, respect, and shared identity influence behavior just as strongly as formal authority.
People are far more likely to support change when they feel secure within it. They are more willing to cooperate when they believe their interests are understood and their sacrifices acknowledged. Most organizations underestimate this profoundly. They continue to believe data alone creates alignment when, in reality, human beings are not spreadsheets waiting to be optimized.
The lesson here is not that rational models are useless. Simplification remains necessary for strategy, economics, and governance.
The danger comes when leaders forget that models are abstractions rather than reality itself.
Organizations fail when executives assume employees will naturally support what benefits the company. Governments fail when policymakers assume citizens will absorb pain for future collective gain without questioning fairness or trust. International strategies fail when nations are treated as unified rational actors rather than unstable coalitions of competing pressures.
The greatest strategic mistake many leaders make is assuming people will behave according to organizational logic rather than personal incentive structures. Because when the rational actor isn’t rational, the problem usually isn’t irrationality at all.
It is that every actor in the system is rational according to a different set of incentives.
