“For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled.”

Richard Feynman


URBAN DYNAMICS by Jay Forrester

This study of urban dynamics was undertaken principally because of discoveries made in modeling the growth process of corporations. It has become clear that complex systems are counterintuitive. That is, they give indications that suggest corrective action which will often be ineffective or even adverse in its results. Very often one finds that the policies that have been adopted for correcting a difficulty are actually intensifying it rather than producing a solution.

Choosing an ineffective or detrimental policy for coping with a complex system is not a matter of random chance. The intuitive processes will select the wrong solution much more often than not. A complex system—a class to which a corporation, a city, an economy, or a government belong—behaves in many ways quite the opposite of the simple systems from which we have gained our experience.

Most of our intuitive responses have been developed in the context of what are technically called first-order, negative-feedback loops. Such a simple loop is goal-seeking and has only one important state variable. For example, warming one’s hands beside a stove can be approximated as a first-order, negative-feedback loop in which the purpose of the process is to obtain warmth without burning one’s hands. The principal state variable of the loop is the distance from the stove. If one is too close he burns his hands, if too far away he receives little heat. The intuitive lesson is that cause and effect are closely related in time and space. Temperature depends on the distance from the stove. Too much or too little heat is clearly related to the position of the hands. The relation of cause and effect is immediate and clear. Similarly, the simple feedback loops that govern walking, driving a car, or picking things up all train us to find cause and effect occurring at approximately the same moment and location.

But in complex systems cause and effect are often not closely related in either time or space. The structure of a complex system is not a simple feedback loop where one system state dominates the behavior. The complex system has a multiplicity of interacting feedback loops. Its internal rates of flow are controlled by nonlinear relationships. The complex system is of high order, meaning that there are many system states (or levels). It usually contains positive-feedback loops describing growth processes as well as negative, goal-seeking loops. In the complex system the cause of a difficulty may lie far back in time from the symptoms, or in a completely different and remote part of the system. In fact, causes are usually found, not in prior events, but in the structure and policies of the system.

To make matters still worse, the complex system is even more deceptive than merely hiding causes. In the complex system, when we look for a cause near in time and space to a symptom, we usually find what appears to be a plausible cause. But it is usually not the cause. The complex system presents apparent causes that are in fact coincident symptoms. The high degree of time correlation between variables in complex systems can lead us to make cause-and-effect associations between variables that are simply moving together as part of the total dynamic behavior of the system. Conditioned by our training in simple systems, we apply the same intuition to complex systems and are led into error. As a result we treat symptoms, not causes. The outcome lies between ineffective and detrimental. [p.p. 8-9]

URBAN DYNAMICS, Jay Forrester; 1969, ISBN 1-56327-058-7. Productivity Press, PO Box 13390, Portland, OR 97213-0390; Phone 503-235-0600 Fax: 503-235-0909;


by Jay Forrester

The intuitively obvious “solutions” to social problems are apt to fall into one of several traps set by the character of complex systems. First, an attempt to relieve one set of symptoms may only create a new mode of system behavior that also has unpleasant consequences. Second, the attempt to produce short-term improvement often sets the stage for a long-term degradation. Third, the local goals of a part of a system often conflict with the objectives of the larger system. Fourth, people are often led to intervene at points in a system where little leverage exists and where effort and money have but slight effect. These four ways in which a social system can mislead us will now be discussed in more detail.

The first kind of system response, wherein a new trouble appears as a result of solving the old, has already been illustrated in Chapter 4. The first mode exhibited by the world system in Section 4.2 developed a population decline and a falling quality of life because natural resources were being exhausted. By assuming a technological solution, the system was freed from dependence on resources in Section 4.3 but a worse kind of crisis developed—the runaway pollution mode. Then in Section 4.4 pollution was assumed to be controllable and crowding became severe enough to drive down the quality of life until population no longer rose. Then in a third attempt to relieve the pressures on the system, crowding was removed in Section 4.5 as a factor in birth and death rates, implying that psychological adjustments could be made to high-density living. The last and ultimate barrier, a food shortage, developed. Each attempt to eliminate a pressure within the system led to a new pressure.

