Reading 23 - University of Richmond

Reading 23

Limits to Growth1

Overshoot

To overshoot means to go too far, to go be-yond limits accidentally—without intention. Peo-ple experience overshoots every day. When yourise too quickly from a chair, you may momen-tarily lose your balance. If you turn on the hotwater faucet too far in the shower, you may bescalded. On an icy road your car might slide pasta stop sign. At a party you may drink much morealcohol than your body can safely metabolize; inthe morning you will have a ferocious headache.Construction companies periodically build morecondominiums than are demanded, forcing themto sell units below cost and confront the possi-bility of bankruptcy. Too many fishing boats areoften constructed. Then fishing fleets grow solarge that they catch far more than the sustainableharvest. This depletes the fish population andforces ships to remain in harbor. Chemical com-panies have produced more chlorinated chemicalsthan the upper atmosphere can safely assimilate.Now the ozone layer will be dangerously depletedfor decades until stratospheric chlorine levels de-cline.

The three causes of overshoot are always thesame, at any scale from personal to planetary.2

First, there is growth, acceleration, rapid change.Second, there is some form of limit or barrier, be-yond which the moving system may not safely go.Third, there is a delay or mistake in the percep-tions and the responses that strive to keep thesystem within its limits. These three are neces-sary and sufficient to produce an overshoot.

Overshoot is common, and it exists in almostinfinite forms. The change may be physical—growth in the use of petroleum. It may beorganizational—an increase in the number ofpeople supervised. It may be psychological—continuously rising goals for personal consump-tion. Or it may be manifest in financial, biological,political, or other forms.

The limits are similarly diverse—they may beimposed by a fixed amount of space; by limitedtime; by constraints inherent in physical, biologi-

cal, political, psychological, or other features of asystem.

The delays, too, arise in many ways. They mayresult from inattention, faulty data, delayed in-formation, slow reflexes, a cumbersome or quar-relling bureaucracy, a false theory about how thesystem responds, or from momentum that pre-vents the system from being stopped quickly de-spite the best efforts to halt it. For example, de-lays may result when a driver does not realize howmuch his car’s braking traction has been reducedby ice on the road; the contractor uses currentprices to make decisions about construction activ-ity that will affect the market two or three years inthe future; the fishing fleet owners base their deci-sions on data about recent catch, not informationabout the future rate of fish reproduction; chemi-cals require years to migrate from where they areused to a point in the ecosystem where they causesevere damage.

Most instances of overshoot cause little harm.Being past many kinds of limits does not exposeanyone to serious damage. Most types of over-shoot occur frequently enough that when they arepotentially dangerous, people learn to avoid themor to minimize their consequences. For example,you test the water temperature with your hand be-fore stepping into the shower stall. Sometimesthere is damage, but it is quickly corrected: Mostpeople try to sleep extra long in the morning aftera late night drinking in the bar.

Occasionally, however, there arises the poten-tial for catastrophic overshoot. Growth in theglobe’s population and material economy con-fronts humanity with this possibility. It is the fo-cus of our work.

Throughout this text we will grapple with thedifficulties of understanding and describing thecauses and consequences of a population andeconomy that have grown past the support capaci-ties of the earth. The issues involved are complex.The relevant data are often poor in quality andincomplete. The available science has not yet pro-

1Donella Meadows, Jorgen Randers, Dennis Meadows, excepts chs 1 and 4 from Limits to Growth: The 30-Year Update,2004. Authors’ references omitted.

2emphasis added

159

Reading 23 Limits to Growth

duced consensus among researchers, much lessamong politicians. Nonetheless, we need a termthat refers to the relation between humanity’s de-mands on the planet and the globe’s capacity toprovide. For this purpose we will use the phraseecological footprint.

The term was popularized by a study MathisWackernagel and his colleagues conducted for theEarth Council in 1997. Wackernagel calculated theamount of land that would be required to providethe natural resources consumed by the popula-tion of various nations and to absorb their wastes.Wackernagel’s term and mathematical approachwere later adopted by the World Wide Fund forNature (WWF), which provides semiannual data onthe ecological footprint of more than 150 nationsin its Living Planet Report. According to thesedata, since the late 1980s the earth’s peoples havebeen using more of the planets resource produc-tion each year than could be regenerated in thatyear. In other words, the ecological footprint ofglobal society has overshot the earth’s capacity toprovide. There is much information to supportthis conclusion.

The potential consequences of this overshootare profoundly dangerous. The situation isunique; it confronts humanity with a variety of is-sues never before experienced by our species on aglobal scale. We lack the perspectives, the culturalnorms, the habits, and the institutions required tocope. And the damage will, in many cases, takecenturies or millennia to correct.

But the consequences need not be catas-trophic. Overshoot can lead to two different out-comes. One is a crash of some kind. Another is adeliberate turnaround, a correction, a careful eas-ing down. We explore these two possibilities asthey apply to human society and the planet thatsupports it. We believe that a correction is possi-ble and that it could lead to a desirable, sustain-able, sufficient future for all the world’s peoples.We also believe that if a profound correction is notmade soon, a crash of some sort is certain. Andit will occur within the lifetimes of many who arealive today.

These are enormous claims. How did we ar-rive at them? Over the past 30 years we haveworked with many colleagues to understand thelong-term causes and consequences of growth inhuman population and in its ecological footprint.We have approached these issues in four ways—ineffect using four different lenses to focus on datain different ways, just as the lenses of a micro-scope and a telescope give different perspectives.

Three of these viewing devices are widely usedand easy to describe: (1) standard scientific andeconomic theories about the global system; (2)data on the world’s resources and environment;and (3) a computer model to help us integrate thatinformation and project its implications. Much ofthis book expands on those three lenses. It de-scribes how we used them and what they allowedus to see.

Our fourth device is our “worldview” an in-ternally consistent set of beliefs, attitudes, andvalues—a paradigm, a fundamental way of look-ing at reality. Everybody has a worldview; it in-fluences where they look and what they see. Itfunctions as a filter; it admits information consis-tent with their (often subconscious) expectationsabout the nature of the world; it leads them todisregard information that challenges or discon-firms those expectations. When people look outthrough a filter, such as a pane of colored glass,they usually see through it, rather than seeing it—and so, too, with worldviews. A worldview doesn’tneed to be described to people who already shareit, and it is difficult to describe to people whodon’t. But it is crucial to remember that everybook, every computer model, every public state-ment is shaped at least as much by the worldviewof its authors as by any “objective” data or analy-sis.

We cannot avoid being influenced by our ownworldview. But we can do our best to describeits essential features to our readers. Our world-view was formed by the Western industrial soci-eties in which we grew up, by our scientific andeconomic training, and by lessons from travellingand working in many parts of the world. But themost important part of our worldview, the partthat is least commonly shared, is our systems per-spective.

Like any viewpoint—for example, the top ofany hill—a systems perspective lets people seesome things they would never have noticed fromany other vantage point, and it may block theview of other things. Our training concentrated ondynamic systems—on sets of interconnected ma-terial and immaterial elements that change overtime. Our training taught us to see the worldas a set of unfolding behavior patterns, such asgrowth, decline, oscillation, overshoot. It hastaught us to focus not so much on single pieces ofa system as on connections. We see the many ele-ments of demography, economy, and the environ-ment as one planetary system, with innumerableinteractions. We see stocks and flows and feed-

160


backs and thresholds in the interconnections, allof which influence the way the system will behavein the future and influence the actions we mighttake to change its behavior.

