Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

72

“To draw attention today to technologicalaffairs is to focus on a concern that is as centralnow as nation building and constitution makingwere a century ago. Technological affairs con-tain a rich texture of technical matters, scientificlaws, economic principles, political forces, andsocial concerns.”

—Thomas P. Hughes, 1983

The American electric utility system hasbeen massively transformed during the last threedecades. Viewed previously as a staid, secure,and heavily regulated natural monopoly, the sys-tem has shed elements of government oversightand now appears to be increasingly susceptibleto terrorist attacks and other disruptions.Overturning the conventional wisdom of the1960s and 1970s, dependence on large-scalegeneration plants and high-voltage transmissionnetworks now seems to decrease reliability ofthe system. The same hardware also subjectsutility companies to pollution concerns at a timewhen environmental sustainability has become aserious political and economic issue worldwide.Clearly, much of the formerly sound thinkingabout the utility system has been overturned.But new thinking has evolved as well. For exam-ple, many people have begun advocating the useof small-scale, decentralized technologiesknown collectively as distributed generation(DG), which offers hope that the electric utilitysystem may become more resilient and environ-mentally friendly in the near future.

This article examines the reversal of almosta century of momentum in the electric utilitysystem. To do so, it employs the nonengineeringversion of the systems approach, which hasproven useful for analyzing sociotechnical enter-prises. As a subtheme, the article explores long-term and cross-industry trends in the use oftechnologies; in particular, it notes that industri-al use of large-scale technologies, which providegreat economies of scale, may have reached lim-its in certain fields, such that small-scale tech-nologies may have become better suited for use.As another subtheme, the article underscores theimportance of nontechnical (i.e., social) actionin the development of technologies. In the

American utility system, the unintended conse-quences of an obscure law passed in 1978spurred deregulatory efforts in the 1990s and thedevelopment of commercially viable renewableenergy and distributed generation technologies.The encouragement of these environmentallypreferable and DG facilities continued duringthe political chaos that emerged early in thetwenty-first century, when utility restructuringbecame questioned in many states.

Using the Systems Approach toUnderstand Technological Change

To help understand the historical develop-ment of electric utilities, one can fruitfully usethe “systems approach” that Thomas Hughesoriginally developed for the social sciences. Inhis seminal Networks of Power: Electrificationin Western Society (Hughes 1983) and otherworks, Hughes argues that the generation, trans-mission and distribution of electricity occurswithin a technological system. That systemextends beyond the engineering realm, and itincludes a “seamless web” of considerations thatcan be categorized as economic, educational,legal, administrative, and technical. Large mod-ern systems integrate these elements into onepiece, with system-builders striving to “con-struct or ... force unity from diversity, centraliza-tion in the face of pluralism, and coherencefrom chaos” (Hughes 1987, 52). If the managerssucceed, the system expands and thrives while,simultaneously, closing itself. In other words,the influence on it of the outside environmentmay gradually recede because the system hasexpanded its reach to encompass factors thatmight otherwise alter it (Hughes 1987, 52).

Hughes’s systems approach also employsthe notion of momentum, which Networks ofPower describes as a mass of “machines,devices, structures” and “business concerns,government agencies, professional societies,educational institutions and other organizations”that “have a perceptible rate of growth or veloci-ty” (Hughes 1983, 15). In fewer words, Hughesdefines momentum elsewhere as a “mass oftechnological, organizational and attitudinalcomponents [that tend] to maintain their steadygrowth and direction” (Hughes 1989, 460). The

Technological Systems and Momentum Change:American Electric Utilities, Restructuring, andDistributed Generation TechnologiesRichard F. Hirsh and Benjamin K. Sovacool

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

73

system’s tendency to continue along a given pathresults from the actions of numerous stakehold-ers, such as educational and regulatory institu-tions, the investment of billions of dollars inequipment, and the work and culture of peopleworking within an industry. In concert, theseelements promote business as usual and the out-ward show of a large degree of momentum.Furthermore, Hughes explains that systemmomentum can be aided through the use of“conservative” inventions, i.e., new technologiesthat preserve the existing system. Managers ofthe system obviously prefer to maintain theircontrol of affairs and, while they may seekincreased efficiencies and profits, they do notwant to see introduction of new and disruptive“radical” technologies. As an example, Internettelephony and cell phones may be viewed todayas radical inventions by managers of the tradi-tional wired communications network.

To summarize, Hughes’s approach suggeststhat systems develop through human managers’control of elements to exploit the existing socialenvironment. Systems acquire momentum, andthey resist alteration as they mature.Momentum, however, is not the equivalent ofdeterminism (which holds that either social ortechnical factors are exclusively responsible fortechnological development) or autonomy (whichsuggests that technologies achieve a level ofaction independent of human contexts). Rather,momentum in systems—even those having highmomentum—can sometimes be altered through“a confluence of contingency, catastrophe andconversion,” as Hughes puts it (1989, 470-471).Through their institutions and actions, humansmay play a role in changing momentum, diffi-cult as that may appear in many situations. Butsuch was the case in the American electric utili-ty system.