As to the second characteristic, social systems usually exhibit fundamental conflict between the short-term and long-term consequences of a policy change. A policy which produces improvement in the short run is usually one which degrades the system in the long run. Short run and long run must be defined in terms of the dynamic responses in the system of interest. In corporate affairs, short run might be one to three years and long run beyond five years. In urban or national issues, short run could be a decade, while long run might be twenty years or more. In world dynamics, short run is several decades, and long run is fifty years to several centuries. Policies and programs which produce long-run improvement may initially depress the behavior of a system. This is especially treacherous. The short run is more visible and more compelling. It speaks loudly for immediate attention. But a series of actions all aimed at short-run improvement can eventually burden a system with long-run depressants so severe that even heroic short-run measures no longer suffice Many ,of the problems which the world faces today are the eventual result of short-run measures taken over the last century. Industrialization has raised the standard of living but now leads to pollution. Improvements in medicine and sanitation facilities improved health and reduced death rate but has led to the pressures from rising population. Greater capital investment and more intensive use of land in agriculture have increased food output in the short run but in the long run have destroyed the productivity of vast land areas by erosion and salt contamination.

As for the third characteristic of complex systems, there is usually a conflict between the goals of a subsystem and the welfare of the broader system. We see this in urban systems where the goal of a city is to expand and to raise its quality of life. But, as the, city strives to meet its goals, the nation encounters increased population, industrialization, pollution, and demand on food supply. The broader social system of a country or the world requires that the goals of the urban areas be curtailed and that the pressures of such curtailment become high enough to keep the urban areas and population within the bounds that are satisfactory to the larger system of which the city is a part. Nations strive for economic growth, higher standard of living, more food, and even increased population. If they succeed, as they may well do, the result will be to deepen the distress of the world as a whole and eventually to deepen the crises in the individual nations themselves. We are at the point where higher pressures in the present are necessary if insurmountable pressures are to be avoided in the future.

As a fourth characteristic, social systems are inherently insensitive to most policy changes that people select in an effort to alter behavior. In fact, a social system draws attention to the very points at which an attempt to intervene will fail. Human experience, which has been developed from contact with simple systems, leads us to look close to the symptoms of trouble for a cause. But when we look, we are misled because the social system presents us with an apparent cause that is plausible according to the lessons we have learned from simple systems, although this apparent cause is usually a coincident occurrence that, like the trouble symptom itself, is being produced by the feedback-loop dynamics of a larger system. In the world system, birth control is likely to be one of those apparent control points that in fact lack leverage for change. At the detailed demographic level, so many factors impinge on birth rate that an active program of birth control will be largely defeated by relaxation of controls that previously existed. As another example, efforts to reduce hunger by greater food production will generally fail unless there are simultaneous counterforces sufficiently large to prevent the population rising to match the new level of food availability.[p.p. 94-95]

WORLD DYNAMICS (second edition), Jay Forrester; Wright-Allen Press, 1971.



Why can public services not get ahead of demands? Why do the best of intentions for improving a city lead, instead, to greater social pressures, more commuting delays, increased drug addiction, higher crime rates, and greater welfare loads? The answer lies in what we have come to call the “attractiveness principle”.

The attractiveness principle states that, to any particular population class, all geographical areas tend to become equally attractive. Or perhaps more realistically stated, all areas tend to become equally unattractive. Why do all areas tend toward equal attractiveness? It is because people move from unattractive areas to areas of greater attractiveness. I use “attractiveness” to encompass every aspect of a city that contributes to its desirability or undesirability. Population movement is an equalizing process. As people move toward a more attractive area, they drive up prices and overload the job opportunities, the environmental capacity, the available housing, and the governmental services. In other words, rising population drives down all the characteristics of an area that made it initially attractive.