The systems perspective is by no means theonly useful way to see the world, but it is one wefind particularly informative. It lets us approachproblems in new ways and discover unsuspectedoptions. We intend to share some of its conceptshere, so you can see what we see and form yourown conclusions about the state of the world andthe choices for the future

....Absolute, global rates of change are greater

now than ever before in the history of our species.Such change is driven mainly by exponentialgrowth in both population and the material econ-omy. Growth has been the dominant behavior ofthe world socioeconomic system for more than200 years.... Individuals support growth-orientedpolicies, because they believe growth will givethem an ever increasing welfare. Governmentsseek growth as a remedy for just about everyproblem. In the rich world, growth is believedto be necessary for employment, upward mobility,and technical advance. In the poor world, growthseems to be the only way out of poverty. Manybelieve that growth is required to provide the re-sources necessary for protecting and improvingthe environment. Government and corporate lead-ers do all they can to produce more and moregrowth.

For these reasons growth has come to beviewed as a cause for celebration. Just con-sider some synonyms for that word: development,progress, advance, gain, improvement, prosperity,success.

Those are psychological and institutional rea-sons for growth. There are also what systemspeople call structural reasons, built into the con-nections among the elements of the population-economy system....

Growth can solve some problems but it createsothers. That is because of limits. The Earth is fi-nite. Growth of anything physical, including thehuman population and its cars and houses andfactories, cannot continue forever. But the lim-its to growth are not limits to the number of peo-ple, cars, houses, or factories, at least not directly.They are limits to throughput—to the continuousflows of energy and materials needed to keep peo-ple, cars, houses, and factories functioning. Theyare limits to the rate at which humanity can ex-tract resources (crops, grass, wood, fish) and emit

wastes (greenhouse gases, toxic substances) with-out exceeding the productive or absorptive capac-ities of the world.

The population and economy depend upon air,water, food, materials, and fossil fuels from theEarth. They emit wastes and pollution back to theEarth. Sources include mineral deposits, aquifers,and the stock of nutrients in soils; among thesinks are the atmosphere, surface water bodies,and landfills. The physical limits to growth arelimits to the ability of planetary sources to providematerials and energy and to the ability of plane-tary sinks to absorb the pollution and waste; seefigure 1.

...the bad news is that many crucial sources areemptying or degrading, and many sinks are fillingup or overflowing. The throughput flows presentlygenerated by the human economy cannot be main-tained at their current rates for very much longer.Some sources and sinks are sufficiently stressedthat they are already beginning to limit growthby, for instance, raising costs, increasing pollutionburdens, and elevating the mortality rate.

The good news is that current high rates ofthroughput are not necessary to support a decentstandard of living for all the world’s people. Theecological footprint could be reduced by loweringpopulation, altering consumption norms, or im-plementing more resource-efficient technologies.These changes are possible. Humanity has theknowledge necessary to maintain adequate levelsof final goods and services while reducing greatlythe burden on the planet. In theory there aremany possible ways to bring the human ecologi-cal footprint back down below its limits.

But theory does not automatically becomepractice. The changes and choices that will bringdown the footprint are not being made, at leastnot fast enough to reduce the growing burden onthe sources and sinks. They are not being madebecause there is no immediate pressure to makethem, and because they take a long time to imple-ment...we [here] discuss the signals that warn hu-man society about the symptoms of its overshoot.And we examine the speed with which people andinstitutions can respond....

Our computer model, World3...permits us toassemble many data and theories, putting thewhole picture—growth, limits, response delays—into an explicit and coherent whole. And it givesus a tool for projecting the future consequencesof our present understanding. We show what hap-pens when the computer simulates the system asit might evolve, assuming no profound changes,

161


Natural

Resources

Materials and

Fuels in Use

Wastes in the

EnvironmentResource Use Emissions

SOURCES SINKS

Figure 1: Natural resource limits. The Earth’s carrying capacity is determined by rates (the arrows in the figure):the rate at which we can sustainably extract natural resources from sources and the rate at which we our wasteemissions can be absorbed by sinks.

no extraordinary efforts to see ahead, to improvesignals, or to solve problems before they become

critical....

The Dynamics of Growth in a Finite World

The factors responsible for growth in popula-tion and industry involve many long-term trendsthat reinforce and conflict with each other. Birthrates are coming down faster than expected, butthe population is still rising. Many people aregetting richer; they are demanding more indus-trial products. But they also want less pollu-tion. The flows of energy and materials requiredto sustain industrial growth are depleting nonre-newable resource stocks and deteriorating renew-able resources. But there is steady progress in de-veloping technologies that discover new reservesand use materials more efficiently. Every societyconfronts a shortage in capital; investments areneeded to find more resources, produce more en-ergy, clean up pollution, improve schools, healthcare, and other social services. But those invest-ments must compete with an ever growing de-mand for more consumer goods.

How will these trends interact and evolve overthe coming decades? To understand their implica-tions, we need a model much more complex thanthe ones in our heads. World3 [is] the computermodel we have created and used. We summa-rize here the main features of World3’s structureand describe several important insights it gives usabout the twenty-first century.

The Purpose of World3

The universal desire for certainty about what is tocome can lead to misunderstanding and frustra-tion when someone presents a model as the basisfor talking about the future...to minimize confu-

sion about our goals, we start with several defini-tions and cautionary notes about models.

A model is a simplified representation of real-ity.3 If it were a perfect replica, it would not beuseful. For example, a road map would be ofno use to drivers if it contained every feature ofthe landscape it represents—it focuses on roadsand omits, for example, most features of build-ings and plants along the way. A small physi-cal airplane model can be useful for exploring thedynamics of a particular airfoil in a wind tunnel,but it gives no information about the comfort ofpassengers in the eventual operational plane. Apainting is a graphic model that may convey amood or the physical placement of features on alandscape. But it does not answer any questionsabout the cost or the insulation of the buildingsit portrays. To deal with those issues, a differ-ent graphic model would be required—an archi-tect’s construction blueprint. Because models arealways simplifications, they are never perfectlyvalid; no model is completely true.

...

To avoid creating impenetrable thickets of as-sumptions, modelers must discipline themselves.They cannot put into their models all they know;they have to put in only what is relevant for thepurpose of the model. The art of modeling, likethe arts of poetry or architecture or engineeringor map-making, is to include just what is neces-sary to achieve the purpose, and no more. That iseasy to say and hard to do.

Therefore to understand a model and judgeits utility, it’s important to understand its pur-

3emphasis added

162


pose. We developed World3 to understand thebroad sweep of the future—the possible modes,or behavior patterns, through which the humaneconomy will interact with the carrying capac-ity of the planet over the coming century. Ofcourse, there are many other important long-termglobal questions to ask: What policies might max-imize the industrial development possibilities forAfrica? What is the best design for a family plan-ning program in a region where many people areilliterate? How can society close the gap betweenthe rich and the poor within and between nations?Will conflict or negotiation become the dominantmeans for resolving disputes among nations? Thefactors and relationships needed to answer thosequestions are largely missing from World3. Othermodels, including other computer models, mighthelp answer some of those questions. But if theyare to be useful, those models must take into ac-count the answers we generate to World3’s corequestion: How may the expanding global popula-tion and material economy interact with and adaptto the earth’s limited carrying capacity over thecoming decades?

