Dynamic and “emergent” complex systems.1 can be found everywhere from ecosystems to economies to our underlying biology. By striving to understand the way these systems work, we can make sense of the world in which we live and better prepare for future events. This approach is known as systems thinking, and it is gaining in popularity in many fields including economics, finance, and investment management.
Read this blog post to learn about systems thinking and how it can be applied to sustainable investing.
Systems Thinking in the Financial Markets
Scholars and practitioners have described financial markets as complex adaptive systems (CAS) in which many different participants within the system “constantly change their actions and strategies in response to the outcome they mutually create.”2
Central to this notion is the idea that discrete actions of system participants or components modify the overall behavior of the system in a way that is more than the sum of those participant or component parts. This concept is called “emergence,” and what emerges at the system level is referred to as the system’s “emergent properties.”
In the realm of financial markets, outcomes are more than the aggregation of individual investor decisions. Interaction among the participants within the investment process — buyers, sellers, brokers, dealers, analysts, managers, or advisers — generates behaviors at the system level. The resulting emergent properties might include market volatility, risk, and return distribution. These patterns are particularly relevant given the increased market share of index funds that track broad market movements.
Financial markets are especially complex because systems are embedded within other systems in ways that produce emergent properties at each level. Equity mutual funds, for example, are collections of different stocks modified by fund managers at varying times that produce emergent, fund-level risks. Likewise, hedge funds are systems of activity of investors and hedge fund managers who produce emergent hedge fund strategies that impact the broader investment ecosystem.
An important thing to remember is that emergent properties then shape the subsequent activities of underlying market participants who, through their interactions, generate market-level volatility, risk, and return characteristics. In turn, the new market characteristics shape future activities like buying/selling of securities, reallocation of portfolios, and the ability of companies to raise capital.
Exhibit 1 demonstrates the emergence of system-level features from individual or component “agent” interactions (bottom-up causality) and the influence of those emergent features back onto agents (top-down causality).
Exhibit 1. Bidirectional Influence of Agent-level and System-level Features
Emergent properties in finance are significant because they allow us to understand noteworthy events in market behavior like financial bubbles and crashes. Notably, emergence is happening all the time, not just during times of high volatility. Sometimes, the underlying dynamics of the system reinforce ongoing stability in the market. In the world of dynamical systems — where all possible states of the system are mathematically modeled as vectors across a state space3 — stability can manifest as an “attractor” toward which the unfolding system gravitates.
Constraints Matter
Systems thinking offers new insights for analyzing past market behavior. Beyond tracking historical trends in the market, we must also consider historical constraints.4 Unlike direct causes, constraints work by shaping the possibility landscape.
Although constraints may carry a negative connotation because they are generally understood as restrictive, some constraints open new possibilities within the system. Referred to as “enabling constraints” by scientists, they influence interactions that drive the system toward a particular emergent state that would otherwise be unavailable.
Consider what happens when a roundabout replaces stop signs at an intersection.5 This constrains the behavior of each car. Stop signs facilitate stop-and-go coordinated behavior from their constituents, whereas roundabouts constrain movement to enable a slow, ongoing flow of traffic within the circle. Stop signs require each driver’s attention to be oriented to cars in multiple directions, whereas roundabouts demand attention to oncoming traffic in one direction.
Importantly, the newly constrained patterns of traffic flow enable a drastic decrease in the likelihood of accidents. Put simply, roundabouts constrain patterns of behavior in ways that alter the probability of car crashes and bring about new system-level interactions that are unavailable with stop signs.
In finance, we often seek out direct causal forces to explain crises. An example is the 2008 housing market crash. We might consider the foreclosures of subprime mortgages in mortgage-backed securities to be such a direct cause. But constraints have a unique role to play in the causal story because they facilitated the likelihood of a system-wide crash.
Lenders were subject to lax underwriting standards and disclosure requirements, which increased the likelihood of offering loans with unconventional, higher-risk terms. While low- and middle-income households depended on home ownership as a primary source of financial security, many of these homeowners were unfamiliar with the risks associated with unconventional loans.
In addition, the low interest rate environment drove a wide range of lenders and consumers across the United States to refinance existing loans with non-traditional and adjustable-rate mortgages (ARMs). Lenders and consumers became entangled in a web of risk layering where unconventional terms such as no-downpayment, interest-only, and piggyback loans were combined.
In systems thinking parlance, market participants engaged in a vast network of loan agreements that constrained their future behavior and produced “a geometric increase in the propensity to default.” 6 The growing network of ARMs established pre-2008 served as enabling constraints within the system, producing levels of risk within the housing market that were unforeseen.7
Importantly, enabling constraints are context dependent. In the roundabout example, the constraints that produce fewer vehicle accidents are well documented. In locations where cyclists are common among cars, however, roundabouts increase bicycle-related accidents. Thus, constraints in one context might have a different effect in another context.
While ARMs themselves are not inherently problematic, when placed in the context of unsustainably low interest rates followed by rate resets and falling housing prices, the chance of mass foreclosures leading to a market crash was high. A recent publication also emphasized that the probability of a market crash was grossly underestimated because the practices used to model risk and predict behavior were not robust for mass risk layering. Rather, they were more suited toward short-term interactions with independent parameters, such as when traders optimize their derivatives portfolio given current market conditions.
