Introduction: The System Lens

Managers are not confronted with problems that are independent of each other, but with dynamic situations that consist of complex systems of changing problems that interact with each other. I call such situations messes.… Managers do not solve problems, they manage messes.

—RUSSELL ACKOFF, operations theorist

Early on in teaching about systems, I often bring out a Slinky. In case you grew up without one, a Slinky is a toy—a long, loose spring that can be made to bounce up and down, or pour back and forth from hand to hand, or walk itself downstairs.

I perch the Slinky on one upturned palm. With the fingers of the other hand, I grasp it from the top, partway down its coils. Then I pull the bottom hand away. The lower end of the Slinky drops, bounces back up again, yo-yos up and down, suspended from my fingers above.

"What made the Slinky bounce up and down like that?" I ask students.

"Your hand. You took away your hand," they say.

So I pick up the box the Slinky came in and hold it the same way, poised on a flattened palm, held from above by the fingers of the other hand. With as much dramatic flourish as I can muster, I pull the lower hand away.

Nothing happens. The box just hangs there, of course.

"Now once again. What made the Slinky bounce up and down?"

The answer clearly lies within the Slinky itself. The hands that manipulate it suppress or release some behavior that is latent within the structure of the spring.

That is a central insight of systems theory.

Once we see the relationship between structure and behavior, we can begin to understand how systems work, what makes them produce poor results, and how to shift them into better behavior patterns. As our world continues to change rapidly and become more complex, systems thinking will help us manage, adapt, and see the wide range of choices we have before us. It is a way of thinking that gives us the freedom to identify root causes of problems and see new opportunities.

There are two fundamental ways to respond to an ever-changing world: when events happen, you can react, or, with knowledge of how events work, you can be a participating actor instead of a passive reactor. Systems thinking and systems analysis allow you to map out a range of different elements and illustrate how they might be interrelated. This can allow you to test your mental models of what is happening and see the unintended consequences of decisions. For instance, you can gain a good overview of the 2007–2008 financial crisis when you look at its causes from a systems analysis perspective.

Systems thinking is a discipline for seeing wholes rather than pieces. It is a framework for seeing interrelationships rather than things, for seeing patterns of change rather than static snapshots. It is also a set of specific tools and techniques, originating in two threads: in feedback concepts of cybernetics and in servo-mechanism engineering theory dating back to the nineteenth century. Beyond this, systems thinking is a sensibility—for the subtle interconnectedness that gives living systems their unique character. There is something in all of us that loves to put together a puzzle, that loves to see the image of the whole emerge. It is no coincidence that the words "whole" and "health" come from the same root (the Old English hal, as in "hale and hearty"). The unhealthiness of our world today is in direct proportion to our inability to see it as a whole.

So, what is a system? A system is a set of things—people, cells, molecules, or whatever—interconnected in such a way that they produce their own pattern of behavior over time. At its most basic, a system is the holistic sum of interdependent parts which interact with one another to fulfill a specific purpose. The crucial point here is that of interaction and interdependence: without these, you merely have a loose collection of parts, not a system. A kitchen in and of itself isn't a system because not all of the parts interact with one another, and it doesn't have one clear overarching purpose; whereas, an oven is a system, in that all of its working parts must interact in order for it to operate for the purpose of cooking food. The system may be buffeted, constricted, triggered, or driven by outside forces. But the system's response to these forces is characteristic of itself, and that response is seldom simple in the real world.

When it comes to Slinkies, this idea is easy enough to understand. When it comes to individuals, companies, cities, or economies, it can be heretical. The system, to a large extent, causes its own behavior! An outside event may unleash that behavior, but the same outside event applied to a different system is likely to produce a different result.

Think for a moment about the implications of that idea:

• Political leaders don't cause recessions or economic booms. Ups and downs are inherent in the structure of the market economy.
• Competitors rarely cause a company to lose market share. They may be there to scoop up the advantage, but the losing company creates its losses at least in part through its own business policies.
• The oil-exporting nations are not solely responsible for oil-price rises. Their actions alone could not trigger global price rises and economic chaos if the oil consumption, pricing, and investment policies of the oil-importing nations had not built economies that are vulnerable to supply interruptions.
• The flu virus does not attack you; you set up the conditions for it to flourish within you.
• Drug addiction is not the failing of an individual and no one person, no matter how tough, no matter how loving, can cure a drug addict—not even the addict. It is only through understanding addiction as part of a larger set of influences and societal issues that one can begin to address it.

This same systemic insight applies to grand geopolitical struggles. Consider the U.S.-U.S.S.R. arms race that dominated the latter half of the twentieth century—a race to see who could get fastest to where no one wants to go. This tragic competition drained the U.S. economy and devastated the Soviet economy. Its roots lay not in rival political ideologies, nor in particular weapons systems, but in a way of thinking both sides shared. Each nation perceived the other as the cause of the threat, while remaining blind to how its own arms buildup provoked precisely the response it feared. The arms race was a systemic phenomenon, driven by a reinforcing feedback loop that neither side could escape until one concluded it was no longer willing to play. Today, much of the world finds itself caught in another race that appears to be heading toward where no one wants to go—the so-called "war on terrorism," governed by a similar systemic dynamic.

Something about statements like these is deeply unsettling. Something else is purest common sense. I submit that those two somethings—a resistance to and a recognition of systems principles—come from two kinds of human experience, both of which are familiar to everyone.

On the one hand, we have been taught to analyze, to use our rational ability, to trace direct paths from cause to effect, to look at things in small and understandable pieces, to solve problems by acting on or controlling the world around us. That training, the source of much personal and societal power, leads us to see presidents and competitors, OPEC and the flu and drugs as the causes of our problems.

On the other hand, long before we were educated in rational analysis, we all dealt with complex systems. We are complex systems—our own bodies are magnificent examples of integrated, interconnected, self-maintaining complexity. Every person we encounter, every organization, every animal, garden, tree, and forest is a complex system. We have built up intuitively, without analysis, often without words, a practical understanding of how these systems work, and how to work with them.

Modern systems theory, bound up with computers and equations, hides the fact that it traffics in truths known at some level by everyone. It is often possible, therefore, to make a direct translation from systems jargon to traditional wisdom.

Because of feedback delays within complex systems, by the time a problem becomes apparent it may be unnecessarily difficult to solve. — A stitch in time saves nine.

According to the competitive exclusion principle, if a reinforcing feedback loop rewards the winner of a competition with the means to win further competitions, the result will be the elimination of all but a few competitors. — For he that hath, to him shall be given; and he that hath not, from him shall be taken even that which he hath (Mark 4:25) or — The rich get richer and the poor get poorer.

A diverse system with multiple pathways and redundancies is more stable and less vulnerable to external shock than a uniform system with little diversity. — Don't put all your eggs in one basket.

Ever since the Industrial Revolution, Western society has benefited from science, logic, and reductionism over intuition and holism. Psychologically and politically we would much rather assume that the cause of a problem is "out there," rather than "in here." It's almost irresistible to blame something or someone else, to shift responsibility away from ourselves, and to look for the control knob, the product, the pill, the technical fix that will make a problem go away.

Serious problems have been solved by focusing on external agents—preventing smallpox, increasing food production, moving large weights and many people rapidly over long distances. Because they are embedded in larger systems, however, some of our "solutions" have created further problems. And some problems, those most rooted in the internal structure of complex systems, the real messes, have refused to go away.

Hunger, poverty, environmental degradation, economic instability, unemployment, chronic disease, drug addiction, and war, for example, persist in spite of the analytical ability and technical brilliance that have been directed toward eradicating them. No one deliberately creates those problems, no one wants them to persist, but they persist nonetheless. That is because they are intrinsically systems problems—undesirable behaviors characteristic of the system structures that produce them. They will yield only as we reclaim our intuition, stop casting blame, see the system as the source of its own problems, and find the courage and wisdom to restructure it.

