THE SPACESHIP ECONOMY

by

R. U. AYRES*

P. FRANKL**

98/47/EPS/CMER

This working paper was published in the context of INSEAD's Centre for the Management of EnvironmentalResources, an R&D partnership sponsored by Ciba-Geigy, Danfoss, Otto Group and Royal Dutch/Shell andSandoz AG.

* Sandoz Professor of Management and the Environment at INSEAD, Boulevard de Constance, 77305Fontainebleau Cedex, France.

** Post Doctoral Fellow at INSEAD, Boulevard de Constance, 77305 Fontainebleau Cedex, France.

A working paper in the INSEAD Working Paper Series is intended as a means whereby a faculty researcher'sthoughts and fmdings may be communicated to interested readers. The paper should be considered preliminaryin nature and may require revision.

Printed at INSEAD, Fontainebleau, France.

THE SPACESHIP ECONOMYRobert U. Ayres and Paolo FranklCMER, INSEAD, France

Forthcoming in Environmental Science & Technology

Abstract

Long-term sustainability of the global economy requires continued growth in value-added butsharply reduced consumption of non-renewable natural resources. Information technology ishelpful, of course, but, in the long-run the global materials cycle must be closed.

We suggest that this imperative necessitates a spaceship economy based on a 'hydrogeneconomy'. The primary source of energy for all purposes (in the long run) must be the sun and themain carriers must be photovoltaic or nuclear electricity (for stationary uses) and photovoltaichydrogen (for mobile users). Fortunately most of the technological prerequisites, e.g. PV cellsand fuel cells are now being developed. We believe a new PV-hydrogen driven industrialrevolution is now in the early stages.

R. U. Ayres Spaceship Economy as printed June 3, 1998

Page 1

THE SPACESHIP ECONOMY

Robert U. Ayres & Paolo FranklCMER, INSEADFontainebleau France

June 1998

Background

In a recent "Viewpoint" article one of us (R. U. A.) suggested that long run sustainability ofhuman civilization on a finite earth is not compatible with perpetual economic growth basedon ever-increasing extraction and processing of natural resources, accompanied by the creationand disposal of wastes that are subsequently returned to the environment. This is becausemost environmental damages are due to emissions and wastes attributable to excessivematerials processing and consumption. We argued that economic growth itself must continuebut that a new and different growth mechanism is needed, consistent with radicaldematerialization and energy conservation — in short, a 'spaceship economy'.

The economic system of the spaceship earth must be one in which virtually allintermediate materials (including fuels) are either eliminated or recycled, while durable goodsare renovated, remodelled and remanufactured indefinitely. In short, our social imperative isto adopt, as a long-term strategy, closing the materials cycle. (Such a goal can never beactually reached, but it is no less reasonable as a goal than the recently highly touted goal,in quality circles, of "zero defects".)

Engineers and economists alike tend to dismiss such radical notions as 'blue sky'.Engineers are mostly concerned with ways and means of improving the performance of theproducts and production systems that exist here and now. Firms, and governments, spend alot of money on minor improvements of existing systems. The catalytic convertor forautomobiles, the automotive fuel modification program, and the flue gas desulfurizationprojects are illustrations of this tendency. Economists are equally skeptical. They tend toargue that if a truly better system were possible it would have been adopted already by someentrepreneur. While it is easy to caricature this argument, there is also some merit in it. Afree economy, even if not perfectly competitive, will not readily adopt radical changesinvolving large and certain up-front investments for uncertain long-term benefits.

The question we pose for ourselves, however, is the following: Is there a plausiblescenario that could get us from our present 'throughput' economy to a reasonableapproximation of a closed-cycle 'spaceship' economy within a century? By that, we mean aneconomy that has eliminated most dissipative uses of most materials, including fuels, whiletreating tangible durable goods as capital assets to be re-used, renovated, remanufactured and— as a last resort — recycled.