Origins and Growth of Momentum inthe Electric Utility System

The electric utility system began to gainmomentum early in the twentieth century. It didso largely as power company managers tookadvantage of incrementally improving technolo-gy. Adopting alternating-current (AC) transmis-sion and steam-turbine technologies, early utilityentrepreneurs quickly found they could emulatetrends in other industries, most notably byincreasing the scale of their business. EconomistAlfred Chandler (1977) has shown how the rail-road industry served as the epitome of modern

enterprises that exploited new technologies toexpand their scale and scope. In the process,railroad companies eliminated regional competi-tion and gained panoptic political and economicpower, while also earning the contempt of manyof their customers. Similarly, through the use oftransformers made available in the late 1880s,AC transmission allowed companies to distrib-ute high-voltage power over long distances. Andin the first decade of the twentieth century, utili-ty managers began employing compact steamturbines (connected to generators) as primemovers that further encouraged centralization,because the new machines offered tremendouseconomies of scale. Put simply, as the turbinesproduced larger capacities of power, the unitcost of electricity declined over a wide range ofoutput. Taking actions that would be imitated byother electrical entrepreneurs, Samuel Insullembraced alternating current and steam turbinesfor his small Chicago Edison Company. Whenhe took over the firm in 1892, Chicago claimed20 competitive electric supply firms. By 1907,Insull employed the new technology and consol-idated them into the renamed “CommonwealthEdison Company,” a virtual monopoly in thecity and its environs (Hughes 1983, 208).

Steam turbine-generators and alternating-current technologies became the core conserva-tive technologies of the utility system, allowingit to produce increasing amounts of power atlower unit cost. Developed for utility use bycompanies such as Westinghouse and GeneralElectric, the technology showed improvementsmainly in two criteria. First, manufacturersdesigned steam turbines using new metal alloysthat could endure higher temperatures and pres-sures, thus yielding better thermal efficiencies.While Thomas Edison’s pioneering 1882 powerstation (which used reciprocating steam engines)converted just 2.5 percent of a raw fuel’s energyinto electricity, steam turbines improved over theyears to achieve a conversion efficiency of about40 percent in the 1960s, though the averageplant demonstrated an efficiency of about 33percent (Figure 1). At the same time, manufac-turers exploited new knowledge about metallur-gy to produce turbines and generators thatbecame ever-larger and took advantage ofeconomies of scale. Following installation of a5-megawatt (MW) steam turbine in 1905, forexample, Samuel Insull’s firm procured 12-MWmachines in 1911 (Insull 1915, 430). More pow-erful equipment followed: Insull’s company

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

74

employed a 208-MW unit in 1929. After theGreat Depression, other American utilitiesbought units that reached 1,000 MW in 1965and 1,300 MW in 1972.

The growth in scale of electric utility tech-nology parallels trends in other industries. As inthe manufacturing of automobiles and consumerdurables and in service industries and sanitation,managers employed technologies that reducedunit costs, increased efficiency, and added toprofitability as the scale of operations increased(Melosi 2000; Noble 1999). Some of the scaleefficiency occurred for “simple” laws of mathe-matics and engineering. For example, as the cir-cumference of a pipe that feeds a steam turbineincreases by a factor of three, the volume ofsteam passing through it increases by a factor ofnine. In other words, for a modest increase inmaterial inputs (and costs), the industrialistobtains disproportionately large outputs (Hirsh1989, 41). More gains emerge when realizingthat many other costs decline as scale increases.Instead of employing two small steam turbines,along with the associated equipment and person-nel needed to operate them, a utility could buildone large turbine. The power output is equiva-lent, but at lower unit cost.

Because of associated improvements intransmission, distribution, and control equip-ment, the use of more fuel-efficient and largerturbine-generators boosted the industry’s totalfactor productivity—a measure of overall oper-ating efficiency (Kendrick 1961; Kendrick1973). In everyday terms, higher productivity

meant declining costs and prices. While residen-tial customers in 1892 paid about 544 cents perkilowatt-hour (kWh) (in adjusted 2004 terms),they only paid 10 cents for the equivalentamount of electricity in 1970 (Edison ElectricInstitute 1988; Edison Electric Institute 1997;U.S. Department of Energy 2005a). Since lowerprices and the availability of attractive electricalappliances stimulated demand, consumptionjumped at a 12 percent annual growth rate from1900 to 1920 and at a 7 percent average annualrate from 1920 to 1973 (Landsberg and Schurr1968).

While the use of conservative, incremental-ly improving technologies contributed to theutility system’s growing momentum, so did non-technical elements. By focusing attention onsuch elements, Hughes’s system approach helpsus understand that technology evolves in a socialmilieu. Perhaps most surprisingly, the creationof regulatory oversight of the utility industrycontributed significantly to momentum. As aca-demics and politicians began to realize duringthe Progressive era in American politics, whichlasted from about 1896 to the beginning ofWorld War I, certain types of businessesappeared unable to operate within the traditionalcompetitive market environment: they requiredconstruction of capital-intensive facilities thatlimited the number of rivals to those few thatcould secure financing. Meanwhile, competitionamong firms had already led to political corrup-tion in attempts to win franchise rights from cityleaders. These businesses, such as railroads andthe increasingly large electric power companies,

Thermal Efficiency of Power Units/Plants1882-2003

40.0%

23.9%

10.3%

13.4%

5.6%

33.5% in 1998

2.5%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

1882

1887

1892

1897

1902

1907

1912

1917

1922

1927

1932

1937

1942

1947

1952

1957

1962

1967

1972

1977

1982

1987

1992

1997

2002

Year

Th

erm

alE

ffic

ien

cy

Maximum Efficiency of Plant or UnitAverage Efficiency of Plant or Unit

Figure 1. Data from Edison Electric Institute 1988, Electric Light and Power, andU.S. Department of Energy 2004.

