Read Goldemberg_85_Kilowatt_Per_Capita.pdf text version

R()", \,

8 ~ I

S.\\t~~..:'iS!j --\rADEMY







Basic Needs and Much More With One Kilowatt Per Capita





Reproduced with permission from AMBIO



and Much.'


Conventional sumption Th ment. global


One Ki lowatt

g holds for h er

II I y,



energy supplies to meet the energy requirements f<;,r this scen.ario at !easonable costs and wIthout major environmental and/or security problems. the Middle East and North African countries, this level of energy use could

not gas. with as

thinkin is a prerequisite b I. f IS e Ie , toget




ener social 0 of

t h

gy condevelopdl .Outside win Ing

economic and h h Wit t e prospect




.~: ,.,i



supply genera



d ea







I . e leve


..gree at It IS

not be supplied

domestically estimated the

to a significant

produced recoverable

oil oil

and and









the resources sIde the

in developing East and

countries North



developing carriers more


countries. and b y ex


p loitin

But by shifting g cost-effective use, It would

P rovide


high-quality opportunities to

energy for satisfy








could supply energy requirements for only about. two decades (3). . If, Instead, coal were emphasIzed, many devel.oping count~ies wo~ld become major

coal of l':!1porters, China, few Since, developing ~Ith the countries e~ceptlon have



improvements ...

Increasing per-capita




energy use


b a ove t h


I e present eve. I


major major expansion global

coal resources

of climatic coal change

(4). Moreover,

use in could a matter cause of


PerhJlps the greatest challenge facing mankind is to find ways to bring a decent standard of living to the majority of the world's population who now live in abject poverty. The poor countries of the "South" account for three-quarters of the world's population but have per-capita incomes averaging only one-tenth that of the rich countries of the "North..' And in the South there are enormous disparities between the elites, who typically account for 10 to 15 percent of the population and one-third to one-half of all income, and the much poorer majorities. It is widely believed that an essential feature of any development program aimed at reducing this poverty must be a significant increase in the level of per capita energy use. This would seem to be selfevident from the strong historical correla.tion between energy use and gross national product (GNP) and the global disparities in energy use levels, While in 1980 per capita primary energy use averaged 6.3 , kilowatts (kW) in industrialized countries. the average in developing countries was only about 1,0 kW, including 0.4 kW of noncommercial energy use, the sources of which are becoming ever scarcer because of ongoing deforestation and the pressures of population growth. Indeed. planning efforts in developing countries have been emphasizing energy supply expansion. and a result of this is that per-capita commercial energy use in d~\eloping countries grew at a vigorous


average annual rate of 3.6 percent per year during the 1970s (1). But this energy growth has been very costly, especially for the oil-importing developing countries, which accounted for half of the increment in energy use by developing countries during this period. For these oil-importing countries. half the increment in energy use came from increased oil imports, which grew at an average rate of 6.3 percent per year in the 1970s. By 1981. low- and middle-income oil-importing developing countries were spending an average 61 and 37 percent of their export earnings on oil imports. respectively (2), If the per-capita commercial energy demand growth rates of the 1970s were to persist, the average per-capita rate of primary commercial energy use in developing countries would increase from 0.55 kW in 1980 to 2,3 kW in 2020. Because the population is expected to nearly double by then. aggregate commercial energy use in developing countries would increase from less than two terrawatts (TW)in 1980 to nearly 15TWin 2020. The increment in energy use by developing countries in this period with this scenario is equivalent to 1,3 times total world energy use in 1980. or three times world oil production. or five times world coal production, or 7.5 times that for natural gas. nearly nine times that for bioenergy. and nearly 60 times that for nuclear energy. It would be exceedingly difficult to increas~

decades (5). To meet half of the developing countries' incremental energy requirements for this scenario with nuclear energy would require building about 100 large nuclear power plants a year between 1985 and 2020. Aside from the financial challenges of this undertaking-owing to the fact that nuclear power has proven to be far m.ore costly than originally expected-such wldespread use of nuclear power would entail' major risks of nuclear-weapon proliferation and nuclear blackmail by terrorists. because the plutonium generated in reactor operations is both a nuclear fuel and a material from which nuclear weapons can be fabricated. By 2020 some 3,5 million kilograms .of we.ap°!1s-usable plutonium ~ould be ~lrculatlng.1n nuclear co~merc.e In dev~loplng countries each. year w,lth this scenarIo. Only 5 to 10 kg I~ re9ulred to !l1ak~ a !1uclear .weapon. It I~ dlffi.cult, to I~aglne internatIonal and natIonal InStl,tUlions capable of adequately s~feguardl!1g essentially 100 percent of this material against occasional diversion to nuclear weapons purposes. Hydropower and biomass are pr?mising renewable resources that already In 1980 accounted for nearly half of primary energy use in developing countries. But without major technological breakthroughs. these and other renewable resources would also be able to make only relatively. minor contributions to the energy supply requirements of this scenario. Although developing countries ha\.e thus far exploiled only seven percent of their hydAMIIIO \"(11. .01 ~() oI-C

.:; ~;.

:::': ,.


Mo re ~ 60 0 .6 6 a:.


a P




0 60




~ !:!









Figure 1. The Physical Quality of Life Index vs. total (commercial non-commercial) per-capita energy use (11).

roelectric resources, the total economic potential is only about 6000 TWhours per year or 0.7 TW (6). The potential for exploiting biomass resources is limited by land use constraints dictated by the low efficiency of photosynthesis. For example, the total rate of above ground wood production in all the forests of developing countries is small compared to the energy requirements of this scenario (on the order of three to four TW) (7), and only a small fraction of this resource could potentially be exploited for bioenergy purposes, as forests must serve multiple purposes, including the preservation of wildlife habitat. Even more modest scenarios for future energy growth would pose formidable challenges. The World Bank in 1983 estimated that in order to bring about a targeted 2.5 percent annual growth in percapita commercial energy use from 1980 to 1995 investments in new energy supplies for all d~eloping countries would have to average some $130 billion per year (in 1982 U.S. dollars) between 1982 and 1992. Half of this investment would have to come from foreign exchange earnings, requi ring an average annual increase of 15 percent in real foreign exchange allocations to energy supply expansion in this period. And despite this ambitious targeted energy supply expansion effort, the Bank projected that oil imports by oilimporting developing countries would still increase by nearly one-third, to nearly eight million barrels of oil per day by 1995 (1). The staggering costs of providing such

Table 1. Per-capita the Latin American

increases in energy supply would lead many to believe (but rarely to state) that it is not feasible to improve living standards substantially in developing countries. The gloomy outlook indicated by this energy analysis is not inevitable, however, because it arises from the assumption that major improvements in human welfare require considerable increases in the level of energy use. This assumption should not be blindly accepted, because the consumption of energy is not an end in itself. Increased energy use is valuable only insofar as it improves the quality of life by providing desired energy services such as cooking, lighting, water heating, space heating and cooling, personal and freight transport, indust rial process heat, motive power. etc. How much energy is needed in the future depends on the underlying goals of developing countries, the energy services required to meet these goals, and the technological choices available for providing these energy services-matters to which we now turn. ENERGY FOR BASIC HUMAN NEEDS In the 1950s,when development strategies were first being articulated, it was generalIy felt that maximizing economic growth was the best way to eradicate poverty; but the benefits of rapid economic growth have not trickled down to the poor to alleviate their plight. While rapid growth is a necessary condition for successful development, it is not sufficient. A more effective way of dealing with poverty is by

directly allocating resources to the satisfaction of basic human needs of the poorest -thereby ensuring that minimum standards for nutrition, shelter, clothing, health, and education are met (8). There is no empirical evidence that targeting basic hum~n needs would lead to slower ec~nomic growth (9), and there are theoretlcal grounds for believing that a basic human needs policy would lead to more rapid growth because of the resulting increase in worker productivity (10). The allocation of sufficient energy to basic needs programs to ensure that the various needs are satisfied is thus of central importance in energy planning. A high priority for analysis is to estimate the energy requirements for such basic needs pro-~ grams. A phenomenological approach to this analytical problem (11) involves examining the correlation between per capita energy use and the Physical Ouality of Life Index (POLl). The POLl is an index that focuses on three very basic measures of well-being: infant mortality rate, life expectancy, and literacy. In developing the POLl every country is first assigned an index for each of these measures in the range 1 to 100, with the lowest and highest values corresponding to the worst and best performances in the world. respectively. The POll for a given country is then obtained as the arithmetic average of these three indices. When the POll is plotted against per-capita energy use (commercial plus non-commercial) for a large number of countries, it is found that on average a

E/GNP correlations and the use of

energy requirements ass~iated with the satisfaction World Model for future economic growth (12).

of basic human needs, (BHN) based on historical

1970 Per-Capita GNP (1960 $) 440 154 112 Energy use' (kW) 1.1 0.7 0.7 Commercial energy intensity in 1970 (Watts per 1960 $) 1.67 1.83 2.89

Required Increment GNP" (1960 $) 369 405 394'

in per-caplta Energy use" (kW) 0.6 0.7 1.1 Date by which BHN could be setlsfledd 1992 2008 2020 Per-capita energy use required to satisfy BHN (kW) 1.7 1.4 1.8