We believe there is such a scenario. It centers on a combination of moderately radicalapproaches to energy conservation and the accelerated development of existing and/orpotential (bio)technologies to exploit solar energy and reduce demand for metals. Thesetechnologies, in combination, can eventually produce massive energy savings, on the onehand, and material savings — especially in the realm of dissipative materials — on the other.

R. U. Ayres

Spaceship Economy as printed June 3, 1998

Page 2

The End-Point

The elimination of most dissipative uses of materials implies, first and foremost, theeventual elimination of hydrocarbon fuels. The 'substitutes' must be either conservation[Lovins 1996], or non-polluting energy. Nuclear power, both fission and fusion, still hasinfluential advocates. But the difficulties of safe nuclear waste disposal and effectiveprotection against misuse for weapons may be insurmountable in the long run. In any case,other non-polluting sources of energy are available, including hydropower, wind power, andboth thermal and photovoltaic (PV) solar energy. Cultivated biomass may not be a long-termsolution but clearly there is no fundamental objection to utilizing waste biomass for electricitygeneration, especially by means of small aircraft-type gas turbines [Williams & Larson 1993].We see no fundamental difficulty in providing adequate non-polluting electric power forstationary uses.

However, there remains a problem of providing a substitute for hydrocarbon liquid fuelsfor mobile power. There is a non-polluting closed cycle starting from and returning to CO2.This is the natural photosynthetic carbon cycle carried out by green plants. Alcohols orvegetable oils from biomass are technically feasible for a relatively low level of usage (butonly at very high prices, unless there is an unexpected a breakthrough in biotechnology). Alsothere will probably be higher value uses for most biomass.

In the long run there is much to recommend hydrogen gas as the ultimate energy carrier.It offers the possibility of a truly non-polluting closed cycle: 2H20 2H2 + 02 2H20.The critical exothermic step requires either thermo-chemical or electrolytic reduction of water.There are a number of possible thermo-chemical processes, one or more of which may provepractical when more research is done. However, solar hydrogen — using solar power fromphotovoltaic (PV) cells — may be the simplest and most practical approach. The electrolyticcell is a hydrogen fuel cell operating in reverse. We return to this topic below.

The problem of substitution/recycling of other intermediates has no universal solution,but there are some obvious directions. For instance, many if not most solvents can beconserved by recycling. This is already happening for commercial dry-cleaning fluids and ona smaller scale for lubricating oil, antifreeze and brake fluids. All of these can be recovered,rerefined and recycled at low cost. This sort of recycling is essentially a matter of collection,filtration and redistillation — a simple substitution of capital and energy for materials. Manymetals are already recycled. Plastics can be recycled also, though with much more difficulty,usually with some degradation. The recycled (and recyclable) fraction for all of these productscan be increased in a variety of ways, however. The problems are primarily related tocollection, identification and sorting. The solutions are partly in the domain of information-technology, which will certainly improve with time, and partly in the domain of governmentregulation. The sorts of regulation needed (take-back, deposit/return, minimum recycledpercentage, etc.) are resisted by primary producers and, often, by retail outlets. But timeschange.

Whereas potentially dissipative materials must be recycled, durable goods and capitalequipment must be maintained, renovated and upgraded on a programmed basis andremanufactured at the end of useful life.

Biotechnology has an important role to play. Chemical pesticides can be largely replacedin the long run by bio-engineered bacterial or viral agents, or organic biocides, tailored tospecific targets organisms on specific crops. These agents should, of course, be administeredonly by trained personnel in the context of "integrated pest management" services. Firms likeNovartis are already pioneering such services, but the "crop doctor" of the late 21st centurywill be a highly educated specialist like the medical doctor of today.

R. U. Ayres Spaceship Economy as printed June 3, 1998

Page 3

Cleaning agents like detergents are being continuously improved, and quantities requiredcan be reduced significantly. Fabrics can be made that are almost impervious to dirt. Buildingmaterials can be made with anti-corrosion surfaces and colors built in, rather than added on.The possibilities are endless. However, this short article cannot cover all the interesting topics.For the remainder we focus strictly on the non-polluting energy system of the late 21stcentury, and how we might "get there from here".