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

75

constituted “natural monopolies”—businessesthat offered services most efficiently and cheap-ly only if they remained free from competition(Hirsh 1999, 17-18). Regulation became viewedas a popular approach for gaining the benefits oflarge-scale companies while protecting the pub-lic from potential abuses. In other words, theexistence of natural monopolies seemed to callfor the regulatory commission as a new, sociallyvaluable institution.

While casual observers might assume thatindustrial tycoons would resist government regu-lation, Samuel Insull and other prescient man-agers realized that regulation actually could helpthem to consolidate a fledgling industry. Asearly as 1898, Insull argued that governmentoversight would confer legitimacy to powercompanies as natural monopolies and, as impor-tant, it would enable utilities to reduce their costof financing by lessening the risk incurred byinvestors. Regulation would also protect utilitiesfrom attempts at municipal takeovers (Insull1915, 34).

Government oversight of utilities added tosystem momentum, ironically by giving utilitymanagers a large measure of control. Becauseregulators guaranteed an almost certain incomestream and financial viability, utilities benefited.Moreover, regulation offered the appearance ofwatchful supervision without really providingmuch, thereby providing a large degree of cloutto utility managers. With the end of theProgressive era, state regulators received littleattention from the news media or state legisla-tors, which provided few resources to combatthe utilities’ well-paid lawyers and consultants.Meanwhile, utility managers courted regulators,and they publicly exaggerated the positive rolethat regulators played in maintaining a success-ful large-scale enterprise. Regulators did little todisabuse the public of this notion. They happilywent along with the charade because theyenjoyed the little prestige and power theyacquired by being associated with an industrythat provided declining prices for an increasing-ly necessary commodity (Hirsh 1999, 41-46). Inshort, utility managers had “captured” the regu-latory apparatus, which contributed to their con-trol of a system that had developed growingmomentum (Stigler 1971; McCraw 1984).

Beyond regulators, utility managers won thesupport of other stakeholders who sought tokeep the system moving in the same direction

and with continuously increasing mass andvelocity. Investment bankers constituted onesupportive party as they profited from supplyingmoney for the capital-intensive industry. Theyalso helped create holding companies, whichoffered power company managers access tofinancial resources and professional manage-ment expertise. Manufacturers of electricalequipment further contributed to momentumbecause they flourished monetarily when utilitycompanies expanded and required moreadvanced technologies. To train utility execu-tives and middle-managers, educational institu-tions (starting with MIT and Cornell University)fashioned degree programs for electrical engi-neers and also became tacit contributors ofmomentum. Utility customers, meanwhile,appeared happy, as electrification benefited peo-ple with higher material standards of living anda sense of social progress (Hirsh 1989, 26-35).They did little to impede the growing momen-tum and, through their increasing purchases ofelectricity, added to it.

By the 1920s (and continuing into the1960s), the utility system had gained a hugeamount of momentum. Serving as an integralpart of the infrastructure that enhanced industri-al productivity and made the “good life” possi-ble, the system employed incrementally improv-ing and conservative technologies. Moreover,utility managers gained control of the system bycapturing the regulatory framework and by win-ning support from financial, industrial, educa-tional, and consumer stakeholders—all of whomseemed to reap tangible benefits (Hirsh andSerchuk, 1996). On the surface, it appeared thatlittle could alter the system’s trajectory.

Momentum Change BeginsThe first challenge to continuously building

momentum in the electric utility system arose inthe 1960s and 1970s. In a process that has beencalled “technological stasis,” the industry wit-nessed the apparent end of productivity-enhancingtechnological improvements. For a host of mana-gerial and technical reasons, the thermal efficien-cy of steam turbine-generators stopped improving.Though some plants reached efficiencies as highat 40 percent, they also proved to be highly unreli-able, which made them unattractive to managerswho sought to keep plants online as long as possi-ble. As bad experience grew with these highlyefficient plants, managers stopped ordering them,remaining satisfied with the less-efficient, butmore trustworthy units (Hirsh 1989, 89).

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

76

Soon thereafter, technological stasisoccurred again, this time as utilities sought tobuild ever-larger, higher-capacity power plants.As units grew to exceed 1,000 MW in the 1960sand 1970s, utilities also discovered operationalproblems as turbine blades suffered distortionsand as furnace problems and other defects ham-pered the units’ performance (Figure 2). Despitethe desire to attain increased economies of scale,especially in light of the industry’s inability toreach higher thermal efficiencies, managersstepped back from the largest units and settledfor smaller, proven units. The problem becameincreasingly serious in light of the 1973 “energycrisis,” which caused fuel to become moreexpensive. In previous decades, the improve-ment of power-producing technologies had miti-gated increased costs in building materials andlabor, allowing the industry to reduce the priceof its product. But because of technological sta-sis—along with greatly increased borrowingrates for an industry that consumed vastamounts of capital—power companies soonfound themselves uncharacteristically requestingrate hikes from utility commissions (Hirsh 1989,111). Additionally, the higher prices createdantipathy among customers who previously sup-ported the structure of the utility system. Moretangibly, higher prices motivated customers topurchase less power: for a few years after theonset of the crisis, electricity consumption actu-ally grew at negative rates, while the longer termannual growth rate (from 1973 to 2003) droppedto about 2.4 percent (Figure 3).

