How everything is advancing except for management science.
At the dawn of the Industrial Revolution, engineers decided that work being done in factories — which were powered by the recent invention of the steam engine — needed to be planned out, and workers supervised, to maximize efficiency and lower costs. Thus, the science of management was born. The status of science as an approach to maximize efficiency and lower costs in engineering efforts is — of course — controversial.
But what is science anyway? Philosophers of science define science (somewhat tautologically) as being whatever: 1) follows the scientific method, and 2) is experimental (points 1 and 2 are related). More precisely: science is a diminutive (i.e., limiting) protocol that controls for confounders and mitigating factors to describe causal relationships in nature. Sometimes, the relationships that science describes are not so casual: quantum physics brings this uncertainty to the forefront. However, this work is about management. What makes management a science?
Management is a science because it is an extension of engineering efforts to reward society through the efficient allocation of goods and services (we will discuss how management is experimental later). Management operates under the assumption that the “marketplace is a dominant means of organizing the exchange of goods.” The economist Adam Smith, likewise, points out that specialization and coordination within a marketplace are what lead to economic growth. Economic growth is not simply a testament to good practices and business sense: it contributes to social well-being (and occasionally detracts from it). If psychology observes the mind, and sociology observes society, management observes efficiency.
Beginning with the Industrial Revolution, production moved increasingly from family-oriented to factory-oriented production. The questions that emerge in this shift include: how can a manager motivate workers in a factory? In a family-oriented work environment, you have to work to eat. Plus, your loved ones depend on you to provide for them. In a factory setting, workers can lay about or even break machines. How does a manager standardize work? How does a manager hire the best workers (i.e., for the best roles)?
Management is required because — without management — work is disorganized and production cannot be tracked adequately. Before management science was increasingly implemented in Western Europe and the U.S., workers produced too little because they were worried that overproduction would lead to them losing their jobs. When ‘managers’ were polled, no one had answers about quotas or margins. They didn’t know how to hire the right people: anyone would.
One can power a family on this kind of organizational model. One cannot power a company or a nation this way. The economist Friedrich Hayek demonstrates that the market is the best form of organization precisely because management practices can be applied there: how does one manage a family? Easily, if the family is functional. How does one manage two, three, or four ‘families’ (i.e., on the market that rewards creative production)? Only through proper management.
How Management has Changed
As corporations have increased and diversified management approaches for them have also. Technology has developed — telecommunications, AI, big data — that makes management both easier (or streamlined) but also more important to get right, since so many factors now depend on proper management. Management approaches need to adapt to these changes; it seems (sadly) like everything has advanced except for management science.
A shift in management practice should follow the shift from linear to complex organizations, but this has also not been the case. Complexity requires a new methodology: particularly Bayesian methodology. In the mathematician Thomas Bayes’ system, each new input into a system changes the algorithm of a system that produces outputs. This is useful in a complex, rather than a linear organization. Algorithms can help computers and human observers determine the best course of action based on what has happened in the past and what one can expect to happen in the future.
These are not ‘learning’ equations (do not confuse Bayesian mathematics with AI) but a set of contingent on-off-switches that account for all permutations of possible events, even ‘fuzzy’ (i.e., probabilistic) ones. Bayesian statistics is what allows a manager to ask: last time I pursued initiative Y, I gained X% in sector C. What would Y look like in sector D, and how would I adapt Y for any sector?
Why Best Practices Have not Kept Pace
The preceding talk of algorithms and Bayes’ mathematics ignores (perhaps purposefully) that organizations are managed by people, and people are biased. Any intelligently-managed system still requires intelligent people at the end of the day to read outputs and make decisions. The prospect of letting a computer make the final decisions for you or your company is the stuff of dystopian horror novels, not a book about management.
Beyond having a human decision-maker, it is pertinent to have a wise human decision-maker. But — barring that (i.e., wisdom is not a cheap commodity) — it is best for someone who can recognize their biases. Bias is neutral in my perspective. Evolutionary psychology suggests we evolved biases like we evolved our hands and feet. However, in professional settings, the onus is on us to be able to recognize and control evolutionary biases. We go the extra mile in presenting ourselves appropriately because we are getting paid. Some aspects of our human nature are disregarded in the work setting, and this is why management is effective in the first place. This is too much of an aside, however.