To be more specific, the carrying capacity is alimit. Any population that grows past its carryingcapacity, overshooting the limit, will not long sus-tain itself. And while any population is above thecarrying capacity, it will deteriorate the supportcapacity of the system it depends upon. If regen-eration of the environment is possible, the deteri-oration will be temporary. If regeneration is notpossible, or if it takes place only over centuries,the deterioration will be effectively permanent.

A growing society can approach its carrying ca-pacity in four generic ways (see figure 2). First, itcan grow without interruption, as long as its limitsare far away or are growing faster than the popu-lation. Second, it can level off smoothly below thecarrying capacity, in a behavior that ecologists calllogistic, or S-shaped, or sigmoid, growth, shown infigure 2(b). Neither of those options is any longeravailable to the global society, because it is alreadyabove its sustainable limits.4

The third possibility for a growing society isto overshoot its carrying capacity without doingmassive and permanent damage. In that casethe ecological footprint would oscillate aroundthe limit before levelling off. This behavior, il-lustrated in figure 2(c), is called damped oscilla-tion. The fourth possibility is to overshoot thelimits, with severe and permanent damage to the

resource base. If that were to occur, the popula-tion and the economy would be forced to declinerapidly to achieve a new balance with the recentlyreduced carrying capacity at a much lower level.We use the phrase overshoot and collapse to des-ignate this option, shown in figure 2(d).

There is pervasive and convincing evidence thatthe global society is now above its carrying capac-ity. What policies will increase the chances of asmooth transition back beneath planetary limits—a transition like 2(c) rather than 2(d)?

Our concept of the “global society” incorpo-rates the effects of both the size of the popula-tion and the size and composition of its consump-tion. To express this concept we use the term eco-logical footprint that has been defined by MathisWackernagel and his colleagues. As we have in-dicated, the ecological footprint of humanity isthe total burden humankind places on the earth.It includes the impact of agriculture, mining, fishcatch, forestharvest, pollution emissions, land de-velopment, and biodiversity reductions. The eco-logical footprint typically grows when the popula-tion grows, because it grows when consumptionincreases. But it can also shrink when appropriatetechnologies are utilized to reduce the impact perunit of human activity.

The concerns motivating our development ofWorld3 may be expressed another way. Given thatthe ecological footprint of the global population ispresently above the earth’s carrying capacity, willcurrent policies lead us to a relatively peaceful, or-derly oscillation, without forcing drastic declinesin population and economy? Or will the globalsociety experience collapse? If collapse is morelikely, when might it come? What policies couldbe implemented now to reduce the pace, the mag-nitude, the social and ecological costs of the de-cline?

These are questions about broad behavioralpossibilities, not precise future conditions. An-swering them requires a different kind of modelthan does precise prediction. For example, if youthrow a ball straight up into the air, you knowenough to describe what its general behavior willbe. It will rise with decreasing speed, then reversedirection and fall faster and faster until it hits theground. You know it will not continue to rise for-ever, nor begin to orbit the earth, nor loop threetimes before landing.

If you wanted to predict exactly how high theball would rise or precisely where and when it

4emphasis added

163


Figure 2: Possible modes of approach of a population to its carrying capacity. The central question addressedby the World3 model is: which of these behavior modes is likely to be the result as the human population andeconomy approach the global carrying capacity?

would hit the ground, you would need precise in-formation about many features of the ball, the al-titude, the wind, the force of the initial throw, andthe laws of physics. Similarly, if we wanted to at-tempt to predict the exact size of the world pop-ulation in 2026, or forecast when world oil pro-duction will peak, or specify precisely the rate ofsoil erosion in 2070, we would need a much morecomplicated model than World3.

To our knowledge no one has come close tomaking such a model; nor do we believe anyonewill ever succeed. It is simply not possible tomake accurate “point predictions” about the fu-ture of the world’s population, capital, and envi-ronment several decades from now No one knowsenough to do that, and there are excellent rea-sons to believe they never will. The global so-cial system is horrendously and wonderfully com-plex, and many of its crucial parameters remainunmeasured. Some are probably unmeasurable.Human understanding of complex ecological cy-cles is very limited.

Moreover, the capacity for humans to observe,adapt and learn, to choose, and to change theirgoals makes the system inherently unpredictable.

Therefore, when we constructed our formalworld model, it was not to make point predictions,

but rather to understand the broad sweeps, thebehavioral tendencies of the system. Our goal isto inform and to influence human choice. To ac-complish these goals, we do not need to predictthe future precisely. We need only identify poli-cies that will increase the likelihood of sustainablesystem behavior and decrease the severity of fu-ture collapse. A prediction of disaster deliveredto an intelligent audience with the capacity to actwould, ideally, defeat or falsify itself by inducingaction to avoid the calamity. For all those reasonswe chose to focus on patterns rather than indi-vidual numbers. With World3 we are engaged, wehope, in self-defeating prophecy.

To achieve our goals, we put into World3 thekinds of information you might use to under-stand the behavioral tendencies of thrown balls(or growing economies and populations), not thekinds of information you would need to describethe exact trajectory of one particular throw of onespecific ball.

Because of the uncertainties and simplifica-tions we know exist in the model (and othersthat we suppose it must contain, though we havenot yet recognized them), we do not put faith inthe precise numerical path the model generatesfor population, pollution, capital, or food produc-

164


tion. Still, we think the primary interconnectionsin World3 are good representations of the impor-tant causal mechanisms in human society Thoseinterconnections, not the precise numbers, deter-mine the model’s general behavior. As a conse-quence, we do have faith in the dynamic behaviorsgenerated by World3.

Limits and No Limits

An exponentially growing economy depletes re-sources, emits wastes, and diverts land from theproduction of renewable resources. As it operateswithin a finite environment, the expanding econ-omy will begin to create stresses. These stressesbegin to grow long before society arrives at thepoint where further growth is totally impossible.In response to the stresses, the environment be-gins to send signals to the economy. These sig-nals take many forms. More energy is neededto pump water from diminishing aquifers, the in-vestment required to develop a hectare of newfarmland goes up, damage suddenly becomes ap-parent from emissions that were thought to beharmless, natural systems of the earth heal them-selves more slowly under the assault from pol-lution. These rising real costs do not necessar-ily show up immediately through increased mone-tary prices, because market prices can be reducedby fiat or subsidies and distorted in other ways.Whether or not they are reinforced by rising mar-ket prices, the signals and pressures function asimportant parts of negative feedback loops. Theyseek to bring the economy into alignment withthe constraints of the sum rounding system. Thatis, they seek to stop the growth of the ecologicalfootprint that is stressing the planets sources andsinks.

...

In the “real world” there are many other kindsof limits, including managerial and social ones.Some of them are implicit in the numbers inWorld3, since our model coefficients came fromthe world’s “actual” history over the past 100years. But World3 has no war, no labor strikes, nocorruption, no drug addiction, no crime, no ter-rorism. Its simulated population does its best tosolve perceived problems, undistracted by strug-gles over political power or ethnic intolerance orby corruption. Since it lacks many social limits,World3 does paint an overly optimistic picture offuture options.

What if we’re wrong about, for example, theamount of nonrenewable resources under theground remaining to be discovered? What if theactual number is only half of what we’ve assumed,or double, or 10 times more? What if the earth’s“real” ability to absorb pollution without harm tothe human population is not 10 times the 1990rate of emissions, but 50 times or 500 times? (Or0.5?) What if technologies are invented that de-crease (or increase) pollution emission per unit ofindustrial production?