Politics and System MomentumThe havoc on the American economy

wreaked by the energy crisis spurred policymakers to take action. On the federal level,President Carter made energy policy his firstmajor initiative. Among a set of five laws pro-posed by Carter and passed by Congress (albeitin greatly diluted form), the Public UtilityRegulatory Policies Act (PURPA) of 1978 hadthe most far-reaching—and least intended—con-sequences for power companies. It spurred cre-ation of radical technologies; it began theprocess of deregulation; and it challenged thecontrol held by power company managers. In theprocess, the law helped change the momentumof the utility system.

Superficially, PURPA appeared to pose littlethreat to utility companies. But one obscure por-tion of it offered incentives for the use of effi-cient cogeneration power plants. These smallunits often produced less than 100 MW ofpower—about 10 percent of what utility-ownednuclear and fossil units churned out—and theyburned coal, natural gas, garbage, or biomass.The resulting heat first produced electricityusing a traditional steam turbine. But instead ofdumping the low-pressure steam into the envi-ronment, as was common practice for utilitypower plants, cogenerators employed it forindustrial processes. In other words, smallcogeneration plants obtained double duty fromraw fuel in the form of two valuable products(electricity and process steam), yielding an

Figure 2. Data from U.S. Department of Energy 2003.

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

77

overall thermal efficiency rate of 50 percent orhigher. (Compare this figure to the 40 percentoptimum achieved by utility power plants.)Because of this feat, cogeneration plants fre-quently matched the economic viability of utilityunits and became increasingly popular, even attheir relatively puny scale. Growing from asmall percentage of American electric capacityin 1980, cogeneration facilities accounted for 39percent of capacity in 2002 (Electric PowerSupply Association, 2005).

A related provision of PURPA also spurredresearch on environmentally preferable tech-nologies that used water, wind, or solar power toproduce electricity. More successful than anyoneoriginally anticipated, PURPA prompted workthat cut the cost of power produced by solarphotovoltaic panels by about 70 percent between1980 and 1995 (Sheer 2001; Clayton 2004).More significantly, it contributed to work thatlowered the cost of power produced by wind tur-bines; by 2002, the average cost of wind-pro-duced electricity dropped to under 5 cents perkWh, a cost that compared favorably with elec-tricity produced by conventional utility plantsburning natural gas or coal (Smith 2004;American Wind Energy Association 2005).

These smaller-scale generation technologieschallenged the established paradigm of the utilityindustry (and many other industries) that previ-ously relied on large-scale equipment to produceeconomies of scale. Now, it appeared, powerplants producing modest amounts of electricity

proved economically viable. Moreover, sincethey were smaller, they required less time tobuild, and they put less capital at risk during a period of rapid price inflation. Finally, theymatched the slower growth rate of consumptionmore appropriately: with growth rates remainingunder 3 percent per year, consumers neededsmaller increments of power to match theirdemand. Had utilities continued to build theirtraditional behemoths, huge chunks of powerwould remain unused when the plants were completed. Small scale, indeed, looked beautiful.

In the parlance of Hughes’s approach, thesesmall-scale technologies constituted radicaltechnologies that helped alter the utility system’smomentum. They did so by enabling competi-tion for the traditional power companies—atleast in the generation sector—and by erodingthe control held by utility managers. No longerthe sole producers of electricity, power companyexecutives watched as industrial firms (employingcogeneration facilities) and renewable energyentrepreneurs sold competitively priced electricity.

This unintended experiment with competi-tion suggested to influential regulators and legis-lators in the 1980s that more competition wouldbenefit stakeholders in the electric utility indus-try. (Not coincidentally, competition had alreadybegun in the airline, telecommunications, andnatural gas industries.) Some academics andpoliticians wondered if utility regulation stillhad merit, seeing that a traditional justificationof government oversight—the fact that power

Growth Rates of Electricity Consumption, 1960 to 2003

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

Year

Per

cen

tAnnual Growth Rate

5-Year Moving Average of Growth Rate

Figure 3. Data from U.S. Department of Energy 2004.

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

78

companies constituted natural monopolies—nolonger appeared valid. After all, if nonutilitiescould produce power as cheaply as could utili-ties, then the big power companies no longerdeemed recognition as natural monopolies. Andif they were not natural monopolies, they nolonger deserved special status as noncompetitiveentities that required regulation (Hirsh 1999,119-142). Why not permit increased competitionto thrive outside the realm of the PURPA-inspired generation companies?

Already feeling their control of the systemthreatened by pressures resulting from imple-mentation of PURPA, utility managers had moreto fear. After the Gulf War focused attentionagain on the cost and security of energy sup-plies, Congress passed the Energy Policy Act of1992. The legislation sought to employ competi-tive forces to increase domestic fuel productionand to improve the efficiency of energy use.One provision gave states the option of openingup their transmission network to use by competi-tors. The network would serve as a common car-rier so any electricity producer could sell powerto any customer. Essentially, the law gave statesthe right to begin competition on the retail level.During the late 1990s, several states (withCalifornia being among the first) passed legisla-tion that established competitive retail frame-works for power. By September 2001, 23 states(and the District of Columbia) had passed simi-lar legislation, while regulatory bodies in severalother states had reduced their oversight and hadintroduced market forces into the system.