To illustrate the attractiveness principle, imagine for a moment the ideal city. Perhaps the ideal city would be one with readily available housing at low cost, a surplus of jobs at high wages, excellent schools, no smoke or pollution, housing located near one’s place of work, no crime, beautiful parks, cultural opportunities, and to this list the reader can add his own preferences. Suppose such a city existed. What would happen? It would be perceived as the ideal place to live. People from everywhere would move into the ideal city until the advantages had been so swamped by rising population that the city would offer no net attractiveness compared with other locations. [p.p. 275-276]


A whole set of pressures is now beginning to inhibit growth. The country faces an oil shortage. Pollution is no longer merely an industrial problem; to reduce pollution created by the individual, his automobile now has less performance, needs more maintenance, and has a higher gasoline consumption. As a result, automotive emissions have been somewhat reduced, but the national oil shortage has worsened and our dependence on other countries has increased. Pollution has also become a major issue in agriculture, as fertilizers and the wastes from animal feed lots pollute rivers and lakes. At the social level, rising crime, drug addiction, mental stress, and community breakdown are all exerting pressure against further growth. Many pressures are developing to stop growth; some we can influence, others we cannot. A most important question is how we would like to have the growth-suppressing pressures distributed.

Pressures to slow the growth process will continue to rise. They will tend to develop from every direction. Some of the pressures can be alleviated. But do we want to alleviate, where we can, the pressures arising from growth? Or, do those pressures serve a valuable purpose?

Unless ever-rising exponential growth can go on forever, and that is generally accepted as impossible, then some set of pressures will eventually stop growth. From whence should the growth-suppressing pressures come? Should the pressures be distributed throughout our society, or should they be concentrated in only a few places within our socioeconomic system? This choice between concentration or distribution of pressures is of the greatest importance. The question arises because we have the power to alleviate pressures in some sectors of the society but not in others. If we alleviate pressures where we can do so, growth will continue until it produces a further rise in the pressures that we cannot control. The way we react to present pressures determines the nature of future pressures.

One set of pressures, such as water shortages and crowded streets, can be alleviated by technological means. We are very good at handling technology, and we can eliminate those pressures if we wish. A second set of pressures, such as job availability, can be alleviated by economic means, and those we know less about but can still influence. A third set of pressures is of a social nature—crime, civil disorder, declining mental health, war, drug addiction, and the collapse of goals and values. These are the ultimate pressures with which we know not how to cope.

If we alleviate the pressures that can now be overcome, those pressures no longer contribute to slowing the growth process. Growth then continues until higher pressures are generated in other sectors. This process has been going on. The first pressures to arise were dealt with technologically by increasing building heights, improving transportation, bringing water from greater distances, developing new sources of energy, and improving medical treatment. As a result of such technological successes, growth continued until a variety of economic malfunctions began to appear—rising unemployment and welfare, worsening balance of trade, and inflation. To a small extent, the economic pressures have been alleviated and their consequences delayed. Growth has thereby continued until the social deterioration resulting from crowding and complexity has begun to manifest itself in serious ways.

In this sequence of technology—solving one problem only to produce an insolvable problem later—is buried the reasons for the antitechnology attitude that has begun to develop. In the past, technology appeared to be solving our problems. The technologists became self-confident. The public came to depend on them. The attitude took root that all problems could be solved by an ever-improving technology. Instead, the rising technology, with its consequent growth in population and industrialization, has carried the society to a complexity and a congestion that are producing rising symptoms of distress in the economic and social sectors. The very fact that technology succeeds in meeting its narrow goals produces greater difficulties in other parts of our social system. The antitechnology feeling grows because of the repeated cycle in which pressures develop, technology produces an excellent solution within its narrow self-perceived goals, the social system becomes more compressed and frustrating and the public perceives that the overall quality of life has failed to respond to the technical solution. The failure to satisfy society results because meeting the subgoals of the technologist is less and less likely to enhance the composite value of all the social goals. For each technical goal that is improved, some social or economic goal is forced to decline.