Today, systems thinking is needed more than ever because we are becoming overwhelmed by complexity. Perhaps for the first time in history, humankind has the capacity to create far more information than anyone can absorb, to foster far greater interdependency than anyone can manage, and to accelerate change far faster than anyone's ability to keep pace. All around us are examples of systemic breakdowns—global warming, climate change, the international drug trade, trade and budget deficits—problems that have no simple local cause. Similarly, organizations break down, despite individual brilliance and innovative products, because they are unable to pull their diverse functions and talents into a productive whole. Complexity can easily undermine confidence and responsibility—as in the frequent refrain, "It's all too complex for me," or "There's nothing I can do. It's the system." Systems thinking is the antidote to this sense of helplessness that many feel as we enter the age of interdependence. It is a discipline for seeing the structures that underlie complex situations, and for discerning high from low leverage change. By seeing wholes, we learn how to foster health.

Obvious. Yet subversive. An old way of seeing. Yet somehow new. Comforting, in that the solutions are in our hands. Disturbing, because we must do things, or at least see things and think about things, in a different way.

This exploration is about that different way of seeing and thinking. It is intended for people who may be wary of the word "systems" and the field of systems analysis, even though they may have been doing systems thinking all their lives. Systems thinking has been developing throughout the twentieth century, first with concepts such as 'the ecosystem' (from the 1930s) and then in earnest after World War II. I have kept the discussion nontechnical because I want to show what a long way you can go toward understanding systems without turning to mathematics or computers.

I have made liberal use of diagrams and time graphs in this book because there is a problem in discussing systems only with words. Words and sentences must, by necessity, come only one at a time in linear, logical order. Systems happen all at once. They are connected not just in one direction, but in many directions simultaneously. To discuss them properly, it is necessary somehow to use a language that shares some of the same properties as the phenomena under discussion.

Pictures work for this language better than words, because you can see all the parts of a picture at once. I will build up systems pictures gradually, starting with very simple ones. I think you'll find that you can understand this graphical language easily.

I start with the basics: the definition of a system and a dissection of its parts (in a reductionist, unholistic way). Then I put the parts back together to show how they interconnect to make the basic operating unit of a system: the feedback loop.

Next I will introduce you to a systems zoo—a collection of some common and interesting types of systems. You'll see how a few of these creatures behave and why and where they can be found. You'll recognize them; they're all around you and even within you.

With a few of the zoo "animals"—a set of specific examples—as a foundation, I'll step back and talk about how and why systems work so beautifully and the reasons why they so often surprise and confound us. I'll talk about why everyone or everything in a system can act dutifully and rationally, yet all these well-meaning actions too often add up to a perfectly terrible result. And why things so often happen much faster or slower than everyone thinks they will. And why you can be doing something that has always worked and suddenly discover, to your great disappointment, that your action no longer works. And why a system might suddenly, and without warning, jump into a kind of behavior you've never seen before.

That discussion will lead us to look at the common problems that the systems-thinking community has stumbled upon over and over again through working in corporations and governments, economies and ecosystems, physiology and psychology. "There's another case of the tragedy of the commons," we find ourselves saying as we look at an allocation system for sharing water resource among communities or financial resources among schools. Or we identify "eroding goals" as we study the business rules and incentives that help or hinder the development of new technologies. Or we see "policy resistance" as we examine decision-making power and the nature of relationships in a family, a community, or a nation. Or we witness "addiction"—which can be caused by many more agents than caffeine, alcohol, nicotine, and narcotics.

Systems thinkers call these common structures that produce characteristic behaviors "archetypes." When I first planned this book, I called them "system traps." Then I added the words "and opportunities," because these archetypes, which are responsible for some of the most intransigent and potentially dangerous problems, also can be transformed, with a little systems understanding, to produce much more desirable behaviors.

From this understanding I move into what you and I can do about restructuring the systems we live within. We can learn how to look for leverage points for change.

I conclude with the largest lessons of all, the ones derived from the wisdom shared by most systems thinkers I know. For those who want to explore systems thinking further, the Appendix provides ways to dig deeper into the subject with a glossary, a bibliography of systems thinking resources, a summary list of systems principles, and equations for the models described in Part One.

When our small research group moved from MIT to Dartmouth College years ago, one of the Dartmouth engineering professors watched us in seminars for a while, and then dropped by our offices. "You people are different," he said. "You ask different kinds of questions. You see things I don't see. Somehow you come at the world in a different way. How? Why?"

That's what I hope to get across throughout this exploration, but especially in its conclusion. I don't think the systems way of seeing is better than the reductionist way of thinking. I think it's complementary, and therefore revealing. You can see some things through the lens of the human eye, other things through the lens of a microscope, others through the lens of a telescope, and still others through the lens of systems theory. Everything seen through each kind of lens is actually there. Each way of seeing allows our knowledge of the wondrous world in which we live to become a little more complete.

At a time when the world is more messy, more crowded, more interconnected, more interdependent, and more rapidly changing than ever before, the more ways of seeing, the better. The systems-thinking lens allows us to reclaim our intuition about whole systems and

• hone our abilities to understand parts,
• see interconnections,
• ask "what-if" questions about possible future behaviors, and
• be creative and courageous about system redesign.

Then we can use our insights to make a difference in ourselves and our world.

---

The Basics

I have yet to see any problem, however complicated, which, when looked at in the right way, did not become still more complicated. —POUL ANDERSON

More Than the Sum of Its Parts

A system isn't just any old collection of things. A system is an interconnected set of elements that is coherently organized in a way that achieves something. If you look at that definition closely for a minute, you can see that a system must consist of three kinds of things: elements, interconnections, and a function or purpose. The whole is greater than the sum of its parts. Therefore, understanding how systems work is of great advantage to anyone trying to assess and solve challenges. By reframing your perspective – you are no longer looking at a broken machine but a damaged system – you are better equipped to understand that each part is significant in its own way, while the whole working together is significant in the highest way.

For example, the elements of your digestive system include teeth, enzymes, stomach, and intestines. They are interrelated through the physical flow of food, and through an elegant set of regulating chemical signals. The function of this system is to break down food into its basic nutrients and to transfer those nutrients into the bloodstream (another system), while discarding unusable wastes.

A football team is a system with elements such as players, coach, field, and ball. Its interconnections are the rules of the game, the coach's strategy, the players' communications, and the laws of physics that govern the motions of ball and players. The purpose of the team is to win games, or have fun, or get exercise, or make millions of dollars, or all of the above.

A school is a system. So is a city, and a factory, and a corporation, and a national economy. An animal is a system. A tree is a system, and a forest is a larger system that encompasses subsystems of trees and animals. The earth is a system. So is the solar system; so is a galaxy. Systems can be embedded in systems, which are embedded in yet other systems.

Is there anything that is not a system? Yes—a conglomeration without any particular interconnections or function. Sand scattered on a road by happenstance is not, itself, a system. You can add sand or take away sand and you still have just sand on the road. Arbitrarily add or take away football players, or pieces of your digestive system, and you quickly no longer have the same system.

When a living creature dies, it loses its "system-ness." The multiple interrelations that held it together no longer function, and it dissipates, although its material remains part of a larger food-web system. Some people say that an old city neighborhood where people know each other and communicate regularly is a social system, and that a new apartment block full of strangers is not—not until new relationships arise and a system forms.

A system is more than the sum of its parts. It may exhibit adaptive, dynamic, goal-seeking, self-preserving, and sometimes evolutionary behavior.

You can see from these examples that there is an integrity or wholeness about a system and an active set of mechanisms to maintain that integrity. Systems can change, adapt, respond to events, seek goals, mend injuries, and attend to their own survival in lifelike ways, although they may contain or consist of nonliving things. Systems can be self-organizing, and often are self-repairing over at least some range of disruptions. They are resilient, and many of them are evolutionary. Out of one system other completely new, never-before-imagined systems can arise.

One dilemma with systems analysis is where to draw the boundary around any investigation or inquiry. Day‑to‑day problems we all know about, such as electric vehicles and EV charging points, can be so much better understood when you approach the problems from a systemic perspective. But deciding what to include and what to exclude from your analysis is both an art and a science.

Look Beyond the Players to the Rules of the Game

You think that because you understand "one" that you must therefore understand "two" because one and one make two. But you forget that you must also understand "and." —Sufi teaching story

The elements of a system are often the easiest parts to notice, because many of them are visible, tangible things. The elements that make up a tree are roots, trunk, branches, and leaves. If you look more closely, you see specialized cells: vessels carrying fluids up and down, chloroplasts, and so on. The system called a university is made up of buildings, students, professors, administrators, libraries, books, computers—and I could go on and say what all those things are made up of. Elements do not have to be physical things. Intangibles are also elements of a system. In a university, school pride and academic prowess are two intangibles that can be very important elements of the system. Once you start listing the elements of a system, there is almost no end to the process. You can divide elements into sub-elements and then sub-sub-elements. Pretty soon you lose sight of the system. As the saying goes, you can't see the forest for the trees.