The Trajectory

For irreducible stationary energy needs the primary contribution to "getting there fromhere" will be 'negawatts' (i.e. conservation). The opportunities are far greater and morepervasive than the energy industry has ever yet acknowledged [Lovins & Lovins 1997].Beyond this, there is an obvious trend towards electrification of the economy and increaseduse of renewable energy sources, including agricultural wastes, wind and solar energy.

The key enabling technologies are photovoltaic PV power, fuel cells and hydrogenstorage. In the near-to-medium term PV power will be an increasingly important source ofelectric power. In the near-to-medium term, hydrogen will gradually become the non-pollutingsubstitute for liquid fuels for motor vehicles and aircraft. It can be obtained easily from fossilhydrocarbons by catalytic means. (Of course this does not eliminate the CO2 but theefficiency of electric drive is high enough to enable significant reductions in overall CO2generation per unit of mechanical work done.)

The PV and fuel cell technologies are already quite well advanced. The technical andeconomic barriers that impeded the development of practical low cost PV cells have now beenlargely overcome [e.g. Frankl 1998]. There are several different competing approaches (e.g.crystalline silicon, amorphous silicon, and thin films of various semiconductor combinations),a number of promising R&D avenues have already been identified, most of the formertechnical barriers have now been overcome (some very recently), engineering developmentis already far advanced, products are on the market and the process of adoption and diffusionhas begun. There is already a small but growing market for rooftop PV units in new structuresand in older structures undergoing major renovation. In these cases the PV units wouldgenerate surplus power during part of the day (on sunny days) as well as providing power forair-conditioning and water heating. This will reduce the need for new and costly fossil fuel-fired central power plants, as well as for fuel.

The main problem now is to reduce manufacturing cost to the point where PV cancompete economically in a sequence of increasingly big market segments. The PV market isgrowing explosively (Figure 1). Economies of scale and the "learning curve" will continueto bring costs down rapidly (Figure 2). That, in turn, will drive further market penetration.As PV sales grow the price will continue to fall. A decade or two hence the PV systems willbe increasingly integrated into the central systems themselves. It is estimated that PV power,utilizing rooftops, highway sound barriers and other public spaces could replace up to 20%or even 25% of total generating capacity by mid-century. This would require no operatingsubsidy. However, the process could be accelerated by (1) reducing the existing direct and(especially indirect) subsidies to conventional power systems which exist in many countries,(2) subsidizing R&D on PV, especially in the area of large-scale manufacturing.

Meanwhile, in the real world, energy companies have made their own calculations andlarge-scale investments have begun to flow. Several big PV projects in the multi-hundredmillion dollar range are now under way. Enron Corp. AMOCO and Siemens-Solar are bigplayers. BP Solar has announced a commitment to achieve revenues of $1 billion per year by

R. U. Ayres

Spaceship Economy as printed June 3, 1998 Page 4

2007 and to bring manufacturing costs down to a level consistent with large-scale use inbuildings.

Governments are beginning to take notice also. The Netherlands has a solar village. Japanunderwrites half the cost of rooftop installations and Japan's MITI expects 400MW ofinstalled capacity in Japan alone by 2000. The European Commission has tabled a ten yearprogram to install 500MW of capacity by 2005, costing $600 million. President Clintonannounced this year a program (unfunded, however) to install 1 million rooftop units in theUS by 2010.1 The state government of Rajasthan in India is hoping to leapfrog all the rest andto become the world leader in generating solar electricity.

The PV story is interesting in itself, but it is not yet a "major" industrial revolution. Whatmakes it potentially revolutionary is the synergistic combination with other relatedtechnologies. The second one is hydrogen fuel cell technology, exploiting the proton exchangemembrane (PEM), which is a solid electrolyte that operates at room temperature. There weretwo technological breakthroughs involved. One was the membrane itself, a polymer thatallows hydrogen ions (i.e. protons) to pass through freely but is impermeable to hydrogenmolecules. The result is a kind of one-way valve for electric current.