Region Latin America Africa Asia

.Includes an estimated 0.4 kW per-capita of non-commercia! energy use. "This is the increment in per-capita GNP. above the 1970 level, required for the satisfaction World Model. " Based on the commercial energy intensity in 1970

of basic human needs, as estimated

with the Latin American

.As estimated with the Latin American World Model, assuming that implementation of the BHN policy begins in 1980. .For the case in which the maximum annual yield of edible products is assumed to be increased from 4 to 6 tons per hectare. socIety.

to avoid collapse

of the




POll of ahout 911 value typical of inuus(a trializ~d countrie\) i\ r~acheu ulr per-capita energy u\e ratcs of 1.0 to 1.2 kW. and that further increases in energy use causc only very marginal further increases in the POll (Figure I). It should be noted. however, that there is a considerable scatter in the data. For an energy use rate of 1.0 kW, the POll ranges from 60 to 90; for 0.5 kW, from less than 20 to more than 80. A more fundamental approach to the pr~blem involves estimating the energy requlrements associated with a future course of economic growth that targets the satisfaction of basic human needs. The Latin American World Model advanced by the Bariloche Foundation (12) can be used to estimate future economic requirements for meeting basic h.uman needs. In this model the path of future economic growth is determined by distributing capital and labor among the various .economic sectors in ways that would ~a~lm~ze lif~ expec~ancy at birth. This o~tlmlzatlon cntenon ISassumed to appro XIma~ethe goal of maximizing output so that basIc human needs are satisfied as quickly as possible. The model also involves sectoral targets for nutrition, shelter, and education: a daily intake of 3,000 kcal and 100 grams of protein per person; seven square meters of housing per person; and 12 years of basic education for all persons between ages six and 17. The increment in per capita GNP required to satisfy basic needs and the time required to reach this level, as determined by the Latin American World Model, are indicated in Table 1 for Latin America, Africa,estimate of th An and Asia. .t lreme~ s b .e henergy requd f . -or meeting aslc uman nee s usIng .this model is calculated as follows: Eb/C+ (CEI x 6GNP/C), where Eb/C is the tota~ (commercial p~usnon-commercia.l) per-capIta energy use m the base year [m (say) Watts per capita], CEI is the present average commercial energy intensity of the ( .lights economy In Watts per dollar), and 6GNP/C is the increment of per-capita GNP (in dollars per capita) required to satisfy basic human needs, as determined via the Latin American World Model. The per capita enerresults of this exercise are gy ,:,serates of 1.7,1.4, and 1,8 kW, for Latm America, Africa, and Asia, respectively (Table 1) or a population-weighted , ave~age energy u~e rate of 1,75 kW per capita, nearly twIce the present average f d I ..anu or eve Oping cou."tnes: , .Raw One problem WIth this analysIs IS that the energy intensity of an economy g basic human needs ma y oriented to ff t d servin f h f h b e quI e I erent rom t at 0 t e present economy, so that the use of present energy intensities in the calculation is questionable.

ate that are tieing developeu. as \\~ \hall ntl\\ sho\\'o -~rn ENERGY-EFFICIENT TECHNOLOGY The en~rgy crises of the IlJ70s have led to a revolution in the technology of energy end use. New end-use technologies that have recently become commercially available or that could become commercially available over the next several years make it possible to provide energy services with far less energy input than is possible with technologies now in widespread use. While some of these developments -notably those relating to more efficient cooking stoves (13}-have taken place in developing countries, most of the innovations have been introduced in industrialized countries and are often perceived as being relevant only to those societies. Yet many of these new technologies are of fundamental importance to developing countries, to the extent that emphasizing energy efficienc"J:' in ~evelopment plann!ng would make It possible not only to provIde the energy required for basic needs but also to provide considerable further improvements in living standards. In short, the pursuit of energy efficiency could free the developing world from the gloomy prospect suggested by the above analysis that energy is a fundamental constraint on the course of future development. To indicate in a very dramatic way the potential importance of energy efficiency for developing countries, we have constructed an energy scenario for a hypothetical developing country with a mix of ener-

gv-u\ing activiti~s similar to that for \\"~\tEurope (I~) in the IlJ711s I~xcluding \pace heating. which is not needed in mo\t developing countries) but is matched to much more efficient end-use technologies than those in common use in Europe. The activity levels for this scenario are indicated for all energy-using sectors in Table 2. The per-capita energy use associated with each activity in our scenario is the product of the activity level shown in Table 2 and an energy intensity corresponding in energy efficiency to either the best technology currently available on the market or to an advanced technologv that could be commercialized in about IO.vears (Table 3). The resulting total final energy use per capita, obtained by summing overall activities, is about 1.0 kW (Table ~), or only about 20 percent more than the average final energy use rate in 1980! It is possible to achieve large improvements in the living standards characterizing t~is scenario without increa~ing energy use, m part because enormous Increases in energy efficiency arise simply by shifting from traditional, inefficiently used. noncommercial fuels (which at present account for nearly half of all energy use in developing countries) to modern energy carriers (electricity, liquid and gaseous fuels, processed solid fuels, etc.). The importance of modern carriers is evident in Western Europe, where non-commercial fuel use is very small. Per-capita final energy use for purposes other than space heating in 1975 was only 2.3 kW, about 2.5 times that in developing countries. even

Table2. Activity levelslor a hypotheticaldevelopingcountryIn a warmclimate,with amenities (except lor spaceheating)comparable those in the WE/JANZ'region in the 1970s. to Activity Residential" Cooking Hot:.vater. R.efrlgeratlon IV ClothesWasher Commercial TranspOr1ation Automobiles Intercity Bus. Passenger Train UrbanMassTransit Air Travel Truck Freight Rail Freight WaterFreighl

I acturlng Steel Cement PrimaryAluminum P~perand Paperboard Nitrogenous Fer1li1zer Agriculture Mining,Construction M

Activity Level 4 persons/HH Typicalcooking levelw/LPGstoves. 50 liters 0,1hotwater/capita/day" One315 liter relrlgerator-!ree~er/HH NewJersey(US)level01lighting' 1 color IV/HH 4 hours/day 1/HH,1 cycle/day 5.4m2of floor space/capita (WE/JANZ '75) av, 0.19autos/capita, 15,000 km/auto/year (WE/JANZ '75) av, 1850passenge~ (p)-km/capita (WE/JANZ '75) av, 3175p-km/caplta(WE/JANZ '75)' av, 520p-km/capita(WE/JANZav,.75)" 345p-km/capita(WE/JANZ '75) av, 1495ton (t)-km/capila (WE/JANZ .75) av, 814t-km/capita(WE/JANZ '75) av, 1/2OECDEuropeav,'78" 320kg/capita(OECDEuropeav, .78) 479 kg/capita(OECD Europe '80) av, 9.7 kg/capi~a (OECD Europeav,'~O) 106kg/caplt.a (OECDEuropeav, ~9) 26 kg N/caplta(OECD Europeav, 79/80) WE/JANZ '75 av, WE/JANZ '75 av,






A more

fu ndamental problem

.' wIth both

Here WE/JANZ stands for Western Europe, Japan, Australia, New Zealand, and South Africa. The WE/JANZ 1975 average values for activity levels and energy intensities given in this table are

this analysis and the POLl analysis now being used is that the estimate of energy re uiremen .b d ...Equivalent <;I ts IS. ase .o~ ~xlstmg e~ergyusIng technologies. This ISInappropnate m l~oking toward the future, because today's high energy prices have rendered many of h. t ese, techn.°logles obsolete and because there ISa wIde range of new. more energyefficient end-use technologies that would be more cost-effective and more appropri192

from Relerence 20. " Activitylevelsfor residences estimates, are owing to poor datalor the WE/JANZ region. in termsof healdelivered the cookingvessels usingone 13kg canister LPG/ to to 01 monthlor a familyof 5, corresponding per-capita to fuel consumption of 49Watts. rate .For waterheatedfrom 20 to sooC ; Seetext, ... In 1975the diesel/electric was!n the ratio 70/30. mIx " In 1975the diesel/electricmixwasIn the ratio60/40. " Theton-kmper-capita waterfreightin 1978in OECD of Europeis assumed be reduced hall to by because reducedoil use (58~'o WesternEuropeanimpor1tonnageand 29% of that of of of exportswereoil in 1977)and emphasis selfreliance. on

A~IBIO \Ol !':-;O ,-~

-though per capita .larl!e.


was 10 times as

Table 3. Technological opportunities for a developing country in a warm climate to use currently best available or advanced energy utilization technologies. Activity . ResIdential Hot Water Refrigeration Lights Clothes Washer Commercial Transportation Automobiles Intercity Bus Train Urban Mass Transit Air Travel Truck Freight Rail Freigbt Water Freight Technology, Performance 70% efficient gas stove' heat pump WH, COP = 2.:;" Electrolux Refrigerator/Freezer, 475 kWh/year' Compact Fluorescent Bulbs. 75 Watt unit 0.2 kWh/cycle" Performance of Harnosand Building (all uses, ex. space heating)' Cummins/~ASA Lewis ,Car at 3,0 1/100 km. 3/4 energy !ntens!ty !n ,75 3/4 energy IntenSIty In 75" 3/4 energy intensity in '75" 1/2 US energy intensity in 'SO; 0.67 MJ/ton (t)-km' Electric rail at 0,18 MJ/t-kmk 60% of OECD energy intensity'