The restructuring process, stimulated byPURPA and pursued by advocates of marketforces, meant that momentum in the utility system had been altered, largely because utilitymanagers lost control of “their” system. Foralmost a century, managers commanded thehuge-scale, incrementally improving conserva-tive technologies that produced and distributedelectricity. But in recent decades, they beganfacing competition from entrepreneurial compa-nies that employed small-scale fossil fuel andrenewable energy technologies. In 1992, nonutil-ity companies controlled 1.5 percent of thenation’s total power capacity; that number roseto 34.7 percent in 2003 (U.S. Department ofEnergy, 2005b). In addition, power companymanagers lost the benefit of traditional regula-tion, which protected companies from competi-tion on the basis of their claim—now chal-lenged—to natural monopoly status. At the same

time, managers lost support from those stake-holders who previously buttressed them: finan-ciers, equipment manufacturers, and educationalinstitutions adapted to the changing environmentand found new opportunities serving the needsof the growing number of nonutility companies.Employing the terms of the systems approach,then, the sociotechnical elements that had previ-ously contributed to utility system momentumno longer had the same speed nor direction.

Assuming some of the political and eco-nomic power that managers once held, otherstakeholders began making new waves.Environmental advocates, for example, gainedimpressive standing in the legislative processthat led to creation of restructuring laws. Whilesupporting deregulation in general, they foughtfor (and in many cases won) provisions in lawsthat guaranteed funding for renewable energyand energy-efficiency initiatives. In California’sversion of restructuring, for example, utilitiesearned the right to receive payment from cus-tomers for building what turned out to be expen-sive power plants during the era of regulation,but which would have little economic value inan era of competition (called “stranded” assets).But environmental advocates won provision forexpenditures on energy efficiency work, renew-able energy technologies, and for developmentof research on new technologies that had not yetshown commercial viability (Hirsh 1999, 258).

The Promise of DecentralizedElectricity Generation

As stakeholders began renegotiating theirpositions in an altered utility system, theCalifornia electricity crisis of 2000 and 2001created a sense of chaos. Subsequent blackoutsin the Midwest during 2002 and in the Northeastand Canada during 2003 contributed further tothat disharmony. In California, where one utilitydeclared bankruptcy as a result of the crisis, thestate suspended competition altogether andexpanded its control of the wholesale market.Such events caused policy makers in other statesto reconsider their previous enthusiasm forrestructuring and to slow down plans to intro-duce market forces. At the same time, theFederal Energy Regulatory Commission pro-posed new initiatives in response to the 2003blackout that may give it greater control over theincreasingly fragile-looking transmission grid.That grid has witnessed serious underinvestmentsince the 1990s as utility companies and nonutil-ity entrepreneurs remained concerned in an

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

79

uncertain policy setting about how the grid willbe employed and which stakeholders will profitfrom its use (Rigby 2003).

The unsettled state of affairs in the powersystem has provided opportunities for advocatesof environmentally-friendly and distributed-gen-eration technologies. Taking advantage of theflux within the utility system, especially instates with strong traditions of politically astuteenvironmental advocates, activists won passageof laws for funding of renewable energy andsmall-scale generation technologies. Customerspaid into “public benefit funds” (also known as“system benefit funds”) regardless of whichcompany (a traditional utility or nonutility com-pany) provided them with electricity (CleanEnergy States Alliance, 2004). And in eighteenstates (plus the District of Columbia), advocatesconvinced legislators to enact laws creating“renewable portfolio standards” that required allcompetitive power companies to produce (or topurchase) a certain amount of power comingfrom small-scale, renewable resources. The lawscreated a market for environmentally preferabletechnologies and spurred research and develop-ment of them so they become more economical-ly viable even without government incentives. InTexas, a law (signed by Governor George W.Bush in 1999) required 2,000 MW of renewableenergy to be constructed by 2009 (Wiser andLangiss, 2001). By the end of 2004, producershad already installed 1,293 MW, most of it con-sisting of competitively priced wind turbines(Garman 2005).

The state-supported move toward small-scale generation technologies reflects the ero-sion of the conventional paradigm of providingpower through large, centralized power genera-tors. Traditionally, policymakers emphasizedusing the discipline of power engineering tofocus on expanding generating capacity to meetaggregated demand at the lowest cost. Thisapproach relied on large plants because utilitymanagers believed that bigger facilities offeredlower fixed costs and were better driven byeconomies of scale (Capehart, Mehta, andTurner 2003; Lovins et al. 2002).