The other breakthrough was really an evolutionary development that owes a great dealto the mass-produced automotive catalytic muffler. It is, in brief, the capability to make alittle platinum catalyst go a long way by plating it in extremely thin monatomic layers. Theresult is that the cost of the catalyst is no longer an insurmountable obstacle: when the PEMtechnology was first developed (by GE) the cost of the platinum needed for a cell big enoughto power a small car would have been $30,000. The subsequent development effort, lastingfifteen years, was led by Ballard Power Systems Inc., a small Canadian firm. Thanks to heavyinvestment (funded largely by Daimler Benz) and collaboration with Johnson-Matthey, acatalyst manufacturer, the cost of platinum catalyst for a car is already down to $140 andfalling, although the fuel cell itself — as opposed to the catalyst — is still far too expensivefor general use [The Economist Oct. 25 1997 p. 104].

For use in a vehicle, the fuel cell would replace the large and cumbersome battery-packin the sort of electric car that is already under development by several auto companies. Thatis, the electric power would power electric motors in the wheels. An electric car usingelectricity generated by a central station can be twice as efficient as a gasoline-powered car.With an on-board fuel cell, there would still be a significant efficiency advantage, probablyin the neighborhood of 50%. This is the major selling point for users.

There is no fuel cell powered electric vehicle on the market yet, but Toyota and Daimler-Benz have both demonstrated prototype vehicles at the Japanese Auto Show, using methanolas fuel. The methanol is catalytically reduced in a small reactor on-board to hydrogen. Butmethanol is not an ideal fuel because the other product is carbon monoxide, which must eitherbe oxidized to CO2 or reacted with steam at high temperatures to produce more hydrogen.Chrysler's version will use gasoline as a source of hydrogen, but it will also generate CO2.Chrysler has announced plans to go into large-scale production of a fuel-cell powered van by2005. Fuel cells will fall in price in much the same way that PV cells and panels will. Thepresent cost of a PEM cell is about $5000 per kilowatt, but this is mostly because each unitis virtually hand-made. The key to cost reduction from this point is increasing scale. Expertsbelieve that $200 per kilowatt is the threshold for profitable large-scale use. Ballard andDaimler-Benz believe this can be achieved at a production volume of 250,000 per year.

At the moment PV-hydrogen technology looks very expensive and rather remote. It willnot be practical for several decades to come. This is partly because it is so much cheaper toobtain hydrogen from hydrocarbons. (However, the latter possibility is one of the incentivesthat is already encouraging accelerated development of fuel cell technology, after many years

R. U. Ayres Spaceship Economy as printed June 3, 1998 Page 5

of neglect. And the development of efficient and mass producible fuel cells for automobileswill generate technological spinoffs that, in turn, help bring electrolytic cells into thecompetitive range.) Indeed, these factors in combination induced Siemens Solar AG andWasserstoff Bayern AG to build the world's first pilot test plant, in Germany, for producingsolar hydrogen (and oxygen) from PV cells. The plant has been designed to test a largenumber of combinations of technologies, from the PV cells themselves to purification of thefinal products, to allow for future upgrades.

Thus, in combination with mass-produced PV cells and PEM fuel cells, the PV-hydrogentechnology could conceivably develop rather rapidly. The necessary trigger for investmentwould be the development of a large market for hydrogen. Hydrogen from water is anobvious way to store solar energy, and also to transport it by ship or pipeline. The first large-scale applications might occur in Scandinavia or Alaska, where hydrogen could be generatedduring the summer and consumed during the winter. An interesting two-way trade inhydrogen and water could develop in some locations. Floating hydrogen factories capable ofmoving from the Arctic to the Antarctic, and back, to exploit the long summer days and theabundant fresh water, might be envisaged for the more distant future.