Growth has continued past the point where suboptimizing is satisfactory. Suboptimizing means the meeting of a local goal without attention to consequences in other parts of the system. During the past period of our industrial growth, the various facets of the technical-social-economic system were sufficiently uncoupled that suboptimizing was a satisfactory procedure for decentralization. Suboptimizing allowed different groups to pursue their own ends independently, with confidence that the total good would thereby improve. But as the system becomes more congested, the solution of one problem begins to create another. The blind pursuit of individually laudable goals can create a total system of degraded utility.


What does this discussion of technology and social goals mean for the American Public Works Association? It means that in the past those who dealt with the technological aspects of urban life were free to suboptimize. The public well-being was increased by the best possible job of drainage, waste disposal, transportation, water supply, and the construction of streets. But it is no longer true that improving each of these will always improve a city. By solving each of these technical problems the technologist risks becoming a party to increasing the population of a city and the densities of the population. He may start social processes that eventually reduce the quality of life. The public is recognizing that improved technology does not always bring an improved society. As a result, men who have sincerely dedicated their efforts to the public good, but perhaps have not foreseen the diversity of social consequences, have already begun to feel the backlash of public criticism.

So far I have developed several propositions. First, pressures are rising that will inevitably stop growth. Second, the national commitment to growth is too strong for the federal government to lead the country in a new direction until a broad constituency for changed expectations has been formed. Third, if the stress-creating nature of growth is to be recognized, and if experiments are to be carried out to find a satisfactory way of moving from growth to a society that can accept a future equilibrium, leadership must come from the local and state levels. Fourth, technical accomplishments no longer appear to be capable of solving our mounting social problems; instead, technology, as now being used, may often lead to expansion in urban population and living densities that become the cause of rising social difficulties. Fifth, all cities do at all times tend toward equal attractiveness in which no one city can remain significantly more attractive to in-migration than other cities. Given this set of propositions, what freedom of action is left to a city?

A city can choose, to a substantial extent, the mix of pressures under which it wishes to exist. There are many components of urban attractiveness, and if one of these is decreased, others can be improved. One cannot create the ideal city. But one can create certain ideal features if he is willing to compensate for them by intentionally allowing other features to worsen. In the past we have improved the technological aspects of cities and have thereby unintentionally contributed to the rise of many of the economic and social problems that plague cities today. There are many facets to a city. There are many things that the public and an urban administration can do. One thing they cannot do is produce the perfect city. They can, however, exercise a wide choice among imperfect cities.

I suggest that a valid goal for local urban leadership is to focus on improving the quality of life for the residents already in the city, at the same time protecting against the kind of growth that would overwhelm the gains. In short, one might raise the attractiveness of a city for the present residents while, at the same time, decreasing the attractiveness to those who might inundate the system from the outside.

Such statements, I recognize, lead to ethical and legal controversy. I am saying that a city should look after itself first. Its own welfare should come ahead of concern for others who are taking no steps to solve the fundamental problems for themselves. If enough cities establish successful policies for themselves, there will be two results. First, a precedent will have been set for coping with the fundamental underlying source of difficulties. Second, the larger the number of areas that solve their problems for themselves, the sooner and more forcefully will the remaining uncontrolled growth impinge on other parts of the country and the more quickly will the nation realistically face the long-range issues of stress arising from excessive growth.

So what can a city do? It can influence its future by choosing among the components of attractiveness. The attractiveness components of a city fall into two categories according to whether they operate more forcefully on the quality of life in the city or on inward migration and growth. These two categories are the “diffuse” and the “compartmentalized” characteristics of a city. The objective should be to maximize the diffuse characteristics of the city in order to improve the quality of urban life while controlling the compartmentalized characteristics in order to prevent the expanded population that would defeat the improvement for present residents.

The diffuse characteristics, such as public safety and clean air, are shared equally by all; their effect is not limited to particular individuals; and they apply alike to present residents and those who might move in. The compartmentalized characteristics of a city, like jobs and housing, are identified with particular individuals; they can be possessed by present residents but are not necessarily available to others from the outside.