The importance of modern energy carriers can be illustrated via an end-use analysis of cooking, which accounts for most non-commercial energy use in de...Cooking velopmg countries. The per-captta energy use rate for fuelwood stoves, some 0.25 to 0.6 kW (0.4 to 1.0 tons of wood per year), . f . f h 0 05 kW -.TV tS ar m e~cess.o t e. .per-~aplt.a rate that IS typical when cooking with 11quid propane gas or natural gas (Figure 2). The much lower energy use rates for cooking with modern energy carriers reflect both the better efficiency (40 to 50 percent ..Passenger versus 12 to 18 percent for traditional fuelwood stoves) and the better controllability of stoves fueled with modern energy

carrIers. tove e .Iclency IS e ne .ere as the total heat delivered to the cooking vess I'd. .d d b h h . I f h Manufacturing e ..V1 e y t e eating va ue 0 t e Raw Steel AI/, Plasmasmelt & Elred Processes. cooking fuel. Cement Swedish ave in 1983" In addition to the energy savings associPrimary Aluminum Alcoa process". . ated with the shift to modern carriers conPaper and Paperb?ard Av of 1977 SwedIsh designs. ' .Nitrogenous Fertilizer Ammonia derived from methane" slderable .addltlonal savings can .be gained Agriculture 3/4 of WE/JANZ energy intensity" by adopting more energy-effiCIent techMining, Construction 3/4 of WE/JANZ energy intensity" nologies that have recently become available. Notes ies assumed here ' Compared to ~n assumed 50% efficiency for existing gas stoves. 70% efficient stoves having Some of the technolo (T bl 3) . 11 g low NO, emiSSIons,have been del/eloped by Thermoelectron Corporation for the Gas Research f h d . or t e omestlc sector a e I ustrate Institute in the United States (36). how large increases in amenities can be " The assumed heat pump performance is comparable to that of the most efficient heat pump achieved without approaching the present .water heaters available in the US in 1982(37). . I eve I s 0 f Western See text. energy co~sumptlon " Typical value for US washing machines. Europe. It IS assumed that each household' The Harn{)sand Building was the most energy-efficient commercial building in Sweden in 1981, has a refrigerator/freezer with energy perat the time it was built. It used. 0.13 GJ of electricity per square meter of floor area for all formance eq uivalent to that of the most purposes other than space heating (18). ..'" .f A 25-percent reduction in energy intensity is assumed relative to the 1975 average of 0.60 MJ/pefficIent 2-door Unit available In Europe In km for intercity buses, owing to the introduction of adiabatic diesels with turbo-compounding. 1982, a 315-liter unit requiring 475 kWh" A 25-percent reduction in energy intensity is assumed r.elativeto the 1975 average of 0.60 (0.20) per year which is less than one-third of the MJ/passenger (p)-km for diesel (electric) passenger trains, owing to the IntroductIon of adlabaI ,. ' .tic diesels with turbo-compounding (electric motor control technology). e .ectnClty requlre.d by .the average. r~" A 25-percent reduction in energy intensity is assumed relative to the 1975 average of 1.13(0.41) fngerator/freezer In use m the U.S. SlmtMJ/p-km for diesel buses (electric mass transit), owing to the introduction of adiabatic diesels larly. compact fluorescent bulbs, which ; with turbo-compoundin~ (electric motor controltechnolog¥). can be screwed into ordina ry incandescent A SO-percentredu~tlon In energy Intensity IS assu~ed relative to the 1980 US average value of 3.8 MJ/p-km for air passenger travel, owing to various Improvements (38). sockets and which have a light quality Slmli The assumed energy intensity is 1/3 less than the simple average today in Sweden for single unit lar to that for incandescents, draw only trucks (1.26 MJ per ton-km) .and 70mbinatio'! trucks (0.76 MJ perton-km), to take into account one-fourth a as th Improvements via use of adiabatIC diesels in Sweden, with an average load of 300 tons and an o m much .electricity f fi Our I t ' 75 k The average energy intensity for electric railwIth turbo-compoundIng. scenari .ssu es e equlva en 0 ve -average load factor of about 40%. watt Incandescent bulbs burned four hours I A 40-percent reduction in fuel intensity is assumed, reflecting innovations such as the adiabatic per day. diesel and turbo-compounding. While most of the technologies indi,. Assumin~ an energy intensity of 3.56 GJ of fuel and 0.40 GJ of electricity per ton, the averagefor Sweden Inan energy intensity of 84 GJ per ton of fuel (the US average in 1978) and 36 GJ of " Assuming 1983. electricity-the requirements for the Alcoa process now being developed (39). .Assuming an energy intensity of 7.3 GJ of fuel and 3.2 GJ of electricity per ton, the average for 1977 Swedish designs (17). " Assuming an energy intensity of 44 GJ of tuel per ton of nitrogen in ammonia, the value with .steam ~eforming of natural gas in.a new fert!lizer plant (40).. . AssumIng a.25-percent reduction In energy IntensIty, owing to InnovatIons such as the use of advanced dIesel engInes.





d fi d h

.:. '..: ,.i :.


:i7. . ,.

The LPG stove (below left) is in an urban slum in Sao Paulo. The fiv&-pot stove (below right) serves a slngl&-family household in Xinbu, a rural village in Guangdong Province In southern China, The three units on the left are for wood firing. The two on the right have been modified for firing with gas.

.-:::..:.?:. ~

AMBIO l"~~





:-,.-' ~-.~-~.--

cated in Tahle ~ are commerciallv availahle todav. a few are still in an advanced stage of development. The average automotive fuel economy specified for this scenario is 3.0 liters per 100km (79 mpg). This fuel economy could be achieved by installing an advanced adiabatic diesel engine with turbocompounding in an average-sized new U.S. automobile (1,300-1,400 kg), along with other energysaving features based on existing technology (reduced aerodynamic drag, low rolling resistance tires, and continuously variable transmission). Researchers at the Cummins Engine Company have advanced a design for a 1,360 kg, f .II> our to five passenger car with these features (15)-a car that could be commercially available within a decade. Alternatively, a lighter weight (775 kg) four to five passenger car with this fuel economy could be built using only present-day technologies-reduced aerodynamic drag, low., rolling resistance tires direct injection diesel eng' d i. 1 .bl n con Inuous y vana e t ..Ine ('I~ ) raTh nsmlsslon , e energy performance assumed for steel-making in our scenario is also characteristic of advanced technologies. Specifically, it is assumed that the energy perfonnance for steel-making is the average for the Elred and Plasmasmelt processes under development in Sweden (17, 18). Despite the potential for energy savings of 50 percent or more with these advanced technologies, the major industrial interest

in either h R case.s th .1. ot. n.

PERCAPITA ENERGY RATES COOKING USE FOR 12 -~ I 00 ~ 0 ~ a: ~ 0 g ~ u. a o. z ~ 0

--~ ~ -.=.

l;j ~ &





~ -~

l;j ~ a ffi

~ ~ ffi ~-

~ i


~ -~






0 . Figure2. Per-caplta energyuse ratesfor cookIng.For both wood stovesand stovesinvolvinghigh-quality energycarriers,the per-capitaenergyuse ratefor

cooking is expressed in watts. The wood-consumption rate is also given in tons

of dry wood per year,Assuming1 ton = 18 GJ,1 ton per year = 570watts(31).


-.n sTable 4. Final energy use scenario for a developing country Ih l vfl g an space heating) comparable to those in the WE/JANZ' region

in a warm climate, with amenities (except for in the 1970s, but with currently best available

suc. .a er, It IS In t e potentia or re duCIng overall costs and environmental pr?blems associated with steelmaking: by beIng able to use powdered ores (concenh h . I -rates Irect y, Wit out avlng to agg omerate the ore into sinter or pellets; by being able to use ordinary steam coal instead of the much more costl y coke. and b y in..' t~gratlng what are now separate operatlons. The other technologies highlighted in this scenario are described elsewhere (18). While not all these technologies are yet commercially available, no "far-out" technologies requiring major technological breakthrou ghs are involved We believe ...Passenger that the entire set of technologIes or a set of technologies with similar energy pert ) d . 1 .

or advanced energyutilization technologies. AverageRateof EnergyUse(Wattsper Capita) EIectr City I . ue 0a

F I T t I

Actl Vlty

Residential Cooking HotWater, RefrIgeratIon Lights TV ClothesWasher Subtotal Commercial Transportation Auto~obiles Intercity Bus, TraIn UrbanMassTransit Air Travel.

Truck FreIght FreIght

29 14 4 3 2 51 22


34 -22 107 26 32 8 21



5 2


fonnance .RaIl



be commercialized

In a matter of a decade or so. But is this scenario relevant to the situation in developing countries? Does this


WaterFreighl(incl. bunkers) Subtotal Manufacturing RawSteel


50 276 77


12 28 11 11 -36 65 121

4 -59


scenario provide meaningful insights regarding the future course of develo pment for developing countnes? Is the strong emphasis on energy efficiency a desirable goal for developing countries? And is it desir..Other" a ble or even realistiC to suggest that developing countries seriously consider the development of technolo g ies not Y et com...Mining. merclally avatlable anywhere In the world? These are questions to which we now turn our attention.



Primary Aluminum PaperandPaper~ard NItrogenous FertIlizer S~btotal.

~rl,culture, Construction


26 24 212 429


. 550

45. 59








.Here WE/JANZ stands tor Western Europe. Japan. Australia. New Zealand. and South Africa. For the activity levels indicated in Table 2 and the energy intensities given in Table 3.