Before going too far in that direction, it's a good idea to stop dissecting out elements and to start looking for the interconnections, the relationships that hold the elements together.

The interconnections in the tree system are the physical flows and chemical reactions that govern the tree's metabolic processes—the signals that allow one part to respond to what is happening in another part. For example, as the leaves lose water on a sunny day, a drop in pressure in the water-carrying vessels allows the roots to take in more water. Conversely, if the roots experience dry soil, the loss of water pressure signals the leaves to close their pores, so as not to lose even more precious water.

As the days get shorter in the temperate zones, a deciduous tree puts forth chemical messages that cause nutrients to migrate out of the leaves into the trunk and roots and that weaken the stems, allowing the leaves to fall. There even seem to be messages that cause some trees to make repellent chemicals or harder cell walls if just one part of the plant is attacked by insects. No one understands all the relationships that allow a tree to do what it does. That lack of knowledge is not surprising. It's easier to learn about a system's elements than about its interconnections.

In the university system, interconnections include the standards for admission, the requirements for degrees, the examinations and grades, the budgets and money flows, the gossip, and most important, the communication of knowledge that is, presumably, the purpose of the whole system.

Many of the interconnections in systems operate through the flow of information. Information holds systems together and plays a great role in determining how they operate.

Some interconnections in systems are actual physical flows, such as the water in the tree's trunk or the students progressing through a university. Many interconnections are flows of information—signals that go to decision points or action points within a system. These kinds of interconnections are often harder to see, but the system reveals them to those who look. Students may use informal information about the probability of getting a good grade to decide what courses to take. A consumer decides what to buy using information about his or her income, savings, credit rating, stock of goods at home, prices, and availability of goods for purchase. Governments need information about kinds and quantities of water pollution before they can create sensible regulations to reduce that pollution. (Note that information about the existence of a problem may be necessary but not sufficient to trigger action—information about resources, incentives, and consequences is necessary too.)

If information-based relationships are hard to see, functions or purposes are even harder. A system's function or purpose is not necessarily spoken, written, or expressed explicitly, except through the operation of the system. The best way to deduce the system's purpose is to watch for a while to see how the system behaves.

If a frog turns right and catches a fly, and then turns left and catches a fly, and then turns around backward and catches a fly, the purpose of the frog has to do not with turning left or right or backward but with catching flies. If a government proclaims its interest in protecting the environment but allocates little money or effort toward that goal, environmental protection is not, in fact, the government's purpose. Purposes are deduced from behavior, not from rhetoric or stated goals.

The function of a thermostat-furnace system is to keep a building at a given temperature. One function of a plant is to bear seeds and create more plants. One purpose of a national economy is, judging from its behavior, to keep growing larger. An important function of almost every system is to ensure its own perpetuation.

System purposes need not be human purposes and are not necessarily those intended by any single actor within the system. In fact, one of the most frustrating aspects of systems is that the purposes of subunits may add up to an overall behavior that no one wants. No one intends to produce a society with rampant drug addiction and crime, but consider the combined purposes and consequent actions of the actors involved:

• desperate people who want quick relief from psychological pain
• farmers, dealers, and bankers who want to earn money
• pushers who are less bound by civil law than are the police who oppose them
• governments that make harmful substances illegal and use police power to interdict them
• wealthy people living in close proximity to poor people
• nonaddicts who are more interested in protecting themselves than in encouraging recovery of addicts

Altogether, these make up a system from which it is extremely difficult to eradicate drug addiction and crime.

Systems can be nested within systems. Therefore, there can be purposes within purposes. The purpose of a university is to discover and preserve knowledge and pass it on to new generations. Within the university, the purpose of a student may be to get good grades, the purpose of a professor may be to get tenure, the purpose of an administrator may be to balance the budget. Any of those sub-purposes could come into conflict with the overall purpose—the student could cheat, the professor could ignore the students in order to publish papers, the administrator could balance the budget by firing professors. Keeping sub-purposes and overall system purposes in harmony is an essential function of successful systems. I'll get back to this point later when we come to hierarchies.

You can understand the relative importance of a system's elements, interconnections, and purposes by imagining them changed one by one. Changing elements usually has the least effect on the system. If you change all the players on a football team, it is still recognizably a football team. (It may play much better or much worse—particular elements in a system can indeed be important.) A tree changes its cells constantly, its leaves every year or so, but it is still essentially the same tree. Your body replaces most of its cells every few weeks, but it goes on being your body. The university has a constant flow of students and a slower flow of professors and administrators, but it is still a university. In fact it is still the same university, distinct in subtle ways from others, just as General Motors and the U.S. Congress somehow maintain their identities even though all their members change. A system generally goes on being itself, changing only slowly if at all, even with complete substitutions of its elements—as long as its interconnections and purposes remain intact.

The least obvious part of the system, its function or purpose, is often the most crucial determinant of the system's behavior.

If the interconnections change, the system may be greatly altered. It may even become unrecognizable, even though the same players are on the team. Change the rules from those of football to those of basketball, and you've got, as they say, a whole new ball game. If you change the interconnections in the tree—say that instead of taking in carbon dioxide and emitting oxygen, it does the reverse—it would no longer be a tree. (It would be an animal.) If in a university the students graded the professors, or if arguments were won by force instead of reason, the place would need a different name. It might be an interesting organization, but it would not be a university. Changing interconnections in a system can change it dramatically.

Changes in function or purpose also can be drastic. What if you keep the players and the rules but change the purpose—from winning to losing, for example? What if the function of a tree were not to survive and reproduce but to capture all the nutrients in the soil and grow to unlimited size? People have imagined many purposes for a university besides disseminating knowledge—making money, indoctrinating people, winning football games. A change in purpose changes a system profoundly, even if every element and interconnection remains the same.

To ask whether elements, interconnections, or purposes are most important in a system is to ask an unsystemic question. All are essential. All interact. All have their roles. But the least obvious part of the system, its function or purpose, is often the most crucial determinant of the system's behavior. Interconnections are also critically important. Changing relationships usually changes system behavior. The elements, the parts of systems we are most likely to notice, are often (not always) least important in defining the unique characteristics of the system—unless changing an element also results in changing relationships or purpose.

Changing just one leader at the top—from a Brezhnev to a Gorbachev, or from a Carter to a Reagan—may or may not turn an entire nation in a new direction, though its land, factories, and hundreds of millions of people remain exactly the same. A leader can make that land and those factories and people play a different game with new rules, or can direct the play toward a new purpose.

And conversely, because land, factories, and people are long-lived, slowly changing, physical elements of the system, there is a limit to the rate at which any leader can turn the direction of a nation.

Bathtubs 101—Understanding System Behavior over Time

A stock is the foundation of any system. Stocks are the elements of the system that you can see, feel, count, or measure at any given time. A system stock is just what it sounds like: a store, a quantity, an accumulation of material or information that has built up over time. It may be the water in a bathtub, a population, the books in a bookstore, the wood in a tree, the money in a bank, your own self-confidence. A stock does not have to be physical. Your reserve of good will toward others or your supply of hope that the world can be better are both stocks.

A stock is the memory of the history of changing flows within the system.

Stocks change over time through the actions of a flow. Flows are filling and draining, births and deaths, purchases and sales, growth and decay, deposits and withdrawals, successes and failures. A stock, then, is the present memory of the history of changing flows within the system.

For example, an underground mineral deposit is a stock, out of which comes a flow of ore through mining. The inflow of ore into a mineral deposit is minute in any time period less than eons.

Water in a reservoir behind a dam is a stock, into which flow rain and river water, and out of which flows evaporation from the reservoir's surface as well as the water discharged through the dam.

The volume of wood in the living trees in a forest is a stock. Its inflow is the growth of the trees. Its outflows are the natural deaths of trees and the harvest by loggers. The logging harvest flows into another stock, perhaps an inventory of lumber at a mill. Wood flows out of the inventory stock as lumber is sold to customers.

