Many of us are interested in explaining and predicting why organizations behave in the way that they do, and how that behavior influences and is influenced by their external environment. Complexity theory offers one way to accomplish these tasks. The fundamental idea behind complexity theory is that individuals, groups, and organizations make up systems of highly interdependent actors that interact in nonlinear ways. As these actors adapt to their local environment, they influence how other actors of the system behave, creating a perpetual cycle of feedback loops. The positive feedback loops generated from these interactions amplify and crowd out other behavior resulting in a pattern that defines the system. Once a pattern is established, systems become locked into self-reinforcing cycles of highly predictable collective behavior. Thus not only do system actors unknowingly influence the broader system, the system itself influences actors, creating a self-reinforcing loop.
Consider a simple comparison between something that is complicated and something that is complex. A motorized vehicle equipped with thousands of moving parts is complicated but not complex because these parts function in a very well-orchestrated and predictable manner. Although parts of the vehicle interact, their localized behavior does not vary in any nonlinear way as a result of the localized behavior of other parts, keeping the entire system—the vehicle—quite predictable. In comparison, the cause of gridlock is highly complex because each driver adapts to his/her local immediate surroundings (e.g., one way streets, traffic lane availability, short cuts), which impacts the behaviors and decisions of other drivers, and so on. After dozens or hundreds of these feedback loops across drivers, traffic congestion emerges and reemerges at random places and points in time.
Organizations as Systems
Suppose you want to understand why businesses are characterized by different cultures despite operating in a similar industry or context. Complexity theory suggests that you can answer that question by posing a series of simple inquiries: 1) what are the actors of the organization, 2) how are these actors connected with one another, and 3) what are the norms and rules that govern the interaction of these actors. Employees, teams, departments, and leaders represent actors that behave based on their own beliefs, personalities, preferences, and skills to optimize their goals and objectives (e.g. sales employees aim to maximize commissions while operations managers aim to maximize efficiency). Formal and informal channels connect actors of the organization in ways that make interaction among them more or less likely, such as reporting lines between managers and executives, inter-departmental teams, or even the geographic proximity and social relationships among organizational members. Finally, rules and norms guide the interaction among these parts as reflected in standard operating procedures, decision-making processes, routines, and policies. Organizational culture can therefore be seen as a highly unpredictable and entrenched pattern that results from cycles of positive feedback loops as organizational members who are formally and informally connected follow norms and rules of interaction to achieve their own needs. Once established, the culture facilitates a self-reinforcing loop of predictable collective behavior within the organization, which is why, for example, employees tend to personify their organizational culture despite behaving quite differently in other environments.
Other Instances of Complexity Theory in Action
Complexity theory can and has been used to understand many aspects of contemporary economic life beyond just the organization. For instance, teams are made up of members who have specific skills, objectives, and preferences that interact with those of other members. Complexity theory would suggest that the outcome of teamwork is not only a function of who is on the team, but how and to what extent the members are interdependent. Indeed, this is precisely what Baruck College Professor William Millhiser found. Because high dynamism between system actors leads to unpredictable outcomes, teams are more effective when members are chosen based on high levels of interdependence rather than skill.
At the industry level, complexity theory is useful in understanding the role of organizations in contributing to a very volatile financial system prior to the 2008 financial crisis. The highly interdependent organizations included five major investment banks, two financial lending conglomerates, three securities insurance companies and three rating agencies. With unprecedented financial innovation in the backdrop of deregulation, investment banks found ways to make bets on all sorts of financial products by purchasing common loans from lenders, such as mortgages and car loans, and repackaging them as collateralized debt obligations (CDOs) to sell to investors. The key rating agencies—influenced by the fact that they were paid by the investment banks to rate CDOs—rated most of them the favorably despite many of them containing low grade or ‘junk’ loans. Securities insurance companies offered investment banks the opportunity to insure against the potential default of these CDOs, encouraging investment banks to develop and sell default-prone CDOs to unsuspecting investors while betting against them by purchasing insurance.