.This is the residual.


<?ur an~l~sls here IS not Intended to estabIIsh activity level targets for developing countries, to be achieved at some future date. Indeed, the appropriate mix and


valueassumed hereis 45%less.sincethe average capitaGDPwas45%lessforW. Europe per than tor Swedenin 1975.Also. 22% of final manufacturingenergy use is assumed be to electriCIty. Swedishvaluefor 1975 the

has been estimated that at Sweden's 1975 level of GDP. final energy demand in manufacturing have been 1.0 kW (half the actual value), had advanced technology been used (17). The

~IBIOVOL I' ~o -1-5

.able 2030

5. in

Per-capita the IIASA

activity scenarios.






















1975 for

Activily Developing

Levels' Countries

Actlvlly Scenario

Levels of

for Tables

the 2-4

Actlvily Low

Levels Scenario


the High

IIASA' Scenario

Domestic Hot Water Service Sector Development Auto Use

Liter per

of Hot Water Capita per Day Commercial Space per





m' of Floor






Number of per Capita Passenger-km per Year Ton-km Ton-km Kg Kg Kg per per per of per per Capita Capita

Automobiles 0.0107 per Capita Capita per per N Year 3' 26 N.A. (d) N.A. (d) Capita 14 1)er per Year Year Year Year 545 189 21' 77' 345 1495 814 320 479 N.A. 82 1378 625 64 (d) 217 2993 1063 125 N.A. (d) 0.19 0.047 0.081

Air Truck Rail Steel

Travel Freight Freight Production Production

Cement Nitrogenous t!:enilizer Production

Contained per


Notes .These the " The

are parameters

population-weighted shown presented are from in the

average the IIASA IIASA

values study analysis

for (20). are


developing End-use convened

countries analysis from was kcal

except not per carried capita


centrally out for of the

planned centrally water

Asian planned per

economies Asian

Unless economies the

indicated in the water is

otherwise, IIASA heated study. by30

parameters Celsius. Reference






, d

degrees From Not

41. Steel is the only basic material for which demand levels are explicitly indicated in the IIASA scenarios.


levels veloping different goals. Rather sho:"" basIc provements beyond without energy ty as


activities countries to be

for may consistent

the well

future have with

in to overall

debe tor cent 28,

The in of four

prominent our total and and 26, Brazil, six, capita (Figure is consistent building. deserves and as




industrial for 58

secperto Tanalternaener-

certainly largely ing period The development growth of basic in

true completed of its

that the own

Western infrastructure development.


has build-

scenario-accounting final 37 energy percent respectively, 1.75 in times these as same use, for

compared India, or, much

infrastructure-building is the materials which construction roads, this period provide of characterized and such the factories, railroads, the production as

period by consumption steel building and blocks commercial bridges and rapid


the that it

purpose is possible but


our not

analysis only to provide that go to

is meet


zania, tively gy today scenario frastructure question per



needs in living satisfaction

also standards of


countries that our inthis as it is


3)-indicates with considerable Nevertheless, close attention,

ment, for

the significant use. such on value

basic in per

needs, capita

increases energy not be a supply

buildings, During

etc. use


Thus need



straint The experimenr' less sive if

development. of in our scenario making this out that there are are likely as point a "thought would






energy-intento be the imporcoming by ac1200






tant for decades

developing and that

countries in are characterized

tivity levels this We scenario. believe


excess that in

of those assumed most instances

for activi1000









,,:,ould In sectors our the



higher In levels and implicit energy example. countries

than all are

those energyfar in ~ 800

~ ~ ~



IndIcated using

scenario. activity

excess of average values for developing

countries stances known ?ext ments 40 in today excess (Table of of values future For 5) in most in inwellforecasts to for 50 years. developing requirethe over activthe


~ () ffi U") a.











Ity 1981

levels energy


developing projections Institute (IIASA) at are generally with per IIASA our capita scenarios higher for

countries for 2030 Applied less

in by

the the SysAusthan

~ ~ 400

International terns tria those 5), leve1s able our values to ev<:n In Analysis (19, 20) associated though the or

Laxenburg. far scenario final are the country 1.7.kW


0 INDUSTRY RESIDENTIAL TRANSPORTATION Final 1982 in energy (34), at the right, the 1 kW on give use and per-capita kW the 1 tops which are give the total (1), scenarIo by sector sectoral final the are presentedener~y In carrier, and Table shares energy (in use at the shares share percent) per tops capita are of 4. for India For the total by the final


99 <9O) ~ ~ . . -

(Table energy com value use parfor


;6 6S)



scenario-developing for 2030




average for the





IIASA low and hIg h scenanos respec. 1 ,

tlve y. mediate critical times. question A during" ment h. g~ t t an th ere wh~t I wl?uld IS of I be now modern P need required Europe. is the whether intensive h ase for 0 more to f at d eve inter'"inI openersupport (33), Figure Brazil 3. left, the three 1978sets(32), in of energy for final India energy use. (I), numbers on (B), the the


columns Tanzania For Tanzania use per the on 1981 in the four (T), capita, on ~.: : .

columns Brazil while these

carrier total The of total

and numbers

scenario the columns

numbers the carrier energy

(in percent). (in percent)

numbers final

in parentheses energy use.

e economies




AMBIO 1"~'



co' 600 6J

~ ,1'


...: ~te, Thonumb't. ., the r"" ..f O,1th h.,r


500 5a



0 20


repr" the """trIIM'tl..". '0 m..,,f."""i". fi".1 pe",.nt) , en.,s:y d"',1nd (In


0 0 ~ :J


.../~ .c '"'300 Co



A""O~;;A -, ---' "

/ '

'" """




-:-\ .".-:-"., ". C) "0 ..


/ :

/ '..7'




~ ~




'6 100 a:



~ 200 lsn CI.y, (:1...


",.".' ,-'

' C;c.ORtNE


.,.,., , ,

, :' 'r :::'f ...' /




~ -





0 0

,J 100.../


1970 1975




Q. ffi


~-"" ~-::;:.:.::'


5 Food











~ S ~ ~



Figure 5. Basic materials use in Western Europe. The data are three-year running averages of aggregates.of apparent consumption for France, Germany. and the U.K. The ammonia data are for France and Germany only (35). The arrows point to the scale to be used in quantifying each curve,

:n '.j) ,n '.n CONTRIBUTIONTO MANUFACTURINGVALUE ADDED (PERCENT) Figure4. The final energy intensify of manufacturing vs. manufacturing value-added for the United States in 1978. Here the value-added measure used Is Gross Product Originating (GPO) by Industry. The sum of GPO values for all activities in the economy equals the Gross National Product (21).

of these materials will tend to grow much faster than the economy as a whole (21). Because the basic materials processing industries are so energy-intensiveaccounting for most industrial energy use even in highly industrialized countries like the U.S. (Figure 4)-industrial energy use -would typically grow much more rapidly than GNP during this period. However, the absolute levels of basic materials production and use are not likely to be higher during the infrastructurebuilding period than in the beginnings of the post-industrial phase (e.g., the mid1970s for Western Europe), because although basic materials playa diminishing relative role in economic activity as the economy matures, these materials continue to play an increasing absolute role for a long time thereafter, as wider uses are found for these materials (21). The history of the per-capita use of seven important basic materials (both traditional and modern) for some Western European countries is shown in Figure 5. The values chosen for the per capita consumption rates of basic materials in our scenario are from the period near the peaks of these curves. This analysis thus suggests that it is feasible to achieve a standard of living in developing countries anywhere along a continuum from the present one up to a level of amenities typical of Western Europe today, without departing significantly from the average energy use per capita for developing countries today, depending on the level of energy efficiency that is emphasized. THREE QUESTIONS RAISED To date most new energy-efficient end-use technologies have become available main196

Iy in industrialized countries with market economies. Their relevance to developing countries is an issue of considerable controversy. Three questions have been raised concerning such technologies: First, can emphasis on energy conservation be justified in countries that are so poor that they have little energy to save? Second, even if there were significant energy conservation opportunities in poor countries, would not the pursuit of these opportunities imply technological dependency on industrialized countries, since much of the needed energy-efficient technology is not now manufactured in developing countries? Third, in light of the extra investment usually required to obtain improvements in energy efficiency, is not more energy-efficient end-use technology inappropriate for developing countries, where capital is so scarce? We now address these questions and show why we believe that investments in energy efficiency are not only relevant to developing countries but essential to bringing about rapid development. What Is There To Save? While it is true that the rich industrialized countries can save far more energy than developing countries, it does not follow that there is little energy to save in developing countries. The elites, who account for most commercial energy use in developing countries, have energy-wasting habits very similar to those of citizens of industrialized countries-and often there is even greater waste in the developing country. Consider two-door refrigerators that are now becoming popular among the elites in Brazil. The new Brazilian units of this type are smaller (340 to 420 liters) than typical units in the US (about 500 liters) and con-

sume between 1,310 and 1,660 kWh per year (22). By comparison, the average consumption of new units in 1983 in the US was 1,150 kWh per year, and the most efficient model available in the US, introduced commercially in March 1985, was a 490-liter unit requiring 750 kWh per year (23). Even poor households dependent largeIy on non-commercial energy and having few if any modern amenities tend to use energy inefficiently. As we have already pointed out, the poor who depend on wood for cooking consume three to 10 times as much energy as those who have access to modern energy carriers (Figure 2). For the urban poor who buy wood or charcoal for their cooking, this inefficiency implies a large expense. In Bangalore, India, for example, the poorest 15 percent of all households spend 17 percent of income on fuelwood (24). For the rural poor, fuel inefficiency implies a great deal of time committed to wood gathering (especially by women and children) that could other. wise be occupied more productively and. that is increasing in many parts of the world, as deforestation makes wood supplies scarcer. . Where WIll The Technology Come From? To the extent that energy-efficient end-use devices are not now available in deve10ping countries, these devices must either be imported, or a local manufacturing capability must be established. Those developing countries with little industrial infrastructure may have to rely on imported technologies, at least for a while. But these countries would have to import the conventional. less-efficient technologies anyway. For these countries the issue is whether the increased foreign exchange expenditures often required for