By contrast, an emerging paradigm of dis-tributed generation emphasizes decentralized,modular, and on-site production of electricity.Proponents of DG argue that small, local plants(often having capacities of between 5 and 30MW) are frequently less expensive, more effi-

cient, more reliable, more flexible, and lessdamaging to the environment than large utility-owned power plants that burn fossil fuels. Sincemany renewable energy technologies—such aswind turbines and photovoltaic systems—aresmaller and decentralized, they serve as exam-ples of DG technologies. Even generation equip-ment that employs fossil fuels, such as naturalgas and coal, can be considered valuable DGresources, especially when they exploit wasteheat generated through combustion or fuel con-version. These comprise cogeneration plants(also known as combined heat and power [CHP]facilities) like those encouraged by PURPA.They may also include very small steam tur-bines known as micro-turbines that producepower in the range of 25 to 500 kW while alsorecycling waste steam for water or space heatingneeds (Capehart 2003). Phosphoric acid fuelcells have become popular too. Being pursuedby public benefit funding programs in somestates, the devices (which are being tested in 200kW to 11 MW sizes) take in hydrogen-rich fueland create electricity through a chemical processrather than by combustion, yielding few particu-late wastes (Yacobucci and Cutright, 2004). Theprocess also generates heat, which can be usedas a profitable by-product. Beyond these DGtechnologies that offer high efficiencies aremodular internal combustion engines that run ondiesel fuel or gasoline. In their smallest incarna-tions (about 1 kW), they provide backup powerfor homes and recreational vehicles. In largerincrements, they supply power for hospitals andcommercial venues. Generally speaking, DGtechnologies can be employed in isolated“islands”—independent of the transmissiongrid—or attached to it, in which case they serveas small power units that contribute power to it.

Consumer advocates who favor DG pointout that distributed resources can improve theefficiency of providing electric power. Theyoften highlight that transmission of electricityfrom a power plant to a typical user wastesroughly 4.2 to 8.9 percent of the electricity as aconsequence of aging transmission equipment,inconsistent enforcement of reliability guide-lines, and growing congestion (Silberglitt,Ettedgui, and Hove 2002; Lovins 2002). At thesame time, customers often suffer from poorpower quality—variations in voltage or electricalflow—that results from a variety of factors,including poor switching operations in the net-work, voltage dips, interruptions, transients, and

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

80

network disturbances from loads. Overall, DGproponents highlight the inefficiency of theexisting large-scale electrical transmission anddistribution network. Moreover, because cus-tomers’ electricity bills include the cost of thisvast transmission grid, the use of on-site powerequipment can conceivably provide consumerswith affordable power at a higher level of quali-ty. In addition, residents and businesses thatgenerate power locally have the potential to sellsurplus power to the grid, which can yield sig-nificant income during times of peak demand.

Industrial managers and contractors havealso begun to emphasize the advantages of gen-erating power on site. Cogeneration technologiespermit businesses to reuse thermal energy thatwould normally be wasted. They have thereforebecome prized in industries that use large quan-tities of heat, such as the iron and steel, chemi-cal processing, refining, pulp and paper manu-facturing, and food processing industries(International Energy Association 2002, 41-49).Similar generation hardware can also deployrecycled heat to provide hot water for use inaquaculture, greenhouse heating, desalination ofseawater, increased crop growth and frost pro-tection, and air preheating (Kleinbach andHinrichs 2002, 311).

Beyond efficiency, DG technologies mayprovide benefits in the form of more reliablepower for industries that require uninterruptedservice. The Electric Power Research Institutereported that power outages and quality distur-bances cost American businesses $119 billionper year (Hinrichs, Conbere, and Lobash 2002).In 2001, the International Energy Agency (2002)estimated that the average cost of a one-hourpower outage was $6,480,000 for brokerageoperations and $2,580,000 for credit card opera-tions (44). The figures grow more impressivelyfor the semiconductor industry, where a two-hour power outage can cost close to $48,000,000(Lin 2004). Given these numbers, it remains no mystery why several firms have alreadyinstalled DG facilities to ensure consistentpower supplies.

Perhaps incongruously, DG facilities offerpotential advantages for improving the transmis-sion of power. Because they produce powerlocally for users, they aid the entire grid byreducing demand during peak times and by min-imizing congestion of power on the network,one of the causes of the 2003 blackout

(Congressional Budget Office 2003). And bybuilding large numbers of localized power gen-eration facilities rather than a few large-scalepower plants located distantly from load centers,DG can contribute to deferring transmissionupgrades and expansions—at a time wheninvestment in such facilities remains constrained(International Energy Agency 2002; Pepermans2003). Perhaps most important in the post-September 11 era, DG technologies mayimprove the security of the grid. Decentralizedpower generation helps reduce the terrorist tar-gets that nuclear facilities and natural gasrefineries offer, and—in the event of an attack—better insulate the grid from failure if a largepower plant goes down (Friedman and Homer-Dixon 2004).