If you understand the dynamics of stocks and flows—their behavior over time—you understand a good deal about the behavior of complex systems. And if you have had much experience with a bathtub, you understand the dynamics of stocks and flows.

Imagine a bathtub filled with water, with its drain plugged up and its faucets turned off—an unchanging, undynamic, boring system. Now mentally pull the plug. The water runs out, of course. The level of water in the tub goes down until the tub is empty.

Now imagine starting again with a full tub, and again open the drain, but this time, when the tub is about half empty, turn on the inflow faucet so the rate of water flowing in is just equal to that flowing out. What happens? The amount of water in the tub stays constant at whatever level it had reached when the inflow became equal to the outflow. It is in a state of dynamic equilibrium—its level does not change, although water is continuously flowing through it.

Imagine turning the inflow on somewhat harder while keeping the outflow constant. The level of water in the tub slowly rises. If you then turn the inflow faucet down again to match the outflow exactly, the water in the tub will stop rising. Turn it down some more, and the water level will fall slowly.

This model of a bathtub is a very simple system with just one stock, one inflow, and one outflow. Over the time period of interest (minutes), I have assumed that evaporation from the tub is insignificant, so I have not included that outflow. All models, whether mental models or mathematical models, are simplifications of the real world. You know all the dynamic possibilities of this bathtub. From it you can deduce several important principles that extend to more complicated systems:

• As long as the sum of all inflows exceeds the sum of all outflows, the level of the stock will rise.
• As long as the sum of all outflows exceeds the sum of all inflows, the level of the stock will fall.
• If the sum of all outflows equals the sum of all inflows, the stock level will not change; it will be held in dynamic equilibrium at whatever level it happened to be when the two sets of flows became equal.

The human mind seems to focus more easily on stocks than on flows. On top of that, when we do focus on flows, we tend to focus on inflows more easily than on outflows. Therefore, we sometimes miss seeing that we can fill a bathtub not only by increasing the inflow rate, but also by decreasing the outflow rate. Everyone understands that you can prolong the life of an oil-based economy by discovering new oil deposits. It seems to be harder to understand that the same result can be achieved by burning less oil. A breakthrough in energy efficiency is equivalent, in its effect on the stock of available oil, to the discovery of a new oil field—although different people profit from it.

A stock can be increased by decreasing its outflow rate as well as by increasing its inflow rate. There's more than one way to fill a bathtub!

Similarly, a company can build up a larger workforce by more hiring, or it can do the same thing by reducing the rates of quitting and firing. These two strategies may have very different costs. The wealth of a nation can be boosted by investment to build up a larger stock of factories and machines. It also can be boosted, often more cheaply, by decreasing the rate at which factories and machines wear out, break down, or are discarded.

You can adjust the drain or faucet of a bathtub—the flows—abruptly, but it is much more difficult to change the level of water—the stock—quickly. Water can't run out the drain instantly, even if you open the drain all the way. The tub can't fill up immediately, even with the inflow faucet on full blast. A stock takes time to change, because flows take time to flow. That's a vital point, a key to understanding why systems behave as they do. Stocks usually change slowly. They can act as delays, lags, buffers, ballast, and sources of momentum in a system. Stocks, especially large ones, respond to change, even sudden change, only by gradual filling or emptying.

Stocks generally change slowly, even when the flows into or out of them change suddenly. Therefore, stocks act as delays or buffers or shock absorbers in systems.

People often underestimate the inherent momentum of a stock. It takes a long time for populations to grow or stop growing, for wood to accumulate in a forest, for a reservoir to fill up, for a mine to be depleted. An economy cannot build up a large stock of functioning factories and highways and electric plants overnight, even if a lot of money is available. Once an economy has a lot of oil-burning furnaces and automobile engines, it cannot change quickly to furnaces and engines that burn a different fuel, even if the price of oil suddenly changes. It has taken decades to accumulate the stratospheric pollutants that destroy the earth's ozone layer; it will take decades for those pollutants to be removed.

Changes in stocks set the pace of the dynamics of systems. Industrialization cannot proceed faster than the rate at which factories and machines can be constructed and the rate at which human beings can be educated to run and maintain them. Forests can't grow overnight. Once contaminants have accumulated in groundwater, they can be washed out only at the rate of groundwater turnover, which may take decades or even centuries.

The time lags that come from slowly changing stocks can cause problems in systems, but they also can be sources of stability. Soil that has accumulated over centuries rarely erodes all at once. A population that has learned many skills doesn't forget them immediately. You can pump groundwater faster than the rate it recharges for a long time before the aquifer is drawn down far enough to be damaged. The time lags imposed by stocks allow room to maneuver, to experiment, and to revise policies that aren't working.

If you have a sense of the rates of change of stocks, you don't expect things to happen faster than they can happen. You don't give up too soon. You can use the opportunities presented by a system's momentum to guide it toward a good outcome—much as a judo expert uses the momentum of an opponent to achieve his or her own goals.

There is one more important principle about the role of stocks in systems, a principle that will lead us directly to the concept of feedback. The presence of stocks allows inflows and outflows to be independent of each other and temporarily out of balance with each other.

Stocks allow inflows and outflows to be decoupled and to be independent and temporarily out of balance with each other.

It would be hard to run an oil company if gasoline had to be produced at the refinery at exactly the rate the cars were burning it. It isn't feasible to harvest a forest at the precise rate at which the trees are growing. Gasoline in storage tanks and wood in the forest are both stocks that permit life to proceed with some certainty, continuity, and predictability, even though flows vary in the short term.

Human beings have invented hundreds of stock-maintaining mechanisms to make inflows and outflows independent and stable. Reservoirs enable residents and farmers downriver to live without constantly adjusting their lives and work to a river's varying flow, especially its droughts and floods. Banks enable you temporarily to earn money at a rate different from how you spend. Inventories of products along a chain from distributors to wholesalers to retailers allow production to proceed smoothly although customer demand varies, and allow customer demand to be filled even though production rates vary.

Most individual and institutional decisions are designed to regulate the levels in stocks. If inventories rise too high, then prices are cut or advertising budgets are increased, so that sales will go up and inventories will fall. If the stock of food in your kitchen gets low, you go to the store. As the stock of growing grain rises or fails to rise in the fields, farmers decide whether to apply water or pesticide, grain companies decide how many barges to book for the harvest, speculators bid on future values of the harvest, cattle growers build up or cut down their herds. Water levels in reservoirs cause all sorts of corrective actions if they rise too high or fall too low. The same can be said for the stock of money in your wallet, the oil reserves owned by an oil company, the pile of woodchips feeding a paper mill, and the concentration of pollutants in a lake.

People monitor stocks constantly and make decisions and take actions designed to raise or lower stocks or to keep them within acceptable ranges. Those decisions add up to the ebbs and flows, successes and problems, of all sorts of systems. Systems thinkers see the world as a collection of stocks along with the mechanisms for regulating the levels in the stocks by manipulating flows.

That means system thinkers see the world as a collection of "feedback processes."

How the System Runs Itself—Feedback

When a stock grows by leaps and bounds or declines swiftly or is held within a certain range no matter what else is going on around it, it is likely that there is a control mechanism at work. In other words, if you see a behavior that persists over time, there is likely a mechanism creating that consistent behavior. That mechanism operates through a feedback loop. It is the consistent behavior pattern over a long period of time that is the first hint of the existence of a feedback loop.

A feedback loop is formed when changes in a stock affect the flows into or out of that same stock. A feedback loop can be quite simple and direct. Think of an interest-bearing savings account in a bank. The total amount of money in the account (the stock) affects how much money comes into the account as interest. That is because the bank has a rule that the account earns a certain percent interest each year. The total dollars of interest paid into the account each year (the flow in) is not a fixed amount, but varies with the size of the total in the account.

You experience another fairly direct kind of feedback loop when you get your bank statement for your checking account each month. As your level of available cash in the checking account (a stock) goes down, you may decide to work more hours and earn more money. The money entering your bank account is a flow that you can adjust in order to increase your stock of cash to a more desirable level. If your bank account then grows very large, you may feel free to work less (decreasing the inflow). This kind of feedback loop is keeping your level of cash available within a range that is acceptable to you. You can see that adjusting your earnings is not the only feedback loop that works on your stock of cash. You also may be able to adjust the outflow of money from your account, for example. You can imagine an outflow-adjusting feedback loop for spending.