A~IBIO VOL 1.1 ~o .1-5


.the more efficient end-use technoll)gies , can be justified, In this instance it is important to calculate the fo~e,ignexchange impacts of the more effIcient technologies 'fro,m the perspective of the entire system of Improved end-us~ technology plus energy s~pply. because In ~any cases the extr,a f?relgn exchange !equ~red for a more efflclent end-use devlce,wllI he more than ?ffset by redul:ed foreign exc~ange requlrements f,or new energy supplies. The Imp?rtance of a systems approach ~o the forel~n exc,hange problem may b.e Illustrated ~,Ith a,IIgh~ In an analysIs of the BrazIlian sltuatl?~ It has been shown .~hat for each d.ollar a citIzen spend~ on ':1~w Inc~n~escent IIg~t bulbs, the e.lectnc ~tlilty ~ust Invest $10 In hydroelectrIc supplies to light t.he bulb (25). For a country that .must, Instead, rely on the more costly th.ermal. power sources, the correspondIng energy supply costs would level, newenerAt the macro-economic be larger. gy supplies for developing countries required $25 billion of foreign exchange in 1982-over one-third of the foreign exchange required for all kinds of investments (1). Clearly, to the extent that net foreign exchange requirements can be reduced by importing more efficient end-use technologies, as alternatives to conventional end-use technologies, a developing country would be better off. In more advanced developing countries, the manufacturing capability for many efficient end-use devices could be readily developed, if manufacturers believed there were sufficient markets. In India this has already happened in the case of automobiles. High efficiency oilusing technoiogy is of crucial importance to India, which spent more than 80 percent of its foreign-exchange earnings in 1981 on oil imports (2). This high cost has motivated a shift to more fuel-efficient cars. While typical five-passenger cars on the road in India have a fuel economy of 10 to

11 liters

,'eloping count~' should he distinguished from the prohlem of the o"erall limited supply of capi,tal, , Under a wide range of cIrcumstances. the extra capital requir~men~s for improved end-use te~hnolog,les will be more than offset by capital sa,:,lngsfor lowered energy nee~s. An anaIY~I~by Geller of t~e ~nergy-sa':lng opport~mtles for ~he BrazllIan, electrIcal s.ector IS suggestive ,of the savings pot~ntlal. Geller h,as estimated that for a (dl,sco~nted) totall~vestment of some $4 ~llIton In more efficlen~ end-use te,chnologles (for, ~ore. ef~cle~t refnge~ators: st.reet lightIng. lightIng In commerclal bulldlng~, and motors, and t~e deployment of speed motor dnv~s), It would be f~aslble to defer construction of some 21. gIgawatts electric .(GW(e)] of new .electrlcal. supply cap.aclty, .correspondlng to. a discounted caplt~1 ~avl~gsfor new supplies of some(22), billion m the period 1986 to 2<XX> $19 The resulting capital savings could be used to speed up the development process in other areas. ." Not only are investments in energy efflciency relevant for developing countries, but also they are often even more relevant than for industrialized countries. This perhaps counter-intuitive result is illustrated in Figure 6, which compares, for the Brazilian situation, the cost of saving one kW with investments in efficient compact fluorescent light bulbs with the cost of the extra hydro-electric supply expansion that would be required if incandescent bulbs were used instead (25). The cost shown in each case, as a function of the discount rate, is the discounted present value of all required investments over a 50-year life cycle. This figure shows that at a 10-percent discount rate the supply expansion cost per kW is about three times the cost of saving one kW by investing in more efficient bulbs. What is perhaps more interesting is the fact that the higher the discount rate the greater the benefit of the energyefficient light bulbs. There are two reasons

THE RELEVANCE OF ADVANCED TECHNOLOGIES There are good reasons to believe that in some instances it would be desirable for developing c~untries to commercialize advanced energy-saving technologies such as those highlighted in our l-kW scenario, The energy savings potential of advanced technologies is often far greater than what can be achieved through the modification of existing technologies. Those technological innovations that society adopts must be attractive to compensate for the dislocations that often accompany the introduction of new technology. Historically, this has been such a powerful phenomenon in the industrial sector that process innovations generally have led to marked improvements in energy efficiency even during periods of constant or declining energy prices, when energy has not been a major concern . 4000 ~:; ;, 3

~ .;: ~ 2 .".



Figure 6. The dIscounted cost 01 peaking electricity


present values 01 the produced via a hydro-

new, domestIcally manufactur.ed fuel economIes of the order of SIX liters per 100 km (40 mpg). In the case of Brazil there is strong evidence that Brazilian manufacturers would be able to manufacture a wide range of energy-efficient, electricity-using technologies (refrigerators, efficient lighting


per 100 km (21 to 24 mpg) .'.


for ~hlSresult:. .electric FIrst, efficient .lIght bulbs, most other energy-effiCIent technologIes, tend to have lifetimes much shorter than energy supply facilities-typically in the range of two to 20 years, versus 20 to 50 years for most energy supply facilities. Investments in conservation equivalent to new supply expansIon are spread out over tIme. Therefore, the present value of the needed investments over the life cycle of the energysupply facility they would displace tends to decline rapidly with an increasing discount rate (25). In sharp contrast, the investment required for energy supply expansion tends to increase

powersystemand the cost01saving electricityvia the Installation efficientcompact 01 fluorescent IIghtbulbs. Thecostsarefor electricitydelivered or saved to at tha household. All capitalInvestments overthe estimated soyearIIle of the hydroelectric pow~rplentareInCIUded

h arB .Th


. .:: -

e ers iss t l u med a co


newpen hYdr


tect s

thrlc a t powe wou

ldr C °s be

ts b u lit

technologIes, heat pumps, motors, motor control devices) in just a few years, if there were sufficient markets for such devices 0 (22). Moreover, it is ironic that the most efficient refrigerator/freezer available in the U.S. achieves its efficiency in large pan by the use of a compressor imported


InBrazil.Specifically It Is assumed the "overthat night" construction~ost the hydroelectric 01 laclllty Is $1,170 kW01Installed per capacity,thattha plantIs constructed overa 6-yeerperlod,andthat the plantIs paidlor with six equalpayments over this construction .P8r1od. provide1 kW01peakTo Ing demand r~ulres 1.16kW01 Installedhydroelectric capacity, ~o allow lor a 16-percent reserve margin. The total Installed capacity required to

t:urer mvo ve exports a hIgh efflctency line an~ markets.a less efficient product domestIcally, partially because of the wide range of design operating voltages and the ~onsiderable voltage fluctuations at BrazilIan houses. Ti)e less efficient compressor is more tolerant of voltage variations. e 0 emo apltal arclty ..resu It IS certaInly true that consumers must mor.e for the purchase of energy-efflclent de:-'lces. But the difficulty the consumer has m obtaining capital in a de"~IBI() I'/!\~

Th Pr bl r C . Sc .We

Brazil (26). . I d




rate, bec~use of Interest c arg~s accu~ulated dunng th~ long constructIon perIod. The net result IS that the benefits of conservation investments tend to increase rapidly with the discount rate. This means that from a societal perspective, investments in energy efficiency improvement look better the higher the discount rate.




with h

the discount

provide1 kW01 residentialbaseloaddemandIs 1,45kW,when allowanceIs madelor 20 percent transmissionand distributionlosses. It is assumed transmission that lacilitieslasting 30yearscost $710per kW,Some this transmls01 slon costcouldbeavoidedII newgeneretlng capaclty.could delerredvia Investments be In moreefficientlight bulbs.

IS assumed that






a rather


..It surprising

' "





t, Investments In energ) e Iclenc~ Improvement will often make even more sense in capital-short developing countries with many pressing needs than in industrialized countries.

li ght bulbslastin 6 000hoursandcosting $9 20 g eachreplace 4O-watt Incandescent bulbslasting 1,000 hoursandcosting$0.50 each. It Is assumed that the lightbulbsareusedlive hoursperday, includingtheearlyeveninghours,so that they contributeto the peak electricitydemand (25).