For purposes of vehicle propulsion, some sort of safe and lightweight hydrogen storageis needed. Methanol is one possibility of course. But several new technologies are emergingthat offer potentially better solutions. One of them is some version of activated carbon thatwill adsorb hydrogen on it's very large internal surface. Metal "sponges" — even "buckyballs"— are another possibility. However the details do not matter here. The point is that there isa large potential market and several plausible possibilities.

Incidentally, if large-scale production of hydrogen precedes large scale production ofsmall portable fuel cells (which is also a possibility), future automobiles may simply burnhydrogen instead of gasoline in internal combustion engines very similar to the present ones.BMW is apparently betting on this.) Hydrogen powered gas turbines have already been takenvery seriously by the aircraft industry. And, of course, hydrogen would replace natural gasin the pipelines that serve cities.

Summary and conclusions

To recapitulate, the core argument of this paper is that this combination of technologicalpotential and demonstrated demand (reflecting a societal need) will trigger a new industrialrevolution of the first magnitude. We believe it will be comparable in with the earlierindustrial revolutions (steam power and railways, electric power, internal combustion engines,and information technology) in terms of unleashing a wave of "creative destruction", replacinglong established technological dinosaurs with new ways of doing things. A host of newperipheral technologies will develop.

Eventually, perhaps by the year 2100, hydrogen could completely replace fossil fuels,either in fuel cells or in conventional ICEs. The potential environmental benefits areenormous. Since the only combustion product of hydrogen is water vapor, the buildup of CO2will end, environmental acidification (from SOX and N0x) will end, there will be no moresmog, or smoke, and cities will finally be clean. Indeed, this prospect is so appealing that,rather than looking for excuses to delay action on CO 2, governments should be looking forways to accelerate it.

What of costs? In the long run, the question is essentially meaningless, in terms ofcurrent monetary units, since the changes envisioned are so radical that a completely differentset of prices would necessarily exist in the new economic equilibrium. In the short term, it

R. U. Ayres

Spaceship Economy as printed June 3, 1998 Page 6

is more a question of redirecting R&D (most of which is still focussed on streamliningdinosaur technology) and of using public policy to accelerate the creation and penetration ofnew markets. Once that is done, the dynamics of the learning curve and economies of scalewill bring costs down dramatically.

To be sure, none of the standard energy-economic forecasting models 2 predict such adevelopment. The reason they do not (cannot) is that most energy forecasting models assumeeither that PV (indeed all) costs are forever fixed at present levels or they assume anexogenously given linearly declining cost curve as a function of time, as shown in Figure 3[Messner 1997]. These models optimize for each time period by linear programming, takinginto account capacity limitations (based on investment assumptions) and prices.

But it is extremely important to note that the common device of introducing a linearlydeclining cost curve that does not depend on usage results in scenarios where a technologyis introduced after some time has passed, without any prior investment. (This is equivalentto the assumption of exogenous technological change without cost). But the assumption ofdeclining costs without corresponding investment also underestimates the potential marketshare of a new technology and the rate of market penetration that can occur.

A recent energy cost minimization model that incorporates the non-linear cost-reductionby learning curve (Figure 3) compares three scenarios [ibid]. In the first scenario, PV costsnever fall below present levels and — as a consequence — the PV technology is neveremployed to a significant extent. In the second scenario, the cost are assumed to fall at aconstant predetermined rate of 3% per year. In this case the model predicts PV penetrationof 8% of the energy market by the year 2050. In the third and most realistic scenario,however, PV costs fall according to a typical experience curve. The result is projected marketpenetration of nearly 20% by 2050. We think this is the most likely future.

Endnotes

1. For background details see the cover story in Tomorrow, VII (6) December 1997.

2. The best-known energy optimization models are ETA-MACRO [Manne 1977], MARKAL [Fishbone et al1983], EFOM [Van der Voort et al 1985], and MESSAGE [Messner & Strubegger 1994]. ETA-MACROwas the first energy optimization model to be linked with a macroeconomic forecasting module. Other suchcombinations include MARKAL-MACRO, which uses the same economy module as ETA-MACRO andETSAP, which also uses MARKAL [Kram 1993].