- The Endowment Effect
The endowment effect is an analog of what in economics is called a “sunk-cost fallacy.” Once, I was waiting in line with my client for lunch. We were working in an office building where the downstairs restaurant was a taquería. If one leaves the building, however, they have the option of sandwich places, pizza, sushi, etc. On a cold day, we decided to wait for tacos. I wanted tacos; my client did not. Near the end of the line, I was ready to put in my order for tacos, and she was not. “I don’t want tacos,” she informed me, “but I waited this long.”
This, in a nutshell, is the sunk-cost fallacy. Because you did something wrong for four minutes (i.e., wait in line), you have to double down and spend money on something you don’t want.
“That’s the sunk-cost fallacy,” I informed her. She left the line and went and got a sandwich. In the office break room, I could tell she was satisfied with her decision.
The endowment effect works similarly because we do not want to let go of what we have ownership of. In the case of the taquería, it is our time lost waiting in line. In more relevant work examples, it is our legacy interest in old-date tools, plans, and consultants. As humans, we have an inclination to reward loyalty to others. We also like (and crave) stability. Combine the two, and you have a bad recipe for advancing new ideas in management. Sometimes — like we have to train our biases — we also have to overcome our tendencies towards comfort in instances where we do have to adopt new practices, including radical ones.
- Prospect Theory
The philosopher David Benatar does not write about management. Rather, he writes about the relevance of utilitarian philosophy — the notion that what is “good for all” is desirable over the absence of any good whatsoever — to the extreme application of the relative value (i.e., or absence of value) of human life on Earth. I will not reveal Benatar’s conclusion (you can read it for yourself), but one supposition he uses to make a utilitarian argument about life’s overall value he calls an “asymmetry matrix.”
The asymmetry matrix holds that what is good is less desirable than avoiding what is bad. Humans will — possibly without exception — opt for less pain over more pleasure if they were offered these options as a trade. That is, I would not trade 30 minutes of intense pleasure for 30 seconds of intense pain. Pain is worse than pleasure is good, according to Benatar.
Prospect Theory — which was developed by the psychologists, Daniel Kahneman and Amos Tversky (Kahneman won the Nobel Prize for his findings on Prospect Theory) — states that people hate losing by a factor of two or three times more than they like winning. For the same reason that humans avoid pain rather than (necessarily) pursue pleasure (i.e., if these were not mutually exclusive), we evolved to be risk-averse.
Another philosopher, W.F.V. Quine, points out that creatures who did not evolve to be risk-averse “have a pathetic but praiseworthy tendency to die [off] before reproducing their kind”. It is not surprising that humans are risk-averse when it comes to trying out new things — we evolved to be that way — and because we perceive risk at a greater scale than we perceive a potential benefit. Even if the risk is low and the benefit is high.
- The risk of change
The majority of management initiatives fail. However, the majority of everything fails, so we are in good company here. Let me qualify my statement: some research demonstrates (consider the irony of this statement in a second) that the majority of clinical and experimental research is non-reproducible or not robust enough to be reproduced. Experimental physics has the highest rate of reproducibility among the sciences, but even then, it is not perfect (psychology dips down below 50%). In life, likewise, a series of personal initiatives may fail: perhaps you did not get the right job the first time, or your first relationship did not lead to marriage.
As humans, we are apt to look at the failures more than the successes. The failures seem to number greater for the same reason we are more averse to pain than we are in pursuit of pleasure. It is evolutionary — natural — to be risk- and change-averse as people, but this principle should not be superimposed on management requirements in certain cases. At least, we should recognize when change is needed and when what is needed is to hold on to what we have.
Why Modern Enterprises Needs New Approaches
Trillions of dollars are wasted each year on applying what usually only works in linear problems to the problems of complex, non-linear organizations. Not only are the statistical algorithms behind these companies’ operations, not Bayesian, the thought patterns of the human minds that interpret algorithms and charts are not adaptable either. I have discussed before how cognitive biases contribute to the problem of stagnation in management science. For fifty years, perhaps since the Great Society programs elevated the science of management into something necessary for good government (i.e., in addition to good private administration), we have not advanced as a science while every other discipline has. How does management move forward in modern enterprises?
The philosopher Thomas Kuhn talks about “paradigm shifts.” You have possibly heard the term ‘paradigm’ thrown around in meetings, but what is a paradigm?
A paradigm is a global organizing model or theory with great explanatory power…. By choosing [paradigm], I mean to suggest that some accepted examples of actual scientific practice — examples which include law, theory, application, and instrumentation together — provide models from which spring particular coherent traditions of scientific research.