Tahle f1illustrates this phenomenon for the ~.S. v;here energy prices were generally constant or falling in the decade~ immediately preceding the 1973 oil cri~is. In this era of high energy costs it can be expected that this phenomenon will persist-that the pursuit of socially attractive innovations will often lead to significant energy efficiency improvements, even if the new technologies are not selected specifically for their energy-saving features. Indeed there is a wide range of promising advanced energy end-use technologies that could be brought to commercialization over the next decade or so and offer major advantages of capital savings, environmental protection, materials savings, etc., in addition to energy savings (18). Despite the promise of major energy savings associated with the adoption of advanced technologies, it is generally thought that the process of commercializing advanced technologies is an inappropriate activity for developing countries, which do not have the institutional infrastructure needed for technological innovation and which cannot afford the risk-taking of innovation when there are so many pressing development needs to attend to. Yet, as we shall now argue, there are many reasons to question this conventional wisdom. Rationale For Seeking New Technologies While less risk would be involved if the technologies adopted by developing countries were those that have already been commercialized in industrialized countries, several considerations weigh against this alternative. First, many of the industrial tech~ nologies now being commercialized in the North are capital-intensive and laborsaving-characteristics that are not wellsuited to industrial activity in most of the South, where labor is cheap and abundant and capital is costly and scarce. Second, the comparative advantages in natural resources are often quite different for many countries of the South from those of the North. In energy, many developing countries are blessed with largely unde-

veloped and relatively low-cost hydroelectric resources. v;hile most industrialized countries must turn to more co~tly thermal po\\.er sources for increased electrical capacity. Similarly. biomass is a promising. source of chemical fuels for many developing countries. requiring decentralized development strategies quite unlike the centralized strategies that have been pursued by the fossil-fuel-rich countries. Third. human needs are quite different in the South from those of the North. because of climatological differences, because of different cultural aspirations, and especially because the satisfaction of basic human needs and infrastructure-building must be given prominent focus in the economic planning of the South, dictating patterns of production markedly different from those of the North. Developing and industrialized countries are at different phases in their industrial development. In the industrialized countries, areas of greatest growth and innovation are electronics, information tech nology. communications, medical technology, etc.-generally areas involving high valueadded fabrication and finishing activities. Both a shift in consumer preferences away from materials-intensive goods and the oil price shocks of the 1970s have curbed growth in the demand for basic materials in industrialized countries (21), as shown for Western Europe in Figure 5. This stagnation in demand provides a poor climate for innovation in the basic materials industries, even though existing capital stocks have been made largely obsolete by the energy price increases of the last decade. Thus in industries of crucial importance for infrastructure-building, the North is not innovating at a pace sufficient to provide for the needs of the South. Finally, the potential for rapid growth in the demand for basic materials in the South suggests that some countries of the South may be more promising theaters for innovation in these areas than the countries of the North, where demand is stagnating. For all of the above reasons, developing countries should not retrace the develop-

ment path of the North hut should pursue new directions and assume the risks of innovating in some area~ of pt)tentially high pay-off. . ExIstence ~roof of Technological LeapfroggIng While there are many theoretical arguments favoring "technological leapfrogging," whereby new technologies are introduced first in developing countries. the idea is still contrary to the conventional wisdom that new technology must be introduced in the industrialized countries. because only these countries have the infrastructure and the risk-absorbing capacity that is needed for introducing new t,echnology. But technological leapfrogging is not just a theoretical construction. It has been tried on various occasions in the past-sometimes with success,and other times not. There are lessons in the historical record, some of which we now review. Ethanol-Success in Brazil and Failure in Kenya: The Brazilian alcohol program in 1984produced about 10 billion liters of ethanol from sugar cane and replaced about 60 percent of the gasoline that would have been required in the absence of the program. Brazil has also developed cars that run on pure alcohol, nearly 600,000 of which were sold in 1983. The cost of ethanol has been estimated to be $50-56 per barrel (1983 U.S. dollars) of gasoline replaced (27), when the subsidies are remove~ and a true exchange rate is used (that IS, a rate based on the parity value of exported goods). This cost is co.mpetitiv~ with gasoline produced in BrazIl from Imported crude at the 1981 world oil price, although it is six to 17 percent higher than t~e c<;>st &asoline of based on the 1984 oil pnce dunng. ~he world-wide oil glut.. But t~e. Brazilian ethanol program provides additional benefits of rural development,. employment generation, incr.e~sed self-rellan~e, a!1dreduced vulnerability to futu~e roses In the worl? oi~ Also, ~hlle. alco~ol productlon IS cam~d out pnmanly ~Ith loc~l currency, gasoline has to b~ paid for In hard currency. Because Brazil has a nega-

Table6. Historicalenergyrequirementsper unit of output for selected malerialsproduced in the U.S.(42). SodaAsh (SolvayProcess) Date

1868 1894 1911 1925

Ammonia (Haber-Bosch Process) Date

1917 1923-50 1965 1972

Chlorine (Diaphragm Celis) Date

1916 1947-73 1980

HydraulicCement".Raw (Wel and Dry Processes) Dale

1947 1955 1960 1965

Steel". (Changing ProcessMix) Date

1947 1954 1962 1971

Energy (GJ/ton)

60 31 28 17

Energy (GJ/ton)

93.0 81.0 52.0 46.5

Electricity (kWh/ton)

4400 3300 2400

Energy' (GJ/ton)

10.3 9.0 8.5 8.2

Energy" (GJ/ton)

37.5 32.4 30.0 27.8 . ,

1942 1970

15 14


41.2 Polyethylene Date

1956 1973 1975

1971 1978

7.5 6.8




Ethylene Dichloride Date

1967 1973

EthyleneOxide Date

1970 1973 1974


100 15


100 85 79


100 40 18


.Data for ali but the last entry are from reference 43. .The 1978data for cement counledat 3.6 MJ 44. , Electricityrequirements are from ReferenceperkWh. are .The 1980data for steel are from Reference 45.



A\IBIO VOL 1-1 NO -1-5



.'Iive halance of pa~ments that is expected to persist for a numher of ycars. this means that oil imports must he financed \\.ith forc:ign debt. The Brazilian ethanol program is especially efficient in generating jobs-requiring an investment of only $6.()()() to $28.000 per job. \\.hich compares with an average of $~2,OOO the Brazilian indusfor trial sector. and $200.000 for the oil-refining, petrochemical complex at Camarcari. It is estimated that a total of 475,000 direct full-time jobs (700.000 at the peak of the harvesting season) in agricultural and industrial activities will be generated by .1985. along with another 100,000 indirect jobs in commerce. services. and government (27). In short the Brazilian program has been , successful in meeting its technical objectives, reducing oil imports, and supporting the development process. Successin this instance is more a reflection on the Brazilian technological development process than on ethanol as an energy source. An abundance of land resources. a climate very favorable to sugar cane, financial incentives to alcohol producers, a high tax on gasoline, and a longterm history of experience with ethanol on a modest scale were ingredients that helped foster technological success. The Kenyan experience with ethanol has not been as positive (28). Because of potential serious conflicts with food production, Kenya cannot readily dedicate significant amounts of agricultural land to sugarcane as an ethanol feed stock. Instead, Kenya has utilized the residual molasses of the sugar inudstry. This choice is a dubious one, however, not only because the economics of ethanol production based on gtolasses are much less favorable than for sugarcane, but also because molasses is a strong foreign-exchange earner and there is strong domestic demand for molasses in cattle and pig production. In short, the Kenyan program is a case of transfer of inappropriate technology, with the situation made worse by poor planning. Charcoal Based Steel-Making in Brazil: In the industrialized countries coke began !o replac~ charcoal as a reducing agent for Iron-makIng at the middle of the 18th century, as a response to rising charcoal costs (29). The shift to coke led to much largerscale.iron-~aking blast furnaces than were possible wIth charcoal, as coke has much , ¥reater mechanical strength to resist crushmg under the load of the blast furnace chargcc:. While most of the world's steel industry is no\v hased on coke. 37 percent of Brazilian steel production (i.e.. ~.Ij million tons) was has~d on charcoal in 191;3. This anomaly reflects the scarcit~. of high quality coking coal in Brazil. Coke-based steel production in Brazil is based on mix of 80 percent imported coal and 20 percent high ash content domestically produced coking coal. Though charcoal-based steel production is widely viewed as anachronistic, it produces a better quality steel because charcoal has less impurities than coke. Brazilian charcoal-based steel is competitive in world markets because the industry is far advanced in relation to the ancient charcoal-based industry abandoned by the now industrialized countries some 200 years ago. That charcoal-based steel with blast furnaces processing hundreds of tons of steel per day can compete with coke-based furnaces processing thousands of tons per day is of great importance to developing countries generally. The technology is labor-intensive; it is well-matched to the resource base of the biomass-rich, fossil-fuel poor developing countries; and the scale of its installations often provides more appropriate increments in productive capacity in relation to the size of local markets than giant cokebased facilities. The major cost item in charcoal-based iron making is charcoal, which accounts for 65 percent of the total pig-iron production cost in Brazil. Charcoal for steel making is produced mainly from planted forests, primarily of eucalyptus and pine. The total plantation area exceeded five million hectares in 1983. Technical developments relating to charcoal-based steel production are advancing at a rapid rate (Table 7). Plantation yield has roughly doubled over the last decade, and a further 50 percent increase in yield is expected. Over the last decade charcoal yields from wood have improved about 20 percent, and still another 10 percent increase is expected in the near term. At the same time charcoal requirements for iron making have been declining. The net result of these improvements is that in the near future. the land area required for a given level of steel production is expected to fall to just onefifth of what was required in the 1970s (Table 7). ...siderable Formulating LeapfroggIng StrategIes The above examples show that it is feasible to have technological leapfrogging in developing countries, but new technologies must be chosen carefully to be compatible \vith the local resourcc hasc and to he supportive of the achievement of hroad development goals (e.g.. enhanced employment generation. increased self-reliance, rural development. etc.) as well as to offer opportunities for direct cost reduction. Despite its bj:nefits, leapfrogging should not be regarded as a universal strategy for industrialization. It is appropriate only where a unique set of circumstances and capabilities offers great enough benefits to justify the taking of risks. The challenge to planners in developing countries and to the international assistance community is to identify and facilitate the exploitation of promising opportunities for leapfrogging and to support the development of the infrastructure needed for innovation. Promoting innovation in this manner could greatly broaden the technological choices available to developing countries to meet their development goals. And while technological success in one developing country will not necessarily be relevant to all developing countries, on balance it is likely that innovations generated via leapfrogging will be more widely applicable in the South than innovations coming from the North, since the countries of the South are often more similar to each other in terms of their needs and capabilities than they are to countries of the North. CONCLUSION Per-capita energy use in developing coun-' tries near the present level would be adequate to support a standard of living ranging from the present all the way to that of Western Europe today, depending on the extent of the shift to modern energy carriers and the emphasis on energy-efficient end-use technology now available and on the commercialization of advanced energy-saving technologies. Modernization of the energy end-use system along the lines discussed in this article is technically and economically feasible and strategically desirable for poor countries. An emphasis on energy-efficiency improvement and modern energy carriers would make it possible to satisfy basic human needs, to radically expand the industrial infrastructure, and to allow for conimprovements in living standards, beyond the satisfaction of basic needs, with little change in the overall percapita level of energy use. This use should not obscure the challenge of bringing this about, however. Large amounts of capital