A computer model is a device for answeringsuch questions. It can be used quickly and cheaplyto conduct tests. All those “what ifs” are testable.It is possible, for example, to set the numbers onWorld3’s limits astronomically high or to programthem to grow exponentially We have tried that.When all physical limits are effectively removedfrom the model system by an assumed technol-ogy that is unlimited in potential, practically in-stantaneous in impact, without cost, and error-free, the simulated human economy grows enor-mously. Figure 3 shows what happens.

In this run, population slows its growth, levelsoff at almost nine billion, and then gradually de-clines, because the entire world population getsrich enough to experience the demographic tran-sition. Average life expectancy stabilizes near 80years worldwide. Average agricultural yield risesby the year 2080 to nearly six times its year-2000value. Industrial output soars off the top of thegraph—it is finally stopped at a very high level bya severe labor shortage, because there is 40 timesas much industrial capital to manage and run asthere was in the year 2000, but only 1.5 times asmany people. (We could take away even that limitby assuming a sufficiently fast exponential rise inlabor’s capacity to use capital.)

By the simulated year 2080, the global econ-omy is producing 30 times as much industrial out-put and 6 times as much food as it did in 2000.To achieve these results it has accumulated dur-ing the first eight decades of the twenty-first cen-tury almost 40 times as much industrial capital asit did during the entire twentieth century. Whileachieving that expansion in capital, the world por-trayed in figure 3 reduces its nonrenewable re-source use slightly and lowers its pollution emis-sions by a factor of eight compared with the year2000. Human welfare increases 25 percent from2000 to 2080, and the ecological footprint de-clines 40 percent. By the end of the scenario, theyear 2100, the footprint is safely back below thesustainable level.

165


Figure 3: If all physical limits to the World3 system are removed, population peaks near 9 billion and starts aslow decline in a demographic transition. The economy grows until by the year 2080 it is producing 30 timesthe year-2000 level of industrial output, while using the same annual amount of nonrenewable resources andproducing only one-eighth as much pollution per year.

Some people believe in this kind of scenario;expect it; revel in it. We know stories of remark-able efficiency increases in particular countriesor economic sectors or industrial processes...wehope and believe that further efficiency improve-ments are possible, even 100-fold improvements.But the data...show no indication of the wholeglobal economy achieving such gains so quickly.If nothing else would prevent such rapid changes,the lifetime of capital plants—the time it takes toreplace or retrofit the vehicle fleet, building stock,and installed machinery of the global economy—and the ability of existing capital to produce thatmuch new capital so fast make this “dematerial-ization” scenario unbelievable to us. The difficul-ties of achieving this infinity scenario would bemagnified in “real life” by the many political andbureaucratic constraints preventing the price sys-tem from signaling that the needed technologiescan be profitable.

We include this run here not because we thinkshows you a credible future of the “real world,”but because we think it tells you something aboutWorld3 and something about modeling.

It reveals that World3 has built into its struc-

ture a self-limiting constraint on population andno self-limiting constraint on capital. The modelis constructed in such a way that the global popu-lation will eventually level off and start declining,if industrial output per capita rises high enough.But we see little “real world” evidence that therichest people or nations ever lose interest in get-ting richer. Therefore, policies built into World3represent the assumption that capital owners willcontinue to seek gains in their wealth indefinitelyand that consumers will always want to increasetheir consumption.

Figure 3 also demonstrates one of the most fa-mous principles of modeling: Garbage In, GarbageOut, or GIGO. If you put unrealistic assumptionsinto your model, you will get unrealistic results.The computer will tell you the logical consequencesof your assumptions, but it will not tell you whetheryour assumptions are true.5 If you assume theeconomy can increase industrial capital accumu-lation 40-fold, that physical limits no longer ap-ply, that technical changes can be built into thewhole global capital plant in only two years with-out cost, World3 will give you virtually unlimitedeconomic growth along with a declining ecological

5emphasis added

166


footprint. The important question about this andevery other computer run is whether you believethe initial assumptions.

We don’t believe the assumptions behind fig-ure 3. We consider this a scenario that portraysan impossible technological utopia. So we labelthat run Infinity In, Infinity Out, or IFI-IFO (pro-nounced iffy-iffo). Under what we think are more“realistic” assumptions, the model begins to showthe behavior of a growing system running into re-sistance from physical limits.

Limits and Delays

A growing physical entity will slow and then stop ina smooth accommodation with its limits (S-shapedgrowth) only if it receives accurate, prompt signalstelling it where it is with respect to its limits, andonly if it responds to those signals quickly and ac-curately (figure 4(b)).

Imagine that you are driving a car and upahead you see a stoplight turn red. Normally youcan halt the car smoothly just before the light,because you have a fast, accurate visual signaltelling you where the light is, because your brainresponds rapidly to that signal, because your footmoves quickly as you decide to step on the brake,and because the car responds immediately to thebrake in a fashion you understand from frequentpractice.

If your side of the windshield were fogged upand you had to depend on a passenger to tell youwhere the stoplight was, the short delay in com-munication could cause you to shoot past the light(unless you slowed down to accommodate the de-lay). If the passenger lied, or if you denied whatyou heard, or if it took the brakes two minutes tohave an effect, or if the road had become icy, sothat it unexpectedly took the car several hundredmeters to stop, you would overshoot the light.

A system cannot come into an accurate and or-derly balance with its limit if its feedback signal isdelayed or distorted, if that signal is ignored ordenied, if there is error in adapting, or if the sys-tem can respond only after a delay. If any of thoseconditions pertain, the growing entity will correctitself too late and overshoot (figures 4(c) and 4(d)).

We have already described some of the infor-mation and response delays in World3. One ofthem is the delay between the time when a pollu-tant is released into the biosphere and the time atwhich it does observable harm to human health orthe human food supply. An example is the 10- to

15-year lag before a chlorofluorocarbon moleculereleased on the earth’s surface begins to degradethe stratospheric ozone layer. Policy delays arealso important. There is often a delay of manyyears between the date when a problem is firstobserved and the date when all important playersagree on it and accept a common plan for action.

One illustration of these delays is provided bythe percolation of PCBs through the environment.Since 1929 industry has produced some two mil-lion tons of the stable, oily, nonflammable chem-icals called polychlorinated biphenyls, or PCBs.They were used primarily to dissipate heat inelectrical capacitors and transformers, but alsoas hydraulic fluid, lubricants, fire retardants, andconstituents of paints, varnishes, inks, carbon-less copy paper, and pesticides. For 40 yearsusers of these chemicals dumped them in land-fills, along roads, into sewers and water bod-ies, without thinking of the environmental con-sequences. Then in a landmark study in 1966,designed to detect DDT in the environment, Dan-ish researcher Soren Jensen reported that in addi-tion to DDT, he had found PCBs to be widespreadas well. Since then other researchers have foundPCBs in almost all the globe’s ecosystems.

Most PCBs are relatively insoluble in water butsoluble in fats, and they have very long lifetimesin the environment. They move quickly throughthe atmosphere, and slowly through soils or sedi-ments in streams and lakes, until they are takenup into some form of life, where they accumu-late in fatty tissue and increase in concentrationas they move up the food chain. They are foundin the greatest concentrations in carnivorous fish,seabirds and mammals, human fat, and humanbreast milk.