References

[Ayres 1996] Ayres, Robert U., "Limits to the Growth Paradigm", Ecological Economics 19, 1996 :117-134.

[Ayres 1998] Ayres, Robert U., Turning Point: An End to the Growth Paradigm, Earthscan, London, 1998.

[Braungart et al 1990] Braungart, Michael et al. Concept Proposal for an Intelligent Products System, EPEA,Germany, 1990.

[Brundtland 1987] Brundtland, Gro H. (ed), Our Common Future, Oxford University Press, New York, 1987.(Report of the World Council For Economic Development)

[Factor 10 1994] Factor Ten Club, Carnoules Declaration, Factor Ten Club, Wuppertal, Germany, October 1994.(Chairman; F. Schmidt-Bleek)

Spaceship Economy as printed June 3, 1998 Page 7R. U. Ayres

[Fishbone et al 1983] Fishbone, L G., Giesen, G. A., Goldstein, H. A., Hymen, H. A. et al, User's Guide forMARKAL, Technical Report (BNL-46319), Brookhaven National Laboratory, Upton NY, 1983.

[FOTE undated] Friends of the Earth. Ten Times Better: Fair Shares in Environmental Space, Friends of theEarth Netherlands, Amsterdam, undated.

[Frankl 1998] Frankl, Paolo. "Photovoltaics", in: Ayres, Robert U. et al(eds), Ecorestructuring, UNU Press,Tokyo, forthcoming, 1998. (incomplete reference)

[Fussler 1996] Fussler, Claude, with Peter James, Driving Eco-Innovation, Pittman Publishing Company, London,1996.

[Goulder 1995] Goulder, Lawrence H., "Environmental Taxation & the Double Dividend: A Reader's Guide",International Tax & Public Finance 2, 1995 :157-183.

[Kram 1993] Kram, T, National Energy Options for Reducing CO2 Emissions, Annex 1V (190-1993)(ECN-C-93-101), Energy Technology Systems Analysis Program, Netherlands Energy Research Foundation,The Netherlands, 1993.

[Lovins 1996] Lovins, A. B. "Negawatts: Twelve Transitions, Eight Improvements and One Distraction" EnergyPolicy 24(4) 1996

[Lovins & Lovins 1997] Lovins A. B. & L. H. Lovins "Climate: Making Sense and Making Money" RockyMountain Institute, 13 Nov. 1997

[Marine 1977] Manne, Alan S.. "ETA-Macro", in: Hitch, C. J.(ed), Modeling Energy-Economy Interactions: FiveApproaches (Series: Research Paper R5), Resources for the Future, Washington DC, 1977.

[Meadows et al 1972] Meadows, Donella H, Dennis L. Meadows, Jorgen Randers & William W. Behrens III,The Limits to Growth: A Report for the Club of Rome's Project on the Predicament of Mankind, UniverseBooks, New York, 1972.

[Messner 1997] Messner, Sabine, "Endogenized Technological Learning in an Energy Systems Model", Journalof Evolutionary Economics, 7, 1997 :291-313

[Messner & Strubegger 1994] Messner, Sabine & M. Strubegger. "The Energy Model MESSAGE DI", in: Hake,J. F. et al(eds), Advances in Systems Analysis: Modelling Energy-Related Emissions on a National &Global Level :29, Forschungszentrum Juelich Gmbh, Juelich, Germany, 1994.

[Rogner 1998] Rogner, Hans-Holgar. "Global Energy Futures: The Long-Term Perspective for Ecorestructuring",in: Ayres, R. U. et al(eds), Ecorestructuring, UNU Press, Tokyo, forthcoming, 1998. (incomplete reference)

[Schmidheiny 1992] Schmidheiny, Stephen with the Business Council for Sustainable Development, ChangingCourse: A Global Perspective on Development & the Environment (ISBN 0-262-69153-1), The MIT Press,Cambridge MA, 1992.