Environmentalists and academics suggestthat DG technologies can provide ancillary ben-efits to society. Large, centralized power plantsemit significant amounts of carbon monoxide,sulfur oxides, particulate matter, hydrocarbons,and nitrogen oxides (Kleinbach and Hinrichs2002, 249). The Environmental ProtectionAgency has long noted the correlation betweenhigh levels of sulfur oxide emissions and thecreation of acid rain. Because they concentratethe amount of power they produce, large powerplants also focus their pollution and waste heat,frequently destroying aquatic habitats andmarine biodiversity (Kleinbach and Hinrichs2002, 307-308). On the other hand, recent stud-ies have confirmed that widespread use of DGtechnologies substantially reduces emissions: ABritish analysis estimated that domestic com-bined heat and power technologies reduced car-bon dioxide emissions by 41 percent in 1999; asimilar report on the Danish power systemobserved that widespread use of DG technolo-gies have cut emissions by 30 percent from 1998to 2001 (International Energy Association 2002,92). Moreover, because DG technologies remainindependent of the grid, they can provide emer-gency power for a huge number of public servic-es, such as hospitals, schools, airports, fire andpolice stations, military bases, prisons, watersupply and sewage treatment plants, natural gastransmission and distribution systems, and com-munications stations (Liss 1999). Finally, DGcan help the nation increase its diversity of ener-gy sources. Some of the DG technologies, suchas wind turbines, solar photovoltaic panels, andhydroelectric turbines, consume no fossil fuels,while others, such as fuel cells, microturbines,

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

81

and some internal combustion units burn naturalgas, much of which is produced in the UnitedStates. The increasing diversity helps insulatethe economy from price shocks, interruptions,and fuel shortages (Herzog, Lipman, andEdwards 2001).

Although estimates vary, total-customerowned generation of distributed resourcesremains comparatively small, although it isexpected to grow rapidly. The EnergyInformation Administration classified only 0.5percent of electricity generation in 2000 as“nonutility” generation while noting that cogen-eration systems in industrial sectors produced3.6 percent of electricity for their own use(Congressional Budget Office 2003). However,an Electric Utility Consultants (2004) study con-cluded that distributed generation units—classi-fied as decentralized generators between 1 and60 MW—provide an aggregate capacity of 234gigawatts (GW) through approximately 12.3million units. (The total capacity of Americanpower plants in 2003 was 948 GW). TheAmerican Gas Association (2000), meanwhile,predicted that renewable energy technologieswill provide over 15 percent of electricity in theUnited States by 2020, and the U.S. Departmentof Energy’s Energy Information Administration(2000) projected that distributed generationtechnologies will likely surpass 25 percent ofelectricity generation by the same year.

Regardless of these projections, proponentsof distributed generation caution that DG tech-nologies are not designed or intended to replacethe power grid. The United States CombinedHeat and Power Association notes that DG facil-ities could not reliably serve as substitutes forall large, centralized power plants since theycannot meet all customers’ needs (especiallycustomers in urban areas with severe land, per-mitting, and citing constraints) (Fallek 2004).The American Wind Energy Association (2004)adds that wind turbines will primarily augment,rather than replace, the grid.

As compelling as DG appears, the transitionto a new paradigm that employs it widely willnot necessarily occur smoothly due to a host ofsocial, technical, and political impediments. Byfar the largest social constraint consists of thebelief that renewable energy systems and decen-tralized units have higher operating costs thanlarge centralized facilities. As a result, manypolicymakers and consumers believe that renew-

able technologies have not become cost compet-itive with traditional fossil fuel sources of elec-tricity (Pepermans 2003; Colorado RenewableEnergy Society 2002). In addition, the financialpayback period for renewable energy systemsstill appears too long for fiscally conservativecustomers. Businesses often look for two tothree year timeframes for obtaining a positivecash flow after making capital investments. Yetwind turbines require about six to nine years forpayback, and their installation sometimes suffersfrom long project lead times, lack of suitableservice infrastructure, and poorly developedsales channels (Rendon 2003; Liss 1999).

The biggest technical impediment consistsof connecting DG technologies to the grid. Thisinterconnection challenge really constitutes a setof related smaller problems: voltage control(keeping voltages within a certain range), thebalancing of reactive power (using power to reg-ulate the flow of alternating current to maintainproper grid synchronization), and safety (ensur-ing the adequate protection of people workingon the grid) (International Energy Agency 2002,73-75). Currently, those using DG technologiesoften depend on custom-designed electronicspackages to solve these problems. The greatexpense in developing such packages obviouslycreates a disincentive for new users (Ostergaard2003; Pepermans 2003).

Perhaps more significant are the challengesrelating to regulatory and political inertia withinthe United States. Lingering monopoly rules anddiscriminatory rate structures still create politi-cal obstacles towards investing in DG technolo-gies. Customers who seek to use DG often facehigh exit fees imposed by utilities that seek torecover stranded costs. Further giving an unfairadvantage to existing producers, utilities andother owners of old coal-fired power plants havewon exemption from 1970 Clean Air Act. These“grandfathered” plants can often operate morecheaply than any type of new generation tech-nology. They therefore enjoy an artificial advan-tage over some DG technologies, which mustmeet current and future environmental standards(Casten 1998, 203; Meyer 2003).

ConclusionThe tentative move toward DG represents

a somewhat paradoxical return to the electricutility industry’s roots. Thomas Edison inaugu-rated the industry in 1882 when he supplieddirect-current (DC) power to a host of nearby

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

82

businesses in the financial district of New York.Because DC power could not be transmittedover long distances, Edison hoped to sell small-scale generation equipment to commercial cus-tomers (such as hotels and industrial firms) thathad large demands for electricity. Individualhomeowners would be served by small powerstations like Edison’s original, which would bedotted throughout cities at regular intervals. Thismodel, of course, became displaced by one thatemployed large centralized power stations tocreate huge amounts of power sent over alternat-ing-current transmission networks. Overcomingthe problem of limited distance distributionposed by Edison’s DC arrangement, the ACapproach lent itself better to the economies ofscale that became apparent in generating equip-ment over the next eight decades.