Feedback loops can cause stocks to maintain their level within a range or grow or decline. In any case, the flows into or out of the stock are adjusted because of changes in the size of the stock itself. Whoever or whatever is monitoring the stock's level begins a corrective process, adjusting rates of inflow or outflow (or both) and so changing the stock's level. The stock level feeds back through a chain of signals and actions to control itself.

Not all systems have feedback loops. Some systems are relatively simple open-ended chains of stocks and flows. The chain may be affected by outside factors, but the levels of the chain's stocks don't affect its flows. However, those systems that contain feedback loops are common and may be quite elegant or rather surprising, as we shall see.

A feedback loop is a closed chain of causal connections from a stock, through a set of decisions or rules or physical laws or actions that are dependent on the level of the stock, and back again through a flow to change the stock.

Systems thinking and causal loop analysis highlight how problematic it is to imagine one root cause. Instead of a single linear chain of causation, feedback loops reveal circular, interconnected patterns of influence. When you conduct an analysis using a systems thinking perspective, you should expect that some lines of inquiry may raise questions about leaders' role in problems which they may think should be attributed to someone else. The system's behavior emerges from its structure, not from any single actor's intentions.

Two fundamental types of feedback operate in systems. Positive feedback amplifies an effect, while negative dampens it. The stock market is an example of positive feedback: the market falls causing a sell-off, which creates a ripple effect of further sell-off and price declines. The opposite occurs in a stock market bubble. A thermostat is an example of negative feedback. We will explore both types in detail.

Stabilizing Loops—Balancing Feedback

One common kind of feedback loop stabilizes the stock level, as in the checking-account example. The stock level may not remain completely fixed, but it does stay within an acceptable range. What follows are some more stabilizing feedback loops that may be familiar to you.

If you're a coffee drinker, when you feel your energy level run low, you may grab a cup of hot black stuff to perk you up again. You, as the coffee drinker, hold in your mind a desired stock level (energy for work). The purpose of this caffeine-delivery system is to keep your actual stock level near or at your desired level. (You may have other purposes for drinking coffee as well: enjoying the flavor or engaging in a social activity.) It is the gap, the discrepancy, between your actual and desired levels of energy for work that drives your decisions to adjust your daily caffeine intake.

Notice that feedback loops often can operate in two directions. In this case, the feedback loop can correct an oversupply as well as an undersupply. If you drink too much coffee and find yourself bouncing around with extra energy, you'll lay off the caffeine for a while. High energy creates a discrepancy that says "too much," which then causes you to reduce your coffee intake until your energy level settles down. The diagram is intended to show that the loop works to drive the stock of energy in either direction.

I could have shown the inflow of energy coming from a cloud, but instead I made the system diagram slightly more complicated. Remember—all system diagrams are simplifications of the real world. We each choose how much complexity to look at. In this example, I drew another stock—the stored energy in the body that can be activated by the caffeine. I did that to indicate that there is more to the system than one simple loop. As every coffee drinker knows, caffeine is only a short-term stimulant. It lets you run your motor faster, but it doesn't refill your personal fuel tank. Eventually the caffeine high wears off, leaving the body more energy-deficient than it was before. That drop could reactivate the feedback loop and generate another trip to the coffee pot. (See the discussion of addiction later in this book.) Or it could activate some longer-term and healthier feedback responses: Eat some food, take a walk, get some sleep.

This kind of stabilizing, goal-seeking, regulating loop is called a balancing feedback loop, so I put a B inside the loop in the diagram. Balancing feedback loops are goal-seeking or stability-seeking. Each tries to keep a stock at a given value or within a range of values. A balancing feedback loop opposes whatever direction of change is imposed on the system. If you push a stock too far up, a balancing loop will try to pull it back down. If you shove it too far down, a balancing loop will try to bring it back up.

Here's another balancing feedback loop that involves coffee, but one that works through physical law rather than human decision. A hot cup of coffee will gradually cool down to room temperature. Its rate of cooling depends on the difference between the temperature of the coffee and the temperature of the room. The greater the difference, the faster the coffee will cool. The loop works the other way too—if you make iced coffee on a hot day, it will warm up until it has the same temperature as the room. The function of this system is to bring the discrepancy between coffee's temperature and room's temperature to zero, no matter what the direction of the discrepancy.

Starting with coffee at different temperatures, from just below boiling to just above freezing, the graphs show what will happen to the temperature over time (if you don't drink the coffee). You can see the "homing" behavior of a balancing feedback loop. Whatever the initial value of the system stock (coffee temperature in this case), whether it is above or below the "goal" (room temperature), the feedback loop brings it toward the goal. The change is faster at first, and then slower, as the discrepancy between the stock and the goal decreases.

This behavior pattern—gradual approach to a system-defined goal—also can be seen when a radioactive element decays, when a missile finds its target, when an asset depreciates, when a reservoir is brought up or down to its desired level, when your body adjusts its blood-sugar concentration, when you pull your car to a stop at a stoplight. You can think of many more examples. The world is full of goal-seeking feedback loops.

Balancing feedback loops are equilibrating or goal-seeking structures in systems and are both sources of stability and sources of resistance to change.

The presence of a feedback mechanism doesn't necessarily mean that the mechanism works well. The feedback mechanism may not be strong enough to bring the stock to the desired level. Feedbacks—the interconnections, the information part of the system—can fail for many reasons. Information can arrive too late or at the wrong place. It can be unclear or incomplete or hard to interpret. The action it triggers may be too weak or delayed or resource-constrained or simply ineffective. The goal of the feedback loop may never be reached by the actual stock. But in the simple example of a cup of coffee, the drink eventually will reach room temperature.

Runaway Loops—Reinforcing Feedback

The second kind of feedback loop is amplifying, reinforcing, self-multiplying, snowballing—a vicious or virtuous circle that can cause healthy growth or runaway destruction. It is called a reinforcing feedback loop, and will be noted with an R in the diagrams. It generates more input to a stock the more that is already there (and less input the less that is already there). A reinforcing feedback loop enhances whatever direction of change is imposed on it.

For example:

• When we were kids, the more my brother pushed me, the more I pushed him back, so the more he pushed me back, so the more I pushed him back.
• The more prices go up, the more wages have to go up if people are to maintain their standards of living. The more wages go up, the more prices have to go up to maintain profits. This means that wages have to go up again, so prices go up again.
• The more rabbits there are, the more rabbit parents there are to make baby rabbits. The more baby rabbits there are, the more grow up to become rabbit parents, to have even more baby rabbits.
• The more soil is eroded from the land, the less plants are able to grow, so the fewer roots there are to hold the soil, so the more soil is eroded, so less plants can grow.
• The more I practice piano, the more pleasure I get from the sound, and so the more I play the piano, which gives me more practice.

The U.S.-U.S.S.R. arms race serves as a poignant macro-scale illustration of a reinforcing feedback loop in geopolitics. Each side perceived the other's arms buildup as a threat, which justified further arms buildup of its own—which the other side then perceived as a threat, justifying yet more buildup. This was a race to see who could get fastest to where no one wants to go. The roots of this tragic dynamic lay not in rival political ideologies, nor in particular weapons systems, but in a way of thinking both sides shared. The system's behavior emerged from its structure—a reinforcing feedback loop that neither side could escape until one concluded it was no longer willing to play. Today, a structurally similar dynamic can be observed in the so-called "war on terrorism," where each attack and counterattack fuels further rounds of escalation, governed by the same systemic logic that trapped the Cold War superpowers.

Reinforcing loops are found wherever a system element has the ability to reproduce itself or to grow as a constant fraction of itself. Those elements include populations and economies. Remember the example of the interest-bearing bank account? The more money you have in the bank, the more interest you earn, which is added to the money already in the bank, where it earns even more interest.

This is not simple linear growth. It is not constant over time. The growth of the bank account at lower interest rates may look linear in the first few years. But, in fact, growth goes faster and faster. The more is there, the more is added. This kind of growth is called "exponential." It's either good news or bad news, depending on what is growing—money in the bank, people with HIV/AIDS, pests in a cornfield, a national economy, or weapons in an arms race.