[Stahel 1982] Stahel, Walter R.. "The Product Life Factor", in: Orr, S. G.(ed), An Inquiry into the Nature ofSustainable Societies (Series: 1982 Mitchell Prize Papers), Houston Area Research Center, Houston TX,1982.

[Stahel & Reday-Mulvey 1981] Stahel, Walter R. & Genevieve Reday-Mulvey, Jobs for Tomorrow: ThePotential for Substituting Manpower for Energy (ISBN 0 533-04799-4), Vantage Press, New York, 1981.

[Van der Voort et al 1985] Van der Voort, E. et al, Energy Supply Modeling Package - EFOM 12C MARK I,(EUR 8896 EN), CEC, 1985. (3 volumes)

R. U. Ayres Spaceship Economy as printed June 3, 1998

Page 8

[von WeizsAcker et al 1994] von Weizsacker, Ernst Ulrich, Amory B. Lovins & L. Hunter Lovins, Faktor Vier(ISBN 3-426-26877-9), Droemer Knaur, Munich, 1994.

[Ward 1966] Ward, Barbara, Spaceship Earth (LC 66-18062), Columbia University Press, New York, 1966.paperback edition.

[Ward & Dubos 1972] Ward, Barbara & Rend Dubos, One Earth: The Care & Maintenance of a Small Planet,Penguin Books, Harmondsworth, UK, A Pelican Book, 1972. paperback edition.

[WBCSD 1995] Ayres, R. et al (rapporteurs). Report of the World Business Council for SustainableDevelopment, 2nd Antwerp Eco-Efficiency Workshop, WBCSD, Antwerp, Belgium, November 1995.

[Williams & Larson 1993] Williams, Robert H. & Eric D. Larson. "Advanced Gasification-based Biomass PowerGeneration", in: Johansson, T. B., H. Kelly, A. K. Reddy & R. H. Williams(eds), Renewable Energy:Sources for Fuels & Electricity, Chapter 17, Island Press, Washington DC, 1993.

Spaceship Economy as printed June 3, 1998 Page 9R. U. Ayres

Source: Y. Kuwano, "Let's utilize Photovoltaiccells", Kohdansha Blue-Backs 1992 (in Japanese)

Figure 1. Growth of photovoltaics

50ee

World 46.5 •1•.5 •I• AO• 470Ie

eIIII.

11

10

11

11

11

0

10 1

11

11

10 1

11

11

11t

II

r

,1

11 J1

11

11

ni

l'Al

10

11, 1.111111:

f. .0

0

•••

0:4C.

•4°.r,A

.•

•

I•

1,_-...ir....

••

•0)

1:J

40

pan 16.8

US 14.8

10 EC 10.2

01978 1980 1982 1984 1986 1988 1990

100

50

20is4

10

• 50•E

0.4 2too0CU• 1

7.1

03 0.5cu

0.2

R. U. Ayres

Spaceship Economy as printed June 3, 1998

Page 10

1976

.....--., ......,.........., .

illi • • .• 19944 ••••• Crystalline

silicontechnologies

Thin filmtechnologies

...

/".

10,000

100,0000.1

0.1 1 10 100 1,000

Total installed PV capacity (MWp)Figure 2. The PV "learning curve" Source: [Frankl 1998]

1000 —

6000

5000 —

.,,...„,.....

4000 — ..,..,. .,.....

..,....,.

3000 — .,,.,.............

..,..........

........

••••••., Dynamic.......

2000 —

Learning.....

...•

R. U. Ayres Spaceship Economy as printed June 3, 1998

Page 11

0 I I I I 11990 2000 2010 2020 2030 2040 2050

Year

Figure 3. Specific investment costs for PV systems for the dynamic & learning cases, 1990-2050. Source: [Messner 19971