Every diffuse characteristic of a city that makes it more attractive for the present residents will also make it more attractive for those who might move in, who would increase the population and density. Therefore, every improvement in the diffuse categories of attractiveness must be accompanied by some worsening in the compartmentalized categories of attractiveness to prevent self-defeating growth. The attractiveness characteristics of a city should be categorized in terms of whether they affect all residents or primarily potential newcomers. For example, the vitality of industry, a balanced socioeconomic mix of population, the quality of schools, the freedom from pollution, low crime rates, public parks, and cultural facilities are all desirable to present residents. If there is no counterbalance to restrain an expanding population, such attractive features tend to be self-defeating by causing inward migration. But the compartmentalized characteristics of a city primarily affect growth without necessarily reducing the quality of life for present residents. The number of housing units and the number of jobs tend to be compartments in the sense that they have a one-to-one correspondence with individuals rather than each being shared by all. The absence of an unoccupied house or a job can be a strong deterrent to in-migration, without necessarily driving down the internal quality of life.

I see no solution for urban problems until cities begin to exhibit the courage to plan in terms of a maximum population, a maximum number of housing units, a maximum permissible building height, and a maximum number of jobs. A city must also choose the type of city it wants to be. To become and remain a city that is all things to all people is impossible. There can be many uniquely different kinds of cities, each with its special mix of advantages and disadvantages. However, the policies that create one type of city may destroy another type. A choice of city type must be made, and corresponding policies must be chosen to create the combination of advantages and disadvantages that are characteristic of that type. One might have an industrial city, a commercial city, a resort city, a retirement city, or a city that attracts and traps without opportunity a disproportionate number of unemployed and welfare residents, as some cities are now doing. But there are severe limits on how many types of cities can be created simultaneously in one place. When the choices have been made, and when effort is no longer dissipated in growth, there will be an opportunity to come to grips with social and economic decay.

Why do I bring this message to the American Public Works Association? Because the members are at the center of the two most important issues I have raised. First, leaders in public works are the custodians of the technological aspects of the urban environment. Those responsible for the physical aspects of a city can continue to solve the technological subgoals of roads, water, waste, and transportation and thereby sustain the growth process and cause a continual shifting of pressures into the social realm of rising crime, increasing psychological trauma, growing welfare costs, and accelerating community breakdown. Or, they can move to reverse the growth attitudes that in the past we considered good, but are good no more, and help halt further expansion of that part of our technological base on which the urban crisis is growing. A second reason for these issues to be important in public works comes from the unique influence of public works over what I call the compartmentalized characteristics of a city. Public works actions directly affect the number of streets that are built, the number of houses that are erected, and the number of industrial locations that are established. Such physical actions, backed up by zoning and municipal policy, determine the kind of urban growth and whether or not there is to be growth. Through the judicious use of, and indeed the appropriate limitation of, water supply, drainage, building heights, waste disposal, road building, and transportation systems, a city can influence its future.

The reader may be thinking that planning and controlling the size and composition of a city and the migration to it are undemocratic or immoral. It may even seem that I am suggesting control where there has not been control before. Neither is true. Every city has arrived at its present size, character, and composition because of the actions that have controlled the city’s evolution in the past. By adding to the water system, sewers, and streets, a city has, in effect, decided to increase its size. By building a rapid transit system a city is often, in effect, deciding to change the composition of its population by encouraging new construction in outlying areas, allowing inner areas to decay, and attracting low-income and unskilled persons to the inner ring at the same time that job opportunities decline. In other words, a control of growth and migration has been exerted at all times, but it has often been guided by short-term considerations, with unexpected and undesirable long-term results. The issue is not one of control or no control. The issue is the kind of control and toward what end.