The impacts of PCBs on the health of humansand other animals are only slowly being revealed.The story is particularly difficult to unravel be-cause PCB is a mixture of 209 closely related com-pounds, each of which may produce different ef-fects. Nevertheless, it is becoming apparent thatsome PCBs act as endocrine disrupters. Theymimic the action of some hormones, such as es-trogen, and block the action of others, such as thy-roid hormones. The effect—in birds, whales, po-lar bears, humans, any animal with an endocrinesystem—is to confuse delicate signals that gov-ern metabolism and behavior. Especially in de-veloping embryos, even minute concentrations ofendocrine disrupters can wreak havoc. They cankill the developing organism outright, or they canimpair its nervous system, intelligence, or sexual

167


Figure 4: Structural causes of the four possible behavior modes of the World3 model.

function.Because they migrate slowly, last a very long

time, and accumulate in higher levels of a foodchain, PCBs have been called a biological time-bomb. Although PCB manufacture and use hasbeen banned in many countries since the 1970s,a huge stock still exists. Of the total amount ofPCBs ever produced, much is still in use or storedin abandoned electrical equipment. In countrieswith hazardous waste laws, some of these oldPCBs are being buried or disposed of by controlledincineration that breaks up their molecular struc-ture and thus their bioactivity. In 1989 it was es-timated that 30 percent of all PCBs ever manufac-tured had already been released into the environ-ment. Only 1 percent had reached the oceans. The29 percent still unaccounted for was dispersed insoils, rivers, and lakes, where it would go on mov-ing into living creatures for decades.

Figure 5 shows another example of a pollutiondelay, the slow transport of chemicals throughsoil into groundwater. From the 1960s until 1990,when it was finally banned, the soil disinfectant1,2-dichloropropene (DCPe) was applied heavilyin the Netherlands in the cultivation of potatoesand flower bulbs. It contains a contaminant, 1,2-dichloropropane (DCPa), which, as far as scien-tists know, has an infinite lifetime in groundwater.A calculation for one watershed estimated thatthe DCPa already in the soil would work its waydown into groundwater and appear there in sig-nificant concentrations only after the year 2010.Thereafter it was expected to contaminate thegroundwater for at least a century in concentra-

tions up to 50 times the European Union’s drink-ing water standard.

The problem is not unique to the Netherlands.In the United States agricultural use of DCP wascancelled in 1977. Yet the Washington State Pes-ticide Monitoring Program found the chemical atconcentrations assumed to injure human healthwhen it monitored ground water at 243 sites in11 study areas between 1988 and 1995.13

A delay in a different sector of World3 is dueto the population age structure. A populationwith a recent history of high birth rates containsmany more young people than old people. There-fore, even if fertility falls, the population keepsgrowing for decades as the young people reachchild-bearing age. Though the number of chil-dren per family goes down, the number of fam-ilies increases. Because of this “population mo-mentum,” if the fertility of the entire world pop-ulation reaches replacement level (about two chil-dren per family on average) by the year 2010, thepopulation will continue growing until 2060 andwill level off at about eight billion.

There are many other delays in the “real world”system. Nonrenewable resources may be drawndown for generations before their depletion hasserious economic consequences. Industrial capi-tal cannot be built overnight. Once it is placed inoperation, it has a lifetime of decades. An oil re-finery cannot be converted easily or quickly into atractor factory or a hospital. It even requires timeto make it into a more efficient, less polluting oilrefinery.

World3 has many delays in its feedback mech-

168


Figure 5: The slow percolation of 1,2-DCP into ground water. The soil disinfectant DCP was used heavily in theNetherlands in the 1970s, then it was restricted, and finally in 1990 it was banned. As a result, the concentrationof DCP in the upper levels of agricultural soils has declined quickly. It was calculated in 1991 that its concentra-tion in groundwater will not peak until 2020, however, and there will be significant quantities of the chemical inthe water after the middle of the twenty-first century.

anisms, including all those mentioned above. Weassume there is a delay between the release ofpollution and its noticeable effect on the system.We assume a delay of roughly a generation be-fore couples fully trust and adjust their decisionsabout family size to changing infant mortalityrates. It normally takes decades in World3 be-fore investment can be reallocated and new cap-ital plant can be constructed and brought into fulloperation in response to a shortage of food or ser-vices. It takes time for land fertility to be regener-ated or pollution to be absorbed.

The simplest and most incontrovertible phys-ical delays are already sufficient to eliminatesmooth sigmoid as a likely behavior for the worldeconomic system. Because of the delays in the sig-nals from nature’s limits, overshoot is inevitableif there are no self-enforced limits, But that over-shoot might, in theory, lead either to oscillationor to collapse.

Overshoot and Oscillation

If the warning signals from the limits to the grow-ing entity are delayed, or if the response is delayed,and if the environment is not eroded when over-stressed, then the growing entity will overshoot its

limit for a while, make a correction, and under-shoot, then overshoot again, in a series of oscil-lations that usually damp down to an equilibriumwithin the limit (figure 4(c)).

Overshoot and oscillation can occur only if theenvironment suffers insignificant damage duringperiods of overload or can repair itself quicklyenough to recover fully during periods of under-load.

Renewable resources, such as forests, soils,fish, and rechargeable groundwater, are erodable,but they also have a self-regenerating capability.They can recover from a period of overuse, as longas it is not great enough or sustained enough thatdamage to the nutrient source, breeding stock, oraquifer is devastating. Given time, soil, seed, anda suitable climate, a forest can grow back. A fishstock can regenerate if its habitat and food supplyare not destroyed. Soils can be rebuilt, especiallywith active help from farmers. Accumulations ofmany kinds of pollution can be reduced if the en-vironment’s natural absorption mechanisms havenot been badly disturbed.

Therefore the overshoot and oscillation behav-ior mode is a significant possibility for the worldsystem. It has been demonstrated in some local-ities for some resources. New England, for ex-

169


ample, has several times witnessed periods whenmore sawmills were built than could be suppliedby the sustainable harvest of the region’s forests.Each time that happened, the commercial timberstands were eventually depleted, mills had to beshuttered, and then the industry waited decadesuntil the forest grew back and the overbuildingof sawmills could begin again. The coastal Nor-wegian fishery has gone through at least one cy-cle of fish depletion, with the government buyingup and retiring fishing boats until the fish stockscould regenerate.

The decline phase of an overshoot and oscilla-tion is not a pleasant period to live through. It canmean hard times for industries dependent uponan abused resource, or bad health in populationsexposed to high pollution levels. Oscillations arebest avoided. But they are not usually fatal to asystem.

Overshoots can become catastrophic when thedamage they cause is irreversible. Nothing canbring back an extinct species. Fossil fuels arepermanently destroyed in the very act of usingthem. Some pollutants, such as radioactive ma-terials, can’t be rendered harmless by any naturalmechanism. If the climate is significantly altered,geological data suggest that temperature and pre-cipitation patterns probably will not return to nor-mal within a time period meaningful to human so-ciety. Even renewable resources and pollution ab-sorption processes can be permanently destroyedby prolonged or systematic misuse. When trop-ical forests are cut down in ways that precludetheir regrowth, when the sea infiltrates freshwa-ter aquifers with salt, when soils wash away leav-ing only bedrock, when a soil’s acidity is changedsufficiently to flush out the heavy metals it hasstored, then the earth’s carrying capacity is dimin-ished permanently, or at least for a period thatappears permanent to human beings.