Reinforcing feedback loops are self-enhancing, leading to exponential growth or to runaway collapses over time. They are found whenever a stock has the capacity to reinforce or reproduce itself.

In the economy, the more machines and factories (collectively called "capital") you have, the more goods and services ("output") you can produce. The more output you can produce, the more you can invest in new machines and factories. The more you make, the more capacity you have to make even more. This reinforcing feedback loop is the central engine of growth in an economy.

By now you may be seeing how basic balancing and reinforcing feedback loops are to systems. Sometimes I challenge my students to try to think of any human decision that occurs without a feedback loop—that is, a decision that is made without regard to any information about the level of the stock it influences. Try thinking about that yourself. The more you do, the more you'll begin to see feedback loops everywhere.

Watch out! If you see feedback loops everywhere, you're already in danger of becoming a systems thinker! Instead of seeing only how A causes B, you'll begin to wonder how B may also influence A—and how A might reinforce or reverse itself. When you hear in the nightly news that the Federal Reserve Bank has done something to control the economy, you'll also see that the economy must have done something to affect the Federal Reserve Bank. When someone tells you that population growth causes poverty, you'll ask yourself how poverty may cause population growth.

You'll be thinking not in terms of a static world, but a dynamic one. You'll stop looking for who's to blame; instead you'll start asking, "What's the system?" The concept of feedback opens up the idea that a system can cause its own behavior.

So far, I have limited this discussion to one kind of feedback loop at a time. Of course, in real systems feedback loops rarely come singly. They are linked together, often in fantastically complex patterns. A single stock is likely to have several reinforcing and balancing loops of differing strengths pulling it in several directions. A single flow may be adjusted by the contents of three or five or twenty stocks. It may fill one stock while it drains another and feeds into decisions that alter yet another. The many feedback loops in a system tug against each other, trying to make stocks grow, die off, or come into balance with each other. As a result, complex systems do much more than stay steady or explode exponentially or approach goals smoothly—as we shall see.

---

A Brief Visit to the Systems Zoo

One good way to learn something new is through specific examples rather than abstractions and generalities, so here are several common, simple but important examples of systems that are useful to understand in their own right and that will illustrate many general principles of complex systems.

This collection has some of the same strengths and weaknesses as a zoo. It gives you an idea of the large variety of systems that exist in the world, but it is far from a complete representation of that variety. It groups the animals by family—monkeys here, bears there (single-stock systems here, two-stock systems there)—so you can observe the characteristic behaviors of monkeys, as opposed to bears. But, like a zoo, this collection is too neat. To make the animals visible and understandable, it separates them from each other and from their normal concealing environment. Just as zoo animals more naturally occur mixed together in ecosystems, so the systems animals described here normally connect and interact with each other and with others not illustrated here—all making up the buzzing, hooting, chirping, changing complexity in which we live.

Ecosystems come later. For the moment, let's look at one system animal at a time.

One-Stock Systems

A Stock with Two Competing Balancing Loops—a Thermostat

You already have seen the "homing in" behavior of the goal-seeking balancing feedback loop—the coffee cup cooling. What happens if there are two such loops, trying to drag a single stock toward two different goals?

One example of such a system is the thermostat mechanism that regulates the heating of your room (or cooling, if it is connected to an air conditioner instead of a furnace). Like all models, the representation of a thermostat is a simplification of a real home heating system.

Whenever the room temperature falls below the thermostat setting, the thermostat detects a discrepancy and sends a signal that turns on the heat flow from the furnace, warming the room. When the room temperature rises again, the thermostat turns off the heat flow. This straightforward, stock-maintaining, balancing feedback loop is shown on the left side of the diagram. If there were nothing else in the system, and if you start with a cold room with the thermostat set at 18°C (65°F), the furnace would simply come on, the room would warm up, the furnace would go off, and the room would stay right at the target temperature.

However, this is not the only loop in the system. Heat also leaks to the outside. The outflow of heat is governed by the second balancing feedback loop, shown on the right side of the diagram. It is always trying to make the room temperature equal to the outside, just like a coffee cup cooling. If this were the only loop in the system (if there were no furnace), a warm room on a cold day would slowly cool down.

The assumption is that room insulation is not perfect, and so some heat leaks out of the warm room to the cool outdoors. The better the insulation, the slower the drop in temperature would be.

Now, what happens when these two loops operate at the same time? Assuming that there is sufficient insulation and a properly sized furnace, the heating loop dominates the cooling loop. You end up with a warm room, even starting with a cold room on a cold day.

As the room heats up, the heat flowing out of it increases, because there's a larger gap between inside and outside temperatures. But the furnace keeps putting in more heat than the amount that leaks out, so the room warms nearly to the target temperature. At that point, the furnace cycles off and on as it compensates for the heat constantly flowing out of the room.

The thermostat is set at 18°C (65°F) in this simulation, but the room temperature levels off slightly below 18°C (65°F). That's because of the leak to the outside, which is draining away some heat even as the furnace is getting the signal to put it back. This is a characteristic and sometimes surprising behavior of a system with competing balancing loops. It's like trying to keep a bucket full when there's a hole in the bottom. To make things worse, water leaking out of the hole is governed by a feedback loop; the more water in the bucket, the more the water pressure at the hole increases, so the flow out increases! In this case, we are trying to keep the room warmer than the outside and the warmer the room is, the faster it loses heat to the outside. It takes time for the furnace to correct for the increased heat loss—and in that minute still more heat leaks out. In a well-insulated house, the leak will be slower and so the house more comfortable than a poorly insulated one, even a poorly insulated house with a big furnace.

With home heating systems, people have learned to set the thermostat slightly higher than the actual temperature they are aiming at. Exactly how much higher can be a tricky question, because the outflow rate is higher on cold days than on warm days. But for thermostats this control problem isn't serious. You can muddle your way to a thermostat setting you can live with.

For other systems with this same structure of competing balancing loops, the fact that the stock goes on changing while you're trying to control it can create real problems. For example, suppose you're trying to maintain a store inventory at a certain level. You can't instantly order new stock to make up an immediately apparent shortfall. If you don't account for the goods that will be sold while you are waiting for the order to come in, your inventory will never be quite high enough. You can be fooled in the same way if you're trying to maintain a cash balance at a certain level, or the level of water in a reservoir, or the concentration of a chemical in a continuously flowing reaction system.

There's an important general principle here, and also one specific to the thermostat structure. First the general one: The information delivered by a feedback loop can only affect future behavior; it can't deliver the information, and so can't have an impact fast enough to correct behavior that drove the current feedback. A person in the system who makes a decision based on the feedback can't change the behavior of the system that drove the current feedback; the decisions he or she makes will affect only future behavior.

The information delivered by a feedback loop—even nonphysical feedback—can only affect future behavior; it can't deliver a signal fast enough to correct behavior that drove the current feedback. Even nonphysical information takes time to feedback into the system.

Why is that important? Because it means there will always be delays in responding. It says that a flow can't react instantly to a flow. It can react only to a change in a stock, and only after a slight delay to register the incoming information. In the bathtub, it takes a split second of time to assess the depth of the water and decide to adjust the flows. Many economic models make a mistake in this matter by assuming that consumption or production can respond immediately, say, to a change in price. That's one of the reasons why real economies tend not to behave exactly like many economic models.

The specific principle you can deduce from this simple system is that you must remember in thermostat-like systems to take into account whatever draining or filling processes are going on. If you don't, you won't achieve the target level of your stock. If you want your room temperature to be at 18°C (65°F), you have to set the thermostat a little above the desired temperature. If you want to pay off your credit card (or the national debt), you have to raise your repayment rate high enough to cover the charges you incur while you're paying (including interest). If you're gearing up your work force to a higher level, you have to hire fast enough to correct for those who quit while you are hiring. In other words, your mental model of the system needs to include all the important flows, or you will be surprised by the system's behavior.

A stock-maintaining balancing feedback loop must have its goal set appropriately to compensate for draining or inflowing processes that affect that stock. Otherwise, the feedback process will fall short of or exceed the target for the stock.

Before we leave the thermostat, we should see how it behaves with a varying outside temperature. Over a twenty-four-hour period of normal operation, with the outside temperature dipping well below freezing, the inflow of heat from the furnace nicely tracks the outflow of heat to the outside. The temperature in the room varies hardly at all once the room has warmed up.