Psychologist John McLeod describes that “Thomas Kuhn argued that science does not evolve gradually towards truth. …A scientific revolution occurs when: (i) the new paradigm better explains the observations, and offers a model that is closer to the objective, external reality; and (ii) the new paradigm is incommensurate with the old.”
In other words, a paradigm shift in management requires a radical reinvention of management practices in line with evidence and tempered but brave intuition. Management will not evolve without a push in the right direction. An advancement in management science requires experimentation — literally and figuratively — and courageous action on the part of individual managers who want to improve efficiency and productivity with new approaches. Google has certainly been creative in its experimentation in management: to the extent, it pays off, it is a cause for discussion.
I will be courageous enough to describe a new paradigm for management, even if it is a broad one for which we can find new laws and approaches: systems-oriented thinking.
Systems Thinking is the New Paradigm
There are two types of reasoning styles employed within organizational decision-making (analysis and synthesis); let us see if we can shift the paradigm from one to the other. Analysis views individuals, somewhat correctly, as being part of a greater system. In this sense, analysis is what is called a reductionist approach: analysis contends that the whole is not any more than the sum of its parts.
Synthesis is a reasoning style that considers individuals to be correlated in a complex ecosystem where — critically — the relationships between parts are even more important, or revealing, than the whole the parts belong to. Both are correct lenses to look at organizations, but one (synthesis) is flexible for the modern enterprise. The philosopher Ken Wilber terms elements that are privy to these kinds of synthetical relationships ‘holons.’ A holon is part of a greater holon that is part of a greater holon. It is holon “all the way down,” according to Wilber.
Analysis attempts to decompose things into their primary particles to consider them in isolation. Synthesis builds from the bottom up of observation and notes the relationships that bring elements together, not only in terms of the category they belong to. Analysis works well, of course. We have gotten modern science from analysis. But analysis works well when there is a low level of interconnectivity, such as the electrons in Bohr’s model of the atom. Many complex organizations have high levels of interconnectivity, more similar to what is known contemporarily as fields or waves.
Synthesis works for items like ecosystems, computer networks, etc; complex organizations, in other words. Synthesis contends that relationships between systems are not the correspondences of static elements. For a certain type of systems-thinker, a relationship is an element itself.
A proper application of synthesis leads to an understanding of holism within organizations: parts of an organization are explicable only by reference to the whole, or — more complexly — to some meta-level that precedes and is encompassed by the whole. There are relationships everywhere like there are holons all the way down.
Managers Need to Tackle their Biases
To overcome a bias to change a habit. All thoughts are habits, as contended by the philosopher Charles S. Peirce. This means: humans only doubt what we experience through a privation (or withdrawal) of action that is caused by an external event (e.g., when a clock strikes at midnight, and we are forced to remember that we had forgotten the time…)
We do not plan for events; we plan for contingencies. This is a crucial lesson for a manager to learn: not just for themselves, but for their employees. To get better at practice — to produce the best practices and to stick with them — one must do them often and (almost-)always until they become routine. To take the lessons in this work: Bayesian adaptability, systems-thinking, and control for cognitive biases mean you have to actually put these lessons into practice. Like there are learning organizations, there are learning managers. This work is the first step to learning better management and advancing the field.
If you want to learn more about the innovative field of systems thinking, click on this article:
Benatar, D. (2009). Better to have never been. Oxford University Press, Oxford University, Oxford.
Bright, D. et al. (2019). Principles of management. Openstax College, Rice University, Ann Arbor, MI.
Haack, S. (2009). Evidence and inquiry. Prometheus Press, Amherst.
Hayek, F. (1996). Individualism and economic order. University of Chicago Press, University of Chicago, Chicago.
Hoopes, J. (1993). Peirce on signs. University of North Carolina Press, University of North Carolina, Chapel Hill.
Ioannidis, J. (2016). Why most clinical research is not useful. PLOS medicine 13(6).
Kuhn, T. (2012). The structure of scientific revolutions. University of Chicago Press, University of Chicago, Chicago.
McLeod, S. (2020). Thomas Kuhn — science as a paradigm. PsychologyToday.
Santos, L. and Rosati, A. (2015). The evolutionary roots of human decision making. Annual review of psychology, 3(66).
Wilber, K. (2000). A theory of everything. Shambhala, New York.