Therefore, the overshoot and oscillation modeis not the only one that could be manifested ashumanity approaches the limits to growth. Thereis one more possibility.

Overshoot and Collapse

If the signal or response from the limit is delayedand if the environment is irreversibly eroded whenoverstressed, then the growing economy will over-shoot its carrying capacity, degrade its resourcebase, and collapse (figure 4(d)).

The result of overshoot and collapse is a per-manently impoverished environment and a ma-terial standard of living much lower than whatwould have been possible if the environment hadnever been overstressed.

The difference between the overshoot and os-cillation and overshoot and collapse is the pres-ence of erosion loops in a system. These are pos-itive feedback loops of the worst kind. Normallythey are dormant, but when a situation gets bad,they make it worse by carrying a system down-ward at an ever-increasing pace.

For example, grasslands all over the worldhave co-evolved with grazing animals such asbuffalo, antelope, llamas, or kangaroos. Whengrasses are eaten down, the remaining stems androots extract more water and nutrients from thesoil and send up more grass. The number of graz-ers is held in check by predation, seasonal migra-tion, and disease. The ecosystem doesnot erode.But if the predators are removed, the migrationsare stymied, or the land is overstocked, an over-population of grazers can eat the grass down tothe roots. That can precipitate rapid erosion.

The less vegetation there is, the less coverthere is for the soil. With loss of cover, the soil be-gins to blow away in the wind or wash away in therain. The less soil there is, the less vegetation cangrow. Loss of vegetation allows still more soil toerode away And so on. Land fertility spirals downuntil the grazing range has become a desert.

There are several erosion loops in World3. Forexample:

• If people become more hungry, they work theland more intensively This produces more foodin the short term at the expense of investmentsin long-term soil maintenance. Lower soil fertil-ity then brings food production down even far-ther.

• When problems appear that require more in-dustrial output—pollution that requires abate-ment equipment, for example, or hunger thatcalls for more agricultural inputs, or resourceshortages that stimulate the discovery and pro-cessing of new resources—available investmentmay be allocated to solving the immediate prob-lem, rather than maintaining existing industrialcapital against depreciation. If the establishedindustrial capital plant begins to decline, thatmakes even less industrial output available inthe future. Reductions in output can lead tofurther postponed maintenance and further de-cline in the industrial capital stock

170


• In a weakening economy, services per capitamay decline. Reduced expenditures on familyplanning can eventually cause birth rates to in-crease. This produces growth in the population,which lowers services per capita even farther.

• If pollution levels increase too much, theycan erode the pollution absorption mechanismsthemselves, reducing the rate of pollution as-similation and raising the rate of pollutionbuildup still more.

This last erosive mechanism, impairment ofthe natural mechanisms for pollution assimila-tion, is particularly insidious. It is a phenomenonfor which we had little evidence when we first de-signed World3 more than 30 years ago. At thetime we had in mind such interactions as dumpingpesticides into water bodies, thereby killing theorganisms that normally clean up organic wastes;or emitting both nitrogen oxides and volatile or-ganic chemicals into the air, which react witheach other to make more damaging photochem-ical smog.

Since then other examples of the degradationof the earth’s pollution control devices have cometo light. One of them is the apparent ability ofshort-term air pollutants, such as carbon monox-ide, to deplete scavenger hydroxyl radicals in theair. These hydroxyl radicals normally react withand destroy the greenhouse gas methane. Whenair pollution removes them from the atmosphere,methane concentrations increase. By destroying apollution cleanup mechanism, short-term air pol-lution can make long-term climate change worse.

Another such process is the ability of air pollu-tants to weaken or kill forests, thereby diminish-ing a sink for the greenhouse gas carbon dioxide.A third is the effect of acidification—from fertiliz-ers or industrial emissions—on soils. At normallevels of acidity, soils are pollution absorbants.They bind with and sequester toxic metals, keep-ing them out of streams and groundwater andthus out of living organisms. But these bonds arebroken under acidic conditions. W M. Stigliani de-scribed this process in 1991.

As soils acidify, toxic heavy metals, accumu-lated and stored over long time periods (say,decades to a century) may be mobilized andleached rapidly into ground and surface wa-ters or be taken up by plants. The ongoingacidification of Europe’s soils from add depo-sition is dearly a source of real concern withrespect to heavy metal leaching.

Besides the ones we included in World3, thereare many other positive feedback loops in the“real world” with the potential to produce rapiderosion. We have mentioned the potential for ero-sion in physical and biological systems. An exam-ple of a very different kind would be breakdownin the social order. When a country’s elites be-lieve it is acceptable to have large differentials inwell-being within their nation, they can use theirpower to produce big differences in income be-tween themselves and most of the citizenry. Thisinequality can lead the middle classes to frustra-tion, anger,and protests. The disruption that re-sults from protests may lead to repression. Exer-cising force isolates the elites even farther fromthe masses and amplifies among the powerful theethics and values that justify large gaps betweenthem and the majority of the population. Incomedifferentials rise, anger and frustration grow, andthis can call forth even more repression. Eventu-ally there may be revolution or breakdown.

It is difficult to quantify erosive mechanisms ofany sort, because erosion is a whole-system phe-nomenon having to do with interactions amongmultiple forces. It appears only at times of stress.By the time it becomes obvious, it isn’t easilystopped. But despite these uncertainties, we cansay confidently that any system containing a la-tent erosion process also contains the possibilityof collapse, if it is overstressed.

On a local scale, overshoot and collapse canbe seen in the processes of desertification, min-eral or groundwater depletion, poisoning of agri-cultural soils or forest lands by ’long-lived toxicwastes, and extinction of species. Abandonedfarms, deserted mining towns, and forsaken in-dustrial dumps all testify to the “reality” of thissystem behavior. On a global scale, overshoot andcollapse could mean the breakdown of the greatsupporting cycles of nature that regulate climate,purify air and water, regenerate biomass, preservebiodiversity, and turn wastes into nutrients. Whenwe first published our results in 1972, the major-ity of people thought human disruption of natu-ral processes on a global scale was inconceivable.Now it is the subject of newspaper headlines, thefocus of scientific meetings, and the object of in-ternational negotiations.

World3: One Possible Scenario

In the simulated world of World3, the primarygoal is growth. The World3 population will stop

171


growing only when it is very rich. Its economy willstop growing only when it runs into limits. Its re-sources decline and deteriorate with overuse. Thefeedback loops that connect and inform its de-cisions contain substantial delays, and its phys-ical processes have considerable momentum. Itshould therefore come as no surprise that themost likely mode of behavior of the model worldis overshoot and collapse.

The graphs in figure 6 show the behavior ofWorld3 when it is run “as is,” with numbers weconsider a “realistic” description of the situationas it appeared on average during the latter partof the twentieth century, with no unusual techni-cal or policy assumptions. In 1972 we called itthe “standard run.” We did not consider it to bethe most probable future, and we certainly didn’tpresent it as a prediction. It was just a place tostart, a base for comparison. But many peopleimbued the “standard run” with more importancethan the scenarios that followed. To prevent thatfrom happening again, we’ll just call it “a refer-ence point.”