Table7. Parameters relatingto the annualproductionof one milliontons of steel basedon charcoalin Brazil.(46) WoodYield on Plantations (tons/ha/vr)" 12.5

25 37.5


1980s Near future


Wood-to-Charcoal ConversionRate (m'/ton) 0.67

0.80 0.87

SpecificCharcoal Consumption (m'/ton pig iron) 3.5

3.2 2.9

RequiredArea for Plantations (1,000ha) 336

128 71

Investment Required to EstablishForest (1,000US$) 201.600

76.800 42,600


.Air dry tons (25-percent moisture).

AMBIO. 1'1115



I~ 'Ir",I"r",ll.ln~ln."I,'llh"",'I",II""""""h,,r,,






t h ~1





to .:fficit:nt

11 I

.,n" ,"."'n


In 1.1"1":: .1' """'~ "h.,r,":I"rNI",~1


..' Fr"nk '.'n IlirPCI. I .\ (r.m'I',.r',I1I"n fno'r::,

IJo',nunol.Pl.,EESR" r .'rt'" 1111("nl.:rmr



r"!".""""'I"rnur..,,,,.lr,ln.lI'tr,II.1 ,.nJ "," Z,:,Ii.lndl







.cn"rg,;ondEmlr..nm""lal~ludl",.Pnn""t.." l:ni'"r,il...141'11



But our analvsis suggests strongly that I.' R R and R..~ Kam.. I(.ummln, I:ngin" for a widt: range-of plal:i~ible sets of-'lctiYity ,,'mr..n, <...,Ium"u', In,dianal and J , \Y..,..I 1.1 ' .'."'.I:';ASAL""i,R"",ar.:hC"nl"r,('I"""land.Oh"'I. e\e s and for a wide r .Inge of end-use .-Id,.",..,,01 .-Id",h"".. /)/..,..1 /:nll""" ftJr 1"1"..nll"r technologie~, it would be less costly to pro(""r,. P"p"r:';.. X-II)'I.1-1,.in"t~,ol Aut~'moli.." vide energv services using the more effiEng,n""r'., .-IoIlaha/u EJIIi/n.." II'Jrld.ll,d.. R... , d' hi' h 'd ,"..",SP-:-7I,(I'IX-I), clent en -use no, ogles t an ~o provi e In Frank ",m Ilipp.:1 and Barhara L"..i. R..,vurc..s the same servIces wIth conventional. less and Cons..r..atiOJI, 103-12-11'!II3) 10, ( efficient end-use technologies and in17 P SI.:.:n,T B Johansson.R Fr.:driks",n. andE. creased energy su pplies. B,'gr.:n. E.n..r~'.-for "'hat a/III Ho" .1/1.'ch:' (~i".:r , ..Forlag.SI'lCkhulm, I'IXI) (In Swcd.,hl Sum.While we have not dealt explIcitly here mariz.:din T. B Johan,son, p, Stc.:n. E Bogr.:n, with the energy supply implications of an .nd R Frcdriks,on. Scitnct, 219, 355 (191'3) energy.efficient future for developin2 IX J~)s':,Gold.:mh.:rg. Thomas B J.','hanssun. Amul~a (, 't' If'd h' h' ~ K~R.:ddy."ndR()hcrtHWlillams,En..r~."for ~oun nes. I, IS se -evi ent t at t e ~eemt' S,."ainablt World. "'x,k m.nuscripl Ingly formidable supply problems deSumm.rizcd in Annual R..,i.." of En..r~v, 10. scribed at the outset would be greatly /113-/1&'1 (19t15) .tCI eased in an ener gy -effic'ent world The 1'1 En.:rg~S~'slemsProgram Group. In!.:rnalll,nallngreater the degree planning,

, lower the level I I of aggregate

of flexibility


in supply

.stllul.: the

gcr. Camhrid~c. l'Jt!I),

for Appllcd Systcm' Analys,s. En..r~v In u Finir.. IVorld A Global S_..SltlnS Allalvsis. (Ballin-

.14 Th"" R B".:k, Imprl"""""'" /II f.n,'rll' 1-:/1/. ..",n..,' "f Inolu"rial 1-:1"..,r"..h,',n,..,,1 I'r;,;'..,"',. tR"p"rlpr"par"dfl'rlh"Olli.:"..IEI"ctr...:h"mi. car Proj"cl ~Ianalt"m"nl. Are,mn" :';ali..nal La""ralor~, A~L 6EP~I-77-2. J:'nu.r~, 1'177) ,/(1 D A W.ilzm.n.., ai" F..nili:..rfra," Coal.P.pcr pr"par"" Di..isionuICh"micaID"..""nl. T"nncss"cV"II.:.. Aulhoril.., ~Iuscl.:Shual" Alahama. and pr.:s.:ntcd"t th.: Facull\ In'titut" un C".I ProductionT.:chnultllt.. and L'tiiizalion. O.k Ridg.: A,s'lCiatcd Uni"'.:.rsili.:.. Oak Ridg", T.:nn.:""c,(Jul..21-Aueu'tll.I'I7XI -II l:nit.:d Nations. 11/79/811 S,a,i"ical ",arho"k. (Unitcd N"liuns, N.:wYork. 19XI). -12 Roh.:rt V. Ayrcs, Final R':p',rt un Futur" En.:rg~ Consumpllonhythclndu,triaICh"micalslndustrv (SIC 2M), Appcndix 10 Th.. Cht/llicals IndILSrr':. ..ul. 5 of 'I. in IndlLSlrial Entr,~.' Prodllctivit.vPro/" Final Rtporl: I?r.:parcd En.:rg~a~d En..ironby. mcnl,,1 AnalysIs. Inc.. Arlln~lon, \", for th"

Asst. s.:crclary for Conserv"llon and R.:n.:wahlc Encrgy, (U,S. D.:partm.:nt of Encr~.., DOE:;CS'


Fcbru"ry, 191!3).


With lower demand levels it be." comes possible to avoid ove~dependence on the more troublesome optIons. Indeed. we have shown elsewhere that via the ,U,rSUI , 0, energy e ICle~C} strateg.les In IndustrIalIzed and developIng CountrIes alike (18, 30) it is feasible to remove energy as a constraint to developmen t an d to evo I ve I ong term go I b a I -ener-

20 Arshad ~I Khan and AIt'IS Holzi. E,'oilitloJI 1).1 -13 Th.: Confcrcnc.: Board. Elltrg." Consulnp'ion in Fuulrt Entrg.v Dtmallds ull 1030 In DI.ff..rt'" ,\Iullufaclur/ng. a rep"rt to the Encrg~PolicyProjlI'orld R..gions All Asst'SSJn..'11 for Ih.. T"o ;lla,lt ':CI of Ihc Ford Foundation. (Ballingcr, CamII.-ISA Sc..narios, (R.:port No. RR.M2.1-I thc 01 hridgc. 197.\). Int.:rnationalln'tilut.: for Appli.:d Sy'tcms An"I~'-1-1J T. DiK"ou. 1978-79 .\Iilltrals Yearbook,Vol,is. L",.:nhurg. Austria, April. I'IK2} Th.: comp"ri",n, indical.:d in Tahlc 5 ar.: fur thu,.: acli..ity UJllt I: ,\Ittals ~Iin.:s. I'I!!II}. and j\1intrals, (U.S Burcau of




f f ..'