As each actor in the system responded to its own localized environment, there were extraordinarily high executive bonuses, a quadrupling of rating agency profits, billions of dollars in investment bank fees and interest earned on CDOs, unprecedented lender revenue on securities, and virtually no risk sustained by lenders and investment banks because any risk of default was now transferred to the investor. These outcomes represented positive feedback loops in the system, crowding out any behavior that might have otherwise created a less volatile financial system. Investment banks borrowed to acquire as much cash as possible to maximize these feedback loops, which further entrenched the system. With risk now transferred to the investor, lenders focused on lending wherever and whenever possible and investment banks were eager to purchase these assets to package them into CDOs, with rating agencies wiling to disguise their true value. As a result, predatory lending skyrocketed leading to a boom in the subprime mortgage market because investment banks earned higher interest rates on these mortgages. Mortgages quadrupled from 2000 to 2003 and subprime loans increased from US$30 billion to US$600 billion in 10 years, resulting in an unprecedented housing bubble.
Systems are highly resilient to change. Bringing in a new CEO or changing how employees are evaluated are probably insufficient to shift the culture of an organization. Similarly, a moral stand taken by executives of an investment bank to avoid subprime loans would be easily absorbed and corrected by the financial system as profit levels nose dive in comparison with competitors. But once systems reach a tipping point, they can undergo dramatic changes in feedback loops that produce an alternative pattern. University of Maryland Professor Thomas Schelling showed that residents who preferred to have up to 30% of their neighbors to be of the same ethnicity as them lived in a neighborhood that was segregated close to this preference. But once the residents’ preferences increased to 33% (the tipping point), the resulting neighborhood was 100% segregated. As each neighbor responded to this seemingly small change in preference, the system shifted dramatically to an alternative pattern which bore no resemblance to the outcomes intended by the individual neighbors. Similarly, as mortgage and loan holders started to default on their loans, lenders were unable to sell their loans to investment banks because the market for CDOs collapsed, leaving investment banks holding hundreds of billions of dollars in loans. Because actors in the financial system were so interdependent, once the system reached a tipping point (certain number of defaults), it was vulnerable to a dramatic and sudden change that saw investors lose tens of trillions of dollars in equity, led to home owners losing their homes, and resulted in up to 30 million people unemployed and a doubling of the US national debt.
Implications and Limitations
Complexity theory can be useful in understanding why some organizations are more or less successful than others. For instance, because system patterns are impossible to predict, competitors, and even managers themselves, find it very difficult to explain how organizational cultures emerge. It then follows that if the culture of an organization is one that fosters high levels of productivity and success, businesses can exploit their culture for competitive advantage because their competitors would be hard pressed to prescribe the right configuration of actors and interactions to achieve a similar culture. It is then logical for managers to foster interdependencies within organizations whether among individuals of a team, teams within a department, or among departments to facilitate unique patterns of collective behavior that are both valuable and difficult to imitate by competitors.
This means though that complexity theory is not very helpful in prescribing how organizations unlock system patterns to facilitate change at the system level, nor is it helpful in finding or creating the critical tipping points required to shift systems. Most importantly, once a tipping point is reached, complexity theory is unhelpful in predicting whether the ensuing pattern is indeed what was intended. This is because individual actors of a system do not have the cognitive capability to fully understand the actors of a system and the underlying connections, norms, and rules that govern interaction.
But while deliberate systems change is less common for this very reason, it has not stopped some organizations from developing parallel system patterns of behavior to bypass the current system. In their efforts to challenge the highly entrenched and linear industrial system of production, consumption, and waste-to-landfill, New Jersey based TerraCycle collects and adds value to waste by upcycling it into new products. To do this, TerraCycle had to understand the key actors of the existing system (e.g., product manufacturers, retailers, consumers, waste collectors), how they are connected, and the norms and rules that govern their interaction. They then worked to build an alternative waste system that brought in new actors, and altered how existing actors were connected with each other. For instance, consumers are not only recipients of products but suppliers of waste for new products. If enough actors of the existing system see benefit to justify switching from disposal to upcycling, a tipping point is possible. To do this, Harvard Business School Professor Michael Porter suggests that businesses need to increase the value pool for actors in the system so that they benefit more from an alternative approach. Organizations like TerraCycle that have succeeded in expanding the value pool have drawn on the principles of complexity theory by building clusters of actors with complementary resources and establishing connections among them with norms and rules for interacting that produce a pattern of behavior that creates more overall value than the incumbent system.
In sum, complexity theory pushes us beyond a ‘cause and effect’ understanding of the relationship between organizations and their environments by explaining that organizations (and their members) operate in systems with multiple actors who, in their respective efforts to optimize goals and objectives, interact with each other resulting in feedback loops that produce a pattern of collective behavior that is nonlinear and impossible to predict. This pattern then locks participants into a particular way of operating, which helps in our understanding of why organizations behave in the way that they do.