In this scenario the society proceeds along avery traditional path as long as possible withoutmajor policy change. It traces the broad outlineof history as we know it throughout the twenti-eth century. The output of food, industrial goods,and social services increases in response to obvi-ous needs and subject to the availability of capi-tal. There is no extraordinary effort, beyond whatmakes immediate economic sense, to abate pol-lution, con-serve resources, or protect the land.This simulated world tries to bring all peoplethrough the demographic transition and into aprosperous industrial economy. The world ac-quires widespread health care and birth controlas the service sector grows; it applies more agri-cultural inputs and gets higher yields as the agri-cultural sector grows; it emits more pollutants,demands more nonrenewable resources, and be-comes capable of greater production as the indus-trial sector grows.

The population rises from 1.6 billion in thesimulated year 1900 to 6 billion in the year 2000and more than 7 billion by 2030. Total industrialoutput expands by a factor of almost 30 between1900 and 2000 and then by 10 percent more by2020. Between 1900 and 2000 only about 30 per-cent of the earth’s total stock of nonrenewable re-sources is used; more than 70 percent of theseresources remain in 2000. Pollution levels in thesimulated year 2000 have just begun to rise signif-icantly, to 50 percent above the 1990 level. Con-

sumer goods per capita in 2000 are 15 percenthigher than in 1990, and nearly eight times higherthan in 1900.

If you cover the right half of the graphs, soyou can see only the curves up to the year 2000,the simulated world looks very successful. Lifeexpectancy is increasing, services and goods percapita are growing, total food production and in-dustrial production are rising. Average humanwelfare is increasing continuously. A few cloudsdo appear on the horizon: Pollution levels arerising, and so is the human ecological footprint.Food per person is stagnating. But generally thesystem is still growing, with few indications of themajor changes just ahead.

Then suddenly, a few decades into the twenty-first century, the growth of the economy stopsand reverses rather abruptly. This discontinua-tion of past growth trends is principally causedby rapidly increasing costs of non-renewable re-sources. This cost rise works its way through thevarious economic sectors in the form of increas-ingly scarce investment funds. Let’s follow theprocess.

In the simulated year 2000, the nonrenewableresources remaining in the ground would havelasted 60 years at the year-2000 consumptionrate. No serious resource limits are then in evi-dence. But by 2020 the remaining resources con-stitute only a 30-year supply. Why does this short-age arise so quickly? It occurs because growth inindustrial output and population raise resourceconsumption while drawing down the resourcestock. Between 2000 and 2020 population in-creases by 20 percent and industrial output by30 per-cent. During those two decades in fig-ure 6, the growing population and industrial plantuse nearly the same amount of nonrenewable re-sources as the global economy used in the entirecentury before! Naturally, more capital is then re-quired to find, extract, and refine what nonrenew-ables remain-in the incessant effort of the simu-lated world to fuel further growth,

As nonrenewable resources become harder toobtain, capital is diverted to producing more ofthem. That leaves less industrial output to investin sustaining the high agricultural output and fur-ther industrial growth. And finally, around 2020,investment in industrial capital no longer keepsup with depreciation. (This is physical investmentand depreciation; in other words, wear and tearand obsolescence, not monetary depreciation inaccounting books.) The result is industrial de-cline, which is hard to avoid in this situation, since

172


Figure 6: A Reference Point. The world society proceeds in a traditional manner without any major deviationfrom the policies pursued during most of the twentieth century. Population and production increase until growthis halted by increasingly inaccessible nonrenewable resources. Ever more investment is required to maintain re-source flows. Finally, lack of investment funds in the other sectors of the economy leads to declining output ofboth industrial goods and services. As they fall, food and health services are reduced, decreasing life expectancyand raising average death rates.

the economy cannot stop putting capital into theresource sector. If it did, the scarcity of materi-als and fuels would restrict industrial productioneven more quickly.

So maintenance and upkeep are deferred, theindustrial plant begins to decline, and along withit go the production of the various industrial out-puts that are necessary to maintain growing cap-ital stocks and production rates in the other sec-tors of the economy. Eventually the declining in-dustrial sector forces declines in the service andagricultural sectors, which depend on industrialinputs. The decline of industry has an especiallyserious impact on agriculture, since land fertilityhas already been degraded somewhat by overuseprior to the year 2000. Consequently, food pro-duction is maintained mainly by compensatingfor this degradation with industrial inputs suchas fertilizer, pesticides, and irrigation equipment.Over time the situation grows increasingly seri-ous, because the population keeps rising due tolags inherent in the age structure and in the pro-cess of social adjustment to fertility norms. Fi-nally, about the year 2030, population peaks andbegins to decrease as the death rate is driven up-ward by lack of food and health services. Averagelife expectancy, which was 80 years in 2010, be-

gins to decline.This scenario portrays a “nonrenewable re-

source crisis.” It is not a prediction. It is notmeant to forecast precise values of any of themodel variables, nor the exact timing of events.We do not believe it represents the most likely“real world” outcome...the strongest statement wecan make about this scenario is that it portraysthe likely general behavior made of the system, ifthe policies that influence economic growth andpopulation growth in the future are similar tothose that dominated the last part of the twenti-eth century, if technologies and values continue toevolve in a manner representative of that era, andif the uncertain numbers in the model are roughlycorrect....

Summary: Why Overshoot and Collapse?

A population and economy are in overshoot modewhen they are withdrawing resources or emittingpollutants at an unsustainable rate, but are notyet in a situation where the stresses on the sup-port system are strong enough to reduce the with-drawal or emission. In other words: Humanity isin overshoot when the human ecological footprint

173


is above the sustainable level, but not yet largeenough to trigger changes that produce a declinein its ecological footprint.

Overshoot comes from delays in feedback. De-cision makers in the system do not immediatelyget, or believe, or act upon information that lim-its have been exceeded. Overshoot is possible be-cause there are accumulated resource stocks thatcan be drawn down. For example, you can spendmore each month than you earn, at least for awhile, if you have stored up funds in a bank ac-count. You can drain water out of a bathtub fasterthan it is replenished by the faucet, at least untilyou have exhausted the initial stock of water inthe tub. You can remove from a forest wood ex-ceeding its annual growth rate as long as you startwith a standing stock of wood that has grown andaccumulated over many decades. You can buildup enough herds to overgraze, or boats to over-fish, if you have initially accumulated stocks offorage and fish that were not exploited in the past.The larger the initial stocks, the higher and longerthe overshoot can be. If a society takes its signalsfrom the simple availability of stocks, rather thanfrom their rates of replenishment, it will over-shoot.

Physical momentum adds to the delay in thewarning signals, and it is another source of delayin the response to them. Because of the time ittakes forests to regrow, populations to age, pollu-tants to work their way through the ecosystem,polluted waters to become clean again, capitalplants to depreciate, people to be educated or re-trained, the system can’t change overnight, evenafter it perceives and acknowledges the problems.To steer correctly, a system with inherent momen-tum needs to be looking ahead at least as far as itsmomentum can carry it. The longer it takes a boatto turn, the farther ahead its radar must see. Thepolitical and market systems of the globe do notlook far enough ahead.

The final contributor to overshoot is the pur-suit of growth. If you were driving a car withfogged windows or faulty brakes, the first thingyou would do to avoid overshoot would be to slowdown. You would certainly not insist on acceler-ating. Delays in feedback can be handled as longas the system is not moving too fast to receivesignals and respond before it hits the limit. Con-stant acceleration will take any system, no mat-ter how clever and farsighted and well designed,to the point where it can’t react in time. Even acar and driver functioning perfectly are unsafe athigh speeds. The faster the growth, the higher the

overshoot, and the farther the fall. The politicaland economic systems of the globe are dedicatedto achieving the highest possible growth rates.