1.:\.:1, ,uffici"ntl~ di,aggr"gal"d in Ih" IrASA anal~,is10 pcrmit a .cump.rison with o.ur ,c.:nariu For Ihcothcr actlvlll':' Ilsl"d In T.hl~ ).an .:'pllclt

comp"nSl,n sibl.: wIth thc IIASA sc.:nant'S " nul P',s-

-15 En.:rgy "nIl En"ironm"ntal Anal~si" IndlLSlriul Producril'i~' PrOjecT Final Rtport-l, pr.:parcd for th.: ASSI,Sccrclar~'for C,m"'f""IIOn "nd R.:n.:w.hlc En"rgv. CSi-ilJt51-1. (US, Februarv. Dcpartment 191!3). of Energv. . DOEI

gy strategies that are economically and en21. E D. Larson. R. H William" and D. Bi"nkow,vironmentally sound and strategically seki. .l/altrial CollSump'ion Pau..rllSUJldInt/rL"rial g t t . th t h t .En..rg.'. D..mandill Induslriali:..d COIIJllri..s,PUt cure-ener , y.S ra egles, a, In s or , are CEES R.:pon No. 17-1.C.:ntcr for Encrgy and compatIble wIth the achievement of a susEn,,'ironmcnlal Sludic,. Princ.:ton Uni".:rsilv. tainable world. Princcton,~J.. 19t1-1 .


-16 Rodrigu.:s dc Almcid.. (Florcslal Acesita. S.A., Bclu Hurizont.:. Brazil). prc",nt.tion at thc Second EI,d-Us..Oriented Global EJltrg). Workshop, Sao Paulo, Br"zil. (Jun.:. 1984).


R f e erences and Notes I. Th.: World B"nk. Tht EJltrgy Tra/lSilion in Otvtloping CounTries. (World Band W.,hinglon DC. 191!3, 2. Thc World Bank. World Dtv..lop"'tnl Rtporl 1984.Oxford University Press. 19!14 3. In Tables2.6 and 2.7 of R.:fcrencc19. thc rcmaining ultimatcly rccovcrablc oil and n"tural gas rcsourccsin dcvcloping counlrics olh.:r than thosc in the Middlc Easl/Norlh African rcgion "rc cslimatcd (as of Ihc laIc 197Us) to be somc 133 TWY, -I. World Energy Confcrcncc. World Entrg,' Rt-

22. Ho"ard S Gellcr. Tht Potential for Eltctricil... Cons..nalionin Bra:il. Companhia En.:rgclica u., Sao P"ulo, SaoPaulo. Brazil. 19K5. 23, Th.: Amcrican Council for an En.:rgv-Efficicnl Economy, Tht /~/oSI Entrg Efficitnl Appliancts, W"shinglon DC, Summcr 19M5. 2-1 Amuly" K, N R.:ddy and B, Sudhakar R"ddy. Energy in a Slralificd Socicly: Ca'cstud~ of FIrewood.n Bangalorc. Economic and PolItIcal IVttkl.v. XVIII. October 8. (19M3). 25 Ju-.;Goldcmberg .nd R. H William,. Th.. Econo,nlCSof Entrgy ConservatIon In Dtvtlopmg Counlnts: Iht CoJlSumtrVersustilt SocIetalPtrsp..cti,t. PU/CEE.S Rcporl No, Itl9. Ccnlcr for Encrgy and Envlronm.:nlal Sludlcs. Pnncclon ., Unl..crsllY,Pnn~cto.n.N.J.. 1985,..

~6. Pn..alc communIcation to R, H. W,lliams from H, S G.:llcr. Apnl. 19t15, Jose Goldemberg

sources 1985-2000. (IPC Scicncc and Tcchnology 27 H".""rd S. G':!".:r. Annual Rtvit',' of Entr.I:)'. 10. Prcss. 197t1), Gcological coal rcS('urccs in L.tin 13)-16-1 (19t1) Amcrica and in Asia outsidc of China arc cSli- 2t1. Phil O~Kecfe "nIl Don Shakow. A,nbio. 10. mated10 be insignificant, Thosc in Africa arc c'li213-21) (19t11), malcd to be only 1.\5TWY. whilc coal rcscf"CSin 29. Charlcs K, Hydc. Ttchnologlcal Changeand the Africa arc cslimatcd to amount tu only 29TWY Brilish Iron Indl"tr),. 1700-1870.(Princcton Uni5. Thc Carbon Dioxidc As",ssmcnl Commill':':. ,.:rslty Prcss.Pnncclon. Ncw Jcrscy. 1'177). Changing Climalt. (Nalional Acadcmy of Sci- 30. Robert H Williams. Potential Roltsfor Bioentrgy cnces W"shinglon DC 19t13 ) In H W.II .' .'.. , P . I R I (Amblo. thIs Issue). ' ' E 3I... R an Entrgy-Efficltnt World. esJor B 10- ntrg,v In 6, Ellis L, Armstrong, Hydraulic Rcsourccs. In RtI lams, Oltnfla 0 ntwable Energy Resourcts:Iht Full Rtports to Iht an Energy.Efficitnl World. PU/CEES ~eport No, Conservation Commission of Iht World En..rgy 1:13,(Ccnlcr for Encrgy, and Envlronmenlal Conference, IPC Scicncc and Technology Pr.:,s. Studl':s. PrInceton UnlvcrsllY. Pnncclon. N,J. 1978, 19t15). 7. D, E, Earl. FOrtSI Entrgy and EconoJllicDtvtlop32. Amul~a K N, Rcddy. An End-Ust .\t..thodology mtnl. (Clarcndon Prcss. Oxford. 1975), for Dtvtlopmtnt-Oritnltd Entrgy Planning in Dt8. D, p, Ghai. A. R. Khan. E. L, H. L"c. and T, "tloping Countries. with India as a Cast Srud_\'. Alfthan. Th~ Busic Nt~ds Approach, 10 D~vtlopPIJCEES R.:port No .ltll. (Ccnter for Energy mtnl. (Intcrnallonal Labor OrganIzation. G.:n.:va. and EnvlronmcnlalStud,.:s. Prlncclon IJnl"ersIIY. 1977). Princcton,~J,. 19M5). 9, P. Slrcclcn. S, J, Burqi, M, U Haq. N. Hicks, and 33, ~I J. ~lwandosyaand M, L. P. Luhanga. Entrg.v F, StCW"rt,FirSI ThiJlgsFirSI: MttliJlg BasicNttds Us.. Patltrns in Tan.:ania.(Alnbio. t~is.jssu.c), In DtvtlopiJlg Counlrit.'. (Oxford UnivcrsilY 3-1,R.:publlca F.:deratlva du Brasli Mlnlslcno das Prcss. N"w York. 19t11) ~Ijnas .: Energia. BalaJI"o Entrgtlico .Valional. 10, M. Q, Quihria. World D.." 10. 2t15-291 Br"silia, Br"zil. (19M3) (19M2) 35. ~I. H Ross. E, D. Lar"'n. and R. H, William,. II. P F, Palm.:dotr al.. Entrgy N ds aJldR..sourc.." En..rgy Dtmand antl/Wal..rialsFlo,,'sin Ih.. EconoIn D..vtlopiJlg Countries. (Bnx, Nal"mal Lah.,ralt)rv R':p',rt No, BLN 5()784, March, 197:11 In" PL' CEES Rcp"rt No. 193, (C.:nt.:r for En.:rIt.. and En"in,nm.:ntal Studi"s. Prin"clon Univ"r-

the Institute of PhYSICS, Un I verslty 0 .f Sao Paulo and president of the energy companies of Sao Paulo (CESP. ELETROPAULO CPFL COMGAS). " He may be reached at Alameda Mlnlstro Rocha Azevedo 25-8th floor, 01410-Sao Paulo, Brazil. Thomas B.0J h ansson IS Pro f essor . of Energy Systems AnalysIs at the University of Lund. His address: Environmental Studies Program Ger' dagatan 13, S-22362 Lund, Sw~en. Amulya K. N. Reddy is Vice-Chairman of Karnataka State Council for Sllence and Technolo and is a, p . fessor at the Indian Institute of SCIence. His address: Department of Inorganic and Physical Chemistry, Ind. I t' t t f S Ian ns I u.e 0 clence, B anga Iore, 560012, India. Robert H. Williams is a senior research scientist at the Center for

is a professor .





, and Environmental

.. StudIes


12. Amilcar o. H.:rr"ra tl ai" ClllttStrtJph.. N..,,' or Soci..,_\': LtJtin Am..ri.'an World ,~/I)d..l..r':p'.rt A uI th.: Fundaclun BanltlCh". (Inl"rn.t",nal 0,,'.:Iupm"nl R"",arch C.:nl"r. Ollaw., 1'17/1) 13 How.rd S G"II"r, Saln Baldwin, Gaulam S Dull. and ~ II R""indran.lh, "Impn'",d C""k"lo\"" Sign,ufSu"c"":'(,4,"hi'J.lhi,i,,uc) 2(XJ

S;I~,Princ':lon. N"w J"rs.:y. July. 19K5) 3/1 K. C: .~hukla .nd J. R, Hurl.:~. D..v..lopm"'~IIJf tlndEJ/I",..""Lol,'NO.Oom..sucGasRanli...Cook Top. R':p"rl pr"par"d h~ Th"rmtlCl,,"lron <'urp')r.tion, Walth.m, ~I",sachus""', fur Ih" I G", R,,",.r.:h In'lilul", Jul~, I'/X.'I. .1" R II \Yilli.m" G S Dull. .nd II S (;"II"r,

Princeton, New Jersey. His address: Center for Energy and Environmental ., n New Jerse Studies, Prlnceto. Y. 08544, U.S,A.


t~ NO ~-5


12 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Microsoft Word - ECFA Report_Final_.doc
Microsoft Word - WETO-H2 report-final.doc