Every balancing feedback loop has its breakdown point, where other loops pull the stock away from its goal more strongly than it can pull back. That can happen in this simulated thermostat system, if I weaken the power of the heating loop (a smaller furnace that cannot put out as much heat), or if I strengthen the power of the cooling loop (colder outside temperature, less insulation, or larger leaks). At very cold temperatures, the furnace just can't keep up with the heat drain. The loop that is trying to bring the room temperature down to the outside temperature dominates the system for a while. The room gets pretty uncomfortable!

See if you can follow, as time unfolds, how the variables relate to one another. At first, both the room and the outside temperatures are cool. The inflow of heat from the furnace exceeds the leak to the outside, and the room warms up. For an hour or two, the outside is mild enough that the furnace replaces most of the heat that's lost to the outside, and the room temperature stays near the desired temperature.

But as the outside temperature falls and the heat leak increases, the furnace cannot replace the heat fast enough. Because the furnace is generating less heat than is leaking out, the room temperature falls. Finally, the outside temperature rises again, the heat leak slows, and the furnace, still operating at full tilt, finally can pull ahead and start to warm the room again.

Just as in the rules for the bathtub, whenever the furnace is putting in more heat than is leaking out, the room temperature rises. Whenever the inflow rate falls behind the outflow rate, the temperature falls. If you study the system changes on this graph for a while and relate them to the feedback-loop diagram of this system, you'll get a good sense of how the structural interconnections of this system—its two feedback loops and how they shift in strength relative to each other—lead to the unfolding of the system's behavior over time.

---

Unintended Consequences and Systemic Effects

Every action has consequences—both intended and unintended. No matter how carefully we plan, we can't anticipate everything. Often we fail to consider what other events are likely to occur as a result of some action. Politics, medicine, welfare programs, technology, military actions, and laws all generate unintended consequences. Whatever we do (or don't do), there are multiple consequences, and they might not be what we expect.

Consider these examples of unexpected outcomes:

• Traffic engineers found that adding new routes (like a more direct road) could make traffic go slower. During merging, cars drive closer to one another and therefore slow down.
• Studies of car safety show that wearing seat belts makes drivers feel more secure, making them drive faster or more recklessly.
• A poor widow who kept a faithful hen decided to double the hen's allowance of barley, expecting two eggs a day instead of one. The hen grew fat and lazy and soon stopped laying altogether.

By solving one problem, we often generate another one—and sometimes create an even worse one. Consider the campus that had a mouse problem. The solution was to pay students one dollar for every dead mouse they delivered. It worked—until students began breeding mice in order to make more money.

Should good intentions eliminate bad consequences? No. Outcomes don't follow from intentions, and intentions by definition apply only to intended consequences. As Samuel Johnson said, "The road to hell is paved with good intentions." Isn't it more important to find out if the consequences are good rather than if the reasons are good? Good thinking is better than good intentions.

Medicare provides a classic example of systemic unintended consequences. All the actuarial studies showed the cost would be X, but the cost turned out to be more than ten times that amount. The planners didn't factor in the fact that effects have effects—they didn't think through the incentive effects of the way they were changing the rules. They created a system wherein they were reimbursing both doctors and hospitals on a cost-plus-percentage-of-cost basis. The moment they did that, hospitals and doctors found wonderful ways to talk patients into buying all kinds of care that got reimbursed—good for the hospital and good for the doctor, but bad for the patient and bad for the taxpayers.

Yet even good thinking can produce undesired consequences. An excess of what seems like professionalism can wind up hurting you horribly, precisely because those careful procedures themselves often lead to overconfidence in their outcome. Smart, hard-working people aren't exempt from professional disasters resulting from overconfidence. Often they just go aground on the more difficult voyages they choose to embark upon based on self-appraisals concluding that they have superior talents and methods. It is irritating that extra care in thinking isn't all good—that it also introduces extra error. But most good things have undesired side effects, and thinking is no exception.

One way to reduce unintended consequences is to stop focusing on isolated factors and instead consider how our actions affect the whole system.

---

Optimizing the Whole System

How a system behaves is a function of all the factors—human and non-human—that make up and influence the system. A business, for example, is a collection of parts that works as a complete system. There are variables like suppliers, employees, customers, demand, competition, and more. There are activities like purchasing, manufacturing, warehousing, logistics, and distribution. There are technological systems and equipment needed to conduct the business. All these parts work together, yet we often optimize one component at a time instead of optimizing the whole.

Consider TransCorp, which fired 200 people to cut costs. They forgot to consider how a change influences the whole system. Cost cutting doesn't automatically translate into higher value. The decision to fire people caused manufacturing and delivery problems, which in turn caused delays in customer production. This created a loss of customers and reputation. The end result was lower profits.

When TransCorp's sales volume dropped, someone suggested reducing prices: "It's a sure-fire way to win back customers and boost sales. We make up on volume what we lose on price and as a result we increase our market share. Profits will surely rise." But they forgot to think about all the factors that influenced the outcome. They didn't anticipate the consequences of cutting prices. Volume increases affect costs and behavior and need for investments in operating assets. There may also have been production problems due to technical constraints, or the price decrease wasn't enough to make customers switch from the competition. Worse, competitors matched the price cut, and TransCorp lost both volume and profits.

Try to optimize the whole and not a system's individual parts. Think through what other variables may change when we alter a factor in a system. Trace out the short and long-term consequences in numbers and effects of a proposed action to see if the net result agrees with our ultimate goal. Ask: What key factors influence the outcome of the system and how do these factors interact? What other things may change as a consequence of some action? Given these conditions, what likely consequences (wanted and unwanted) will the proposed action have on the system, considering all the relevant factors that influence or are part of the system? Will the net result be what we want?

---

The Reaction of Others

When thinking through consequences, consider what other people are likely to do. Since our interests may conflict with others, the final outcome of our decision often depends on what others will do. What other people do may depend on what they think we will do, their available choices, their interests, and how they are thinking—including their misjudgments. Humans don't always act rationally.

Game theory deals with what happens when individuals or groups of people interact with one another to achieve their goals. It is a study of conflict between thoughtful and potentially deceitful opponents. Factors that decide the final outcome of a negotiation include: the number of participants, whether the participants will meet again in the future, the time lapse between interactions, the degree of anonymity and communication, and the relative position of strength, which includes other options, back-up alternatives, and the need to reach an agreement.

---

The Winner's Curse

Several mining firms are bidding on the right to mine silver. No firm knows for certain how much silver there is and hence what the true value is. They each hire an expert to make an educated guess. By definition, these expert guesses will range from too low to too high. Some firm's expert will probably be close—but they won't win. The winning firm had the most optimistic estimate of the value, since the seller accepts the highest bid. But there was less silver in the mine than their expert guessed, and less value than what they paid for the rights. The winning bidder was cursed, since the bid was higher than the value.

The idea behind the Winner's Curse emerged when researchers studied companies bidding for oil fields. Their basic finding was that a lease winner tends to be the bidder who most overestimates reserve potential. Winning is an informative event, telling us whose estimate was most optimistic. When we place a bid on a house, company, project, or negotiate to buy something, we often don't realize what is implied by an acceptance of our offer—that we may have overestimated its value and therefore paid too much.

Research shows that the more bidders there are competing for a limited object, each having the same information, and the more uncertain its value is, the more likely we are to overpay. If our objective is to create value, the more bidders there are, the more conservative our bidding should be. This also implies that the less information we have compared to other bidders, or the more uncertain we are about the underlying value, the lower we should bid. If we participate in auctions, we must ascertain the true value of what's being sold or its value to us.

When we negotiate with one party and want an acceptance for an offer, the other party may have an informational advantage. The other party is most likely to accept our offer when it is least favorable to us, especially if it is a one-time relationship or if the other party is anonymous. Consider the seller's perspective. Ask: Why are they selling? How would I reason if I think it through from the viewpoint of the other person? Why would I make a better decision than someone who has all the information?

---

The Limits of Prediction in Complex Systems

The world is too complicated to predict all the effects of some action. Maybe a business can predict scenarios like reduction in demand and intensified competition, but some events, their timing, magnitude, or consequence, are impossible to anticipate. As Mark Twain said: "The art of prophecy is very difficult, especially with respect to the future."