What finally converts overshoot to collapseis erosion, aided by nonlinearities. Erosionis a stress that multiplies itself if it is notquickly remedied. Nonlinearities...are equivalentto thresholds, beyond which a system’s behaviorsuddenly changes. A nation can mine copper oredown to lower and lower grades, but below a cer-tain grade mining costs suddenly escalate. Soilscan erode with no effect on crop yield until the soilbecomes more shallow than the root zone of thecrop. Then further erosion leads rapidly to deser-tification. The presence of thresholds makes theconsequences of feedback delays even more seri-ous. If you’re driving that car with the fogged win-dows and faulty brakes, sharp curves mean youneed to go even more slowly.

Any population-economy-environment systemthat has feedback delays and slow physical re-sponses; that has thresholds and erosive mech-anisms; and that grows rapidly is literally unman-ageable. No matter how fabulous its technologies,no matter how efficient its economy, no matterhow wise its leaders, it can’t steer itself away fromhazards. If it constantly tries to accelerate, it willovershoot.

By definition, overshoot is a condition in whichthe delayed signals from the environment are notyet strong enough to force an end to growth. How,then, can society tell if it is in overshoot? Fallingresource stocks and rising pollution levels are thefirst clues. Here are some other symptoms:

• Capital, resources, and labor diverted to activi-ties compensating for the loss of services thatwere formerly provided without cost by nature(for example, sewage treatment, air purification,water purification, flood control, pest control,restoration of soil nutrients, pollination, or thepreservation of species).

• Capital, resources, and labor diverted from fi-nal goods production to exploitation of scarcer,more distant, deeper, or more dilute resources.

• Technologies invented to make use of lower-quality smaller, more dispersed, less valuableresources, because the higher-value ones aregone.

• Failing natural pollution cleanup mechanisms;rising levels of pollution.

174


• Capital depreciation exceeding investment, andmaintenance deferred, so there is deteriorationin capital stocks, especially long-lived infras-tructure.

• Growing demands for capital, resources, andlabor used by the military or industry to gainaccess to, secure, and defend resources thatare increasingly concentrated in fewer, more re-mote, or increasingly hostile regions.

• Investment in human resources (education,health care, shelter) postponed in order to meetimmediate consumption, investment, or secu-rity needs, or to pay debts.

• Debts a rising percentage of annual real output.

• Eroding goals for health and environment.

• Increasing conflicts, especially conflicts oversources or sinks.

• Shifting consumption patterns as the popula-tion can no longer pay the price of what it re-ally wants and, instead, purchases what it canafford.

• Declining respect for the instruments of collec-tive government as they are used increasinglyby the elites to preserve or increase their shareof a declining resource base.

• Growing chaos in natural systems, with “natu-ral” disasters more frequent and more severebecause of less resilience in the environmentalsystem.

Do you observe any of these symptoms in your“real world?” If you do, you should suspect thatyour society is in advanced stages of overshoot.

A period of overshoot does not necessarilylead to collapse. It does require fast and de-termined action, however, if collapse is to beavoided. The resource base must be protectedquickly, and the drains on it sharply reduced.Excessive pollution levels must be lowered, andemission rates reduced back to levels below whatis sustainable. It may not be necessary to reducepopulation or capital or living standards. Whatmust go down quickly are material and energythroughputs. In other words, the ecological foot-print of humanity must be lowered. Fortunately(in a perverse way), there is so much waste and in-efficiency in the current global economy that thereis tremendous potential for reducing the footprint

while still maintaining or even raising the qualityof life.

In summary, here are the central assumptionsin the World3 model that give it a tendency toovershoot and collapse. If you wish to disagreewith our model, our thesis, our book, or our con-clusions, these are the points to contest:

• Growth in the physical economy is considereddesirable; it is central to our political, psycho-logical, and cultural systems. Growth of boththe population and the economy, when it doesoccur, tends to be exponential.

• There are physical limits to the sources of mate-rials and energy that sustain the population andeconomy, and there are limits to the sinks thatabsorb the waste products of human activity.

• The growing population and economy receivesignals about physical limits that are distorted,noisy, delayed, confused, or denied. Responsesto those signals are delayed.

• The system’s limits are not only finite, but erod-able when they are overstressed or overused.Furthermore, there are strong nonlinearities—thresholds beyond which damage rises quicklyand can become irreversible.

Listing these causes of overshoot and collapsealso gives a list of ways to avoid them. To changethe system so that it is sustainable and manage-able, the same structural features have to be re-versed:

• Growth in population and capital must beslowed and eventually stopped by human de-cisions enacted in anticipation of future prob-lems rather than by feedback from external lim-its that have already been exceeded.

• Throughputs of energy and materials must bereduced by drastically increasing the efficiencyof capital. In other words, the ecological foot-print must be reduced through dematerializa-tion (less use of energy and materials to ob-tain the same output), increased equity (redis-tribution from the rich to the poor of the ben-efits from using energy and materials), andlifestyle changes (lowering demands or shiftingconsumption towards goods and services thathave fewer negative impacts on the physical en-vironment).

• Sources and sinks must be conserved and,where possible, restored.

175


• Signals must be improved and reactionsspeeded up; society must look farther aheadand base current actions on long-term costs and

benefits.

• Erosion must be prevented and, where it alreadyexists, slowed and then reversed.

Questions

1. Critical analysis. State the authors’ basicpremise and conclusion. How would you goabout verifying or refuting them?

2. What is overshoot and how is it caused? An-swer in some detail, and apply the concept toour use of Earth’s natural resources.

3. What is ecological footprint?

4. Describe the nature of the factors that limit theEarth’s carrying capacity.

5. Explain the concept of delay in systems mod-eling, and its role in producing overshoot.

6. What characterizes a ‘systems perspective’worldview?

7. Describe the four general ways in which agrowing human population can approach theEarth’s carrying capacity.

8. The authors state that ‘we do not put faith inthe precise numerical path the model gener-ates...[but] we do have faith in the dynamicbehaviors generated by World3.’ What do theymean by this distinction? What is the purposeof World3? What are its strengths and limita-tions? How may it help us?

9. The authors believe that, for the global hu-man economy and the natural resources onwhich it depends, overshoot is now unavoid-able. According to the author’s computermodels, what are the two possible modes in

which the human economic ‘system’ can in-teract with the Earth’s carrying capacity? De-scribe each mode in some detail.

10. According to the authors, three factors arenecessary and sufficient to produce overshoot:growth, limits, and delays. Explain how (in theauthors’ view) these three requirements aremet for the human population. In particular,explain the nature of the limits and delays indetail.

11. The authors state that Earth’s natural re-sources are ‘erodable.’ What are the conse-quences of this fact on the output of their com-puter model?

12. Two scenarios tested by World3 are shownhere: one with no limits, and one with ‘busi-ness as usual’ and a fixed carrying capacitythat has already been exceeded. Describe andexplain the behavior of the model in these twoscenarios, highlighting and explaining the dif-ferences.

13. The authors spend some time discussing thenature of the economic and ecological ‘sig-nals’ of natural resource limits, as well as thedelays in our perception of and response tothese signals. What are some examples of sig-nals given in the text? What are some of thereasons for the delays?

14. In some detail, describe and explain the dis-tinction between overshoot and oscillation andovershoot and collapse.

176

Reading 23 - University of Richmond

Documents

Transcript of Reading 23 - University of Richmond