Investing in Resilience for a Sustainable Future
There is tremendous potential for systems thinking when investing in a sustainable future. The CFA Institute Research and Policy Center’s Climate Data in the Investment Process points out that climate hazards pose a threat to every aspect of our lives, from the safety of our homes to our basic social infrastructure. Undoubtedly, climate change will bring long-ranging effects across jobs, industries, and economies. It will also likely exacerbate existing inequalities and pose significant challenges to developing markets.
As one of the most complex problems we face today, the lingering threats brought about by climate change demand thinking about long-term, systemic impacts. Understanding how to effectively channel resources to mitigate climate risk and generate resilience will be of immense value for building a sustainable future.
Like the housing market crash, climate-related hazards involve the emergence of risks that are more than the sum of their parts. These risks can produce direct effects, such as property damage caused by a flood, and indirect effects, such as transportation or business disruptions. And they can propagate beyond the immediate economic system, impacting global supply chains and production in dependent industries.
The risks that emerge from natural hazards constrain the subsequent dynamics of the system, transforming that system. A recent study highlighted the distributional network effects following extreme flood events in Austria. It identified significant negative impacts on public budgets, public goods and services, and final-demand goods and services, all of which are important for reconstruction after a natural disaster. And while capital owners and high-income households were most significantly affected in the short-term, the study found greater long-term effects on low-income households due to rising prices and capital scarcity.
Because the behavior of individuals, institutions, and industries are interconnected, negative effects propagate throughout the system following extreme events. These changing dynamics can not only limit the system’s return to normalcy but also produce additional negative effects (e.g., the inability to rebuild and increasing inequality) and render the system more vulnerable to additional shocks.
Predicting the impact of a natural hazard thus requires looking beyond the strength of the current system and analyzing the system’s future states given unfolding constraints.
As climate change continues, hazards are more likely to occur either concurrently, as witnessed with natural disasters that hit during the COVID-19 pandemic, or one directly after another. In the disaster science literature, these are called compounding and cascading threats that require systems thinking to analyze.
The heightened connectivity of our world means that systemic risk management is needed not only to model future impacts, but also identify ways enabling constraints can be reoriented to drive the system away from vulnerability and toward resiliency. In other words, we need to determine when and where to replace stop signs with roundabouts.
To address flood risks, governments are likely to build dams or levees to protect areas vulnerable to flooding, but this leaves in place many of the enabling constraints that could lead to widespread economic collapse. Barrier construction can even create a false perception of safety and impenetrability, resulting in even fewer resilience and mitigation efforts. If a significant flood event were to break the dam, the community would have no additional capabilities for dealing with the systemic effects.
Setting goals and maintaining priorities that directly impact the system’s current state may be helpful in some contexts but ultimately neglects the evolving possibilities of the dynamic world we live in. Policymakers should therefore develop plans and procedures that proactively address future emergent risks and guard against unwanted system dynamics by selectively modifying system constraints.
Flood-risk mitigation might include strengthening public services, offering protections for producers of certain goods or services, and providing extended capital support to low-income households. Instituting such policies can be challenging as large investments in resilience efforts often produce no immediate returns.
However, these investments have the potential to significantly reduce spending in the long-term. One study modeled 3,000 natural hazard scenarios and found that strengthening additional infrastructure had a benefit-cost ratio greater than 1:1 in 96% of scenarios (this ratio was greater than 2:1 in 77% of scenarios and greater than 4:1 in 55% of scenarios).
For their part, investors can develop economic sustainability strategies oriented toward strengthening systems against exogenous shocks. Because constraints are context-sensitive, attention to the specific dynamics of each system or subsystem is needed to determine exactly where and how to generate resilience and create value.
For example, educational supports that encourage problem-solving and cognitive skills over numeracy and literacy were found to be key in generating resilient labor markets in Ethiopia. Another study found investing in tailored diversity efforts within the US workforce may have a positive impact on the ability to withstand exogenous shocks at the firm level. A complex systems understanding of resilient investment strategies would go beyond traditional (environmental, social, and governance) ESG investment criteria (e.g., company practices or industries) and assess companies based on their potential to modify agent interactions within the firm or market in which they are embedded to strengthen the overall system.
At the same time, improper climate adaptation strategies run the risk of reinforcing structural inequalities within a society, rendering that society more vulnerable to shocks. Because system constraints are context-sensitive, investors who partner with local businesses, organizations, and groups sensitive to the specific needs of the community are more likely to find success building resilient systems. Integrating context-specific knowledge can foster ease of adoption and increase effectiveness when transforming the system toward specific goals or outcomes. Such a lens is particularly useful for impact investing, which aims to produce measurable positive environmental and/or social effects while achieving a financial return.
Investing in resilience means more than just making short-term adaptations to guard against potential shocks. Too often the focus is centered on addressing immediate disruptions, such as business discontinuity and single area supply chain issues, rather than on generating long-term resilience that encompasses system connectivity.
Creative and collaborative solutions, including the development of new financing and investment instruments, may be needed to successfully address the threats posed by climate change. Only through a careful analysis of the emergent future states of the system and corresponding enabling constraints can we hope to develop these long-term climate risk mitigation strategies and identify key opportunities for sustainable investment.
Bitcoin 
Tether 
XRP 
Dogecoin 
USDC 
Lido Staked Ether 
LEO Token 