It is hard to predict something when we don't (or can't) foresee or understand how an entire system works, what key variables are involved, their attributes, how they influence one another, and their impact. Even if we know the key variables, their values may be impossible to estimate. They may also change over time and be dependent on context. It may also be impossible to estimate how they will interact as a whole.

The more parts involved and the more they interact, the more can happen, and the harder it is to determine consequences of individual actions. The brain is an example of a complex system: a complex system is one in which smaller parts form a heterogeneous set of components which are more or less independent. But as these parts connect with each other in larger and larger aggregates, their functions tend to become integrated, yielding new functions that depend on such high-order integration.

As the number of variables grows, the number of possible interactions grows even faster. Assume that two subsystems, A and B, cause the behavior of a system. Each subsystem consists of five parts. If we only consider two-way interactions among parts, there are ten interactions between A-parts, ten between B-parts, and twenty-five interactions between A and B parts. This means that the behavior of the system is composed of fifty-five determinants. Only eighteen percent of all determinants derive from individual effects of parts in A and B, while about eighty-two percent derive from interactions. Now imagine a system where A and B each consists of one hundred parts. There are now over twenty thousand determinants, and ninety-nine percent of system determinants derive from interactions.

We often take too little notice of how variables interact. Take the economy as an example. There are many factors to consider, including interest rates, currency exchange rates, balance of trade figures, unemployment rates, consumer confidence, political factors, the stock market, business cycles, and biases. These factors are interconnected, and it is hard to tell which is most important. Add to this that people's behavior isn't fixed. We are emotional creatures, our preferences change, and we react to each other's actual or expected decisions. A prediction may also make us change our expectations and behavior, making the prediction more or less likely to come true.

Predictions about the future are often just projections of past curves and present trends. This is natural since our predictions about the future are made in the present. We therefore assume the future will be much like the present. But the future can't be known until it arrives. It is contingent on events we can't see. For example, who in the year 1900 could foresee events like World War I and II, the 1929 stock market crash, Chernobyl, or technologies like the television, laser, computer, internet, or the DVD? Many key inventions happened by accident and sagacity.

Economics isn't like physics. There are no reliable or precise formulas where we easily can fill in the values of various economic factors and then have the work done. Economics involves too complex a system—it should emulate physics' basic ethos, but its search for precision in physics-like formulas is almost always wrong in economics. To convert a model into a quantitative formula is to destroy its usefulness as an instrument of thought. We tend to put too much comfort in computer models and the precision they project. The precision they project is a chimera. In fact, such models can lull decision-makers into a false sense of security and thereby increase their chances of making a really huge mistake. The next time someone proposes an elegant model to manage risks and foretell odds, the next time a computer with a perfect memory of the past is said to quantify risks in the future, investors should run—and quickly—the other way.

That an event has happened many times before doesn't mean that it will continue to happen. And just because an event has never happened before doesn't mean it can't happen in the future. Take catastrophic events as an example. Who could have predicted the September 11, 2001 terrorist attack on the World Trade Center? Hijacking four planes simultaneously and using them to attack the U.S. was improbable. Yet it happened. Time and again, history has proved a very bad predictor of future events. This is because history never repeats itself; nothing in human society ever happens twice under exactly the same conditions or in exactly the same way.

Sometimes we can guess that certain things are bound to happen, but we can't predict when they will happen. Some things are possible to predict short-term but impossible to predict long-term. Small changes make a big difference over time. Long-term weather forecasts are an example. Many factors determine the weather—factors that can't be reliably measured ahead of time. Small changes in the temperature and pressure over the ocean can lead to large variations in the future development of storm systems. Weather predictions become more inexact the further out they are.

The difficulty lies in initial condition uncertainties and model error. Small errors in the initial values of variables can grow and produce errors in the forecast. There may also be gaps in the initial data. But even if we know the initial conditions perfectly, the models are not perfect. Small model errors, either in physics or numerics, can also grow and produce different states. Not all atmospheric processes are understood. Furthermore, all weather models operate on a finite grid or limited area, typically in the order of ten to one hundred kilometers, depending on the study area. This means that the numerical resolution and representation is finite. But many physical processes and features that influence the weather occur on smaller scales than those resolved by the grid. The model must treat or "parameterize" the effects of these "sub-grid" features on the resolved scale. These parameterizations are simplifications and approximations and may also account for many of the model errors. So even if we know all the principles behind the weather and what governs the atmosphere, fundamental limitations make it hard to make accurate predictions.

Many meteorologists have therefore given up on predicting whether it will or will not rain beyond a few days into the future. Instead, they have altered their approach and try to quantify the uncertainty in the prediction—"The probability of precipitation is twenty percent this Saturday." This uncertainty is small at short lead times and greater at longer lead times. When meteorologists make forecasts two weeks out, they look at the climatological frequency of precipitation, determined from a history of what has happened in the past.

All prediction is inherently uncertain, and we have a duty to tell people about the uncertainties of our predictions and our past error rates. As Albert Einstein wrote: "The right to search for truth... implies also a duty; one must not conceal any part of what one has recognized to be true."

---

Applying Systems Thinking in Organizations

There is an assortment of ways to actualize systems thinking inside an organization, and every organization must discover what fits well inside the organizational culture. However, there are vital aspects of systems thinking to consider:

• Occasions: Begin the practice of systems thinking by recounting the story of the present situation. It is imperative to hear many points of view, but don't jump to solutions right away. It may be necessary to provide a quick fix with the understanding that it is a stopgap measure not intended for long-term arrangements. In the interim, investigate the nature of the issue in greater detail.

• Patterns of behavior: Track the situation over time. Search for trends and patterns that go beneath the surface. Return to history if you have the data. Those patterns are key indicators for the system.

• Systemic structure: In the patterns you've discovered, look for interrelationships. Search for feedback balancing and strengthening, and identify system delays. Uncover the mental models driving those patterns.

• Mental models: Create new mental models to make system change. Be tolerant—delays will occur, and things could deteriorate before they show signs of improvement. Track the effects of the changes and assess them. Decide whether there are unintended outcomes and determine which adjustments are needed.

This process, though it may seem moderately direct, is far from it. There are numerous challenges in implementing systems thinking, the greatest of which are conventions and habits. Patterns of behavior are difficult to recognize and even harder to change, and we often feel pressure to act quickly with very little systemic data. Finding the balance between taking care of the present and creating the future is never simple.

Systems thinking is also a diagnostic tool. Like in the medical profession, accurate diagnosis accompanies successful treatment. In this sense, system thinking is a systematic method for a more thorough and reliable analysis of issues before acting. This helps us ask informed questions before we conclude.

Systems analysis requires shifting from events or data to behavioral trends over time and the processes underlying these trends and events. By identifying and modifying systems that are not well served—including our mental models and perceptions—we extend the options available and create more meaningful, long-term solutions for chronic issues.

Generally, the systems thinking perspective requires curiosity, insight, compassion, preference, and courage. This approach requires the ability to look at the situation in detail, realize that we are interconnected, understand that an issue often involves multiple solutions, and champion approaches that might not be common.

Systems thinking is the conceptual cornerstone that underlies a broader set of learning disciplines. It is a discipline for seeing the structures that underlie complex situations, and for discerning high from low leverage change. That is, by seeing wholes we learn how to foster health. All such learning disciplines are concerned with a shift of mind: from seeing parts to seeing wholes, from seeing people as helpless reactors to seeing them as active participants in shaping their reality, from reacting to the present to creating the future. Without systems thinking, there is neither the incentive nor the means to integrate these disciplines. As a cornerstone of how learning organizations think about their world, systems thinking offers a language that begins by restructuring how we think.

Organizations practicing systems thinking can:

• Develop new ways of looking at old issues.
• Incorporate new data more effectively.
• See interrelationships and cause and effect more clearly.
• Develop patience with implementing change and enduring delays.
• Step away from habitual pettiness toward shared responsibility.
• See the whole as opposed to the parts.

Systems thinking helps you assess why people act in certain ways or why things happen. It gives you the perspective to step back and make well-informed decisions, expanding your range of options. At last, organizations practicing systems thinking will be able to step up to that next level of thinking to solve issues and create their desired futures. They will have greater command over their fortune and be increasingly adept at responding to environmental changes.