The interurban control of population movement is the internal counterpart of international control of population movement. Except for the legal, coercive, psychological, and economic deterrents to human mobility, the standard of living and the quality of life of all countries would fall to the level set by the population group that accepts the lowest standards. No group can be expected to exert the self-discipline now necessary to limit population and the environmental demands of industrialization unless there is a way to keep the future advantages of such self-discipline from being swallowed up by inward migration. If the control of international movement of population is ethical, then some intercity counterpart must also be ethical. Or, if the justification is only that of practical necessity, then the internal necessity arises in a country that is reaching its growth limit without having established a national means to implement a compromise between quantity and quality. Between nations, countries exert restrictions on population movement that are not allowed internally between urban areas. Even so, the policies of each city have a powerful effect on mobility and on the resulting character of the city. Because controls are implicit in every action taken and every urban policy adopted, a city should understand the future consequences of its present actions. A city affects its local choice between quantity and quality mostly by how it handles the diffuse versus the compartmentalized components of attractiveness.

The difference between diffuse and compartmentalized control of urban population can be illustrated by two extremes of policies that might govern the availability of water. Depending on how it is managed, the availability of water might be either a diffuse or a compartmentalized control on growth. Consider a city with a limited water supply—more and more this will be the actual situation. To illustrate diffuse control, one could distribute water freely and equally to everyone, both present and future residents. New houses could be constructed, new industries could be encouraged, growth could be continued, and the water could be divided among all. If no other growth limits were encountered, growth would continue until the low water pressure, occasional shortages, and the threat of disaster from drought had risen to the point where out-migration equaled in-migration. Under this circumstance of unrestricted access to water, net growth would have been stopped, but the equally distributed nature of the water shortage would have reduced the quality of life for all residents. The water shortage would be diffuse; it would be spread to all, former residents and newcomers alike. Alternatively, the opposite water policy illustrates compartmentalized control. Building permits and new water connections could be denied so that water demand is constrained to lie well within the water supply. Water would be available to present, but not to new, residents. Under these circumstances, the quality of life for the present residents would be maintained, but growth beyond the limit of satisfactory water supply would be restricted.

I believe that such a choice between present residents and potential in-migrants is inherent in a practical solution of our urban problems. Unless control through such self-interest is acceptable, and ways are available to exercise control, there is no incentive for any city or state to solve its own problems. Its efforts will be swamped from the outside. There must be freedom for local action, and the consequent differences between areas, if social experiments are to lead to better futures and if there is to be diversity in the country rather than one gray homogenized sameness. If there is to be any meaning to the president’s hope of preserving “the ability of citizens to have a major voice in determining policies that most directly affect them,” local areas must be able to control their destinies in different ways and toward different ends.

If people are to influence the policies most affecting them, it follows that policies will be different in different places, and the resulting trade-offs between growth and the quality of life will be different. If there is to be any substance to local choice, there must be differences between localities.

In the policies for a city that I am proposing, the ethical and legal issues are substantial. A city, in looking after its own well-being, will no doubt be accused of being selfish because it discriminates against nonresidents. But what are the alternatives? Must it discriminate against its own present residents instead? Must it discriminate against its own long-term interests? Must it be forced to take only a short-range view of its future? Must it be a party to delaying the day when the nation faces the fundamental choice between quality and quantity? Our past policies have not been so successful that they should persuade us against new experiments.

If a sufficient number of cities find new ways of controlling their own destinies in spite of national policy and what other cities do, then pressures to work toward the long-term well-being of the country will be quickly generated. If some cities and states take effective steps to establish an equilibrium with their natural surroundings, and to maintain a viable and proper internal balance of population and industry, then the remaining growth in the country will quickly descend on those communities and states that have taken no such action. A national consensus to establish a viable balance with the capacity of the environment will quickly develop out of the contrasts between those who have and those who have not dealt with the basic issues of overcommitment.

In summary, I believe that the country is now heading more deeply into economic and social difficulty. Technological solutions will no longer suffice. There is no national consensus strong enough to support an effective national policy nor to ensure national leadership in solving the problems that are arising from growth and overcommitment of the nation’s long-term capability. But, fortunately, the problems are solvable piecemeal at the local level independently of other areas and of the national government. Local action can set a precedent for the country as a whole. Those in public works are in a uniquely influential position for exerting that leadership.[p.p. 277-284]