But even consideration of a return to the use of distributed generation resources high-lights the significant change in momentum thathas occurred within the electric power system.Unlike in the period from the early twentiethcentury to the 1970s, the technological basis forthe system remains in flux today: large-scalegeneration and transmission technologies nolonger offer prized economic benefits, and theyrecently have begun to be seen as security andreliability threats to a society that increasinglydepends on electricity. The experience containsparallels with other industries that once valuedlarge scale. In the steel, biotechnology, agricul-ture, microelectronics, pharmaceutical, and mining industries, companies have recentlyemployed small-scale technologies to gainincreased flexibility, security, and lower costs.

The utility system’s momentum has beenaltered by more than the shift away from tradi-tional large-scale technologies. It has also beenchanged by those who wield economic andpolitical power. As utility managers lost the ben-efit of state regulation, which previously guaran-teed their companies’ status as natural monopo-lies, they found themselves competing withentrepreneurs who employed small-scale tech-nologies. Simultaneously, other supportingstakeholders (i.e., financiers, equipment suppli-ers, academic institutions, and politicians) aban-doned their former allies, realizing that theycould benefit from making alliances with newplayers in the system. As utility managers lostcontrol over the system, new players, such asenvironmental advocates and policy makers,

sought to advance their own (and different)visions of a utility paradigm. During the chaoticfirst years of the twenty-first century, the oldand new stakeholders have not yet reached con-sensus for a new paradigm for the utility system.In such an untidy environment, advocates of dis-tributed-generation technologies have seenopportunities to redirect the system’s momentum.

If nothing else, this study of the Americanelectric utility system demonstrates that systemswith momentum do not carry that momentumindefinitely. Evolving systems are not, asHughes (1983) pointed out, “driverless vehiclescarrying society to destinations unknown andperhaps undesired” (462). Rather, human stake-holders play important roles in channelingmomentum. Though they may not always realizethe consequences of their actions—contributingto forces that alter momentum in unanticipatedways—people remain at the core of technologi-cal systems because of their concern for politi-cal control, influence, money, and power. Thisconclusion suggests the value of the systemsapproach for teaching about technology andsociety in school at several levels. In particular,the approach reveals the subtle and explicitinterdependency of hardware and nontechnicalelements in modern society.

Richard F. Hirsh is a professor of history andscience & technology studies at Virginia Tech inBlacksburg, Virginia.

Benjamin K. Sovacool is a PhD student in sci-ence & technology studies at Virginia Tech,Blacksburg, VA.

Acknowledgment

The authors are grateful to the National ScienceFoundation for grants DIR-9012087, SBR-9223727, SES-0112430, and ECS-0323344,which have supported elements of the workreported here. Any opinions, findings, and con-clusions or recommendations expressed in thismaterial are those of the authors and do notnecessarily reflect the views of the NationalScience Foundation.

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

83

References

American Gas Foundation. (2000). Fueling the Future: Natural Gas and New Technology for aCleaner 21st Century, at http://www.fuelingthefuture.org/contents/NaturalGasPowersUp.asp.

American Wind Energy Association. (2004). “Distributed Generation: Revolutionizing the Grid,” athttp://groups.msn.com/AAEA/utilities.msnw.

American Wind Energy Association. (2005). “Wind Energy Costs,” http://www.awea.org/faq/tutori-al/wwt_costs.html.

Capehart, Barney L., Paul Mehta, and Wayne Turner. (2003). “Distributed Generation and YourEnergy Future.” Cogeneration and Distributed Generation Journal 18(4): 17-33.

Capehart, Barney L. (2003). “Micro-turbines.” Whole Building Design Guide (August),http://www.wbdg.org/design/microturbines.php.

Casten, Thomas R. (1998). Turning Off the Heat: Why America Must Double Energy Efficiency toSave Money and Reduce Global Warming. New York: Prometheus Books.

Chandler, Alfred D. (1977). The Visible Hand: The Managerial Revolution in American Business.Cambridge, MA: Harvard University Press.

Clayton, Mark. (2004). “Solar Power Hits Suburbia.” The Christian Science Monitor (February 12),14.

Clean Energy States Alliance. (2004). “CESA Year One: A Report on Clean Energy Funds in the U.S.,2003-2004,” at http://www.cleanenergystates.org/.

Colorado Renewable Energy Society. (2002). “Renewable Portfolio Standard Backgrounder.”February 4, www.cres-energy.org/documents/ RPS%20Backgrounder_MW-2-04-02.pdf.

Congressional Budget Office. (2003). Prospects for Distributed Electricity Generation,http://www.cbo.gov/showdoc.cfm?index=4552andsequence=2.

Edison Electric Institute. (1988). EEI Pocketbook of Electric Utility Industry Statistics. Washington,DC: Edison Electric Institute.

Edison Electric Institute. (1997). EEI Pocketbook of Electric Utility Industry Statistics. Washington,DC: Edison Electric Institute.

Edison Electric Institute. (1988). Statistical Yearbook of the Electric Utility Industry/1987Washington, DC: Edison Electric Institute.

Electric Light and Power. Various issues with statistics on heat rates (thermal efficiency) of steampower plants.

Electric Power Supply Association. (2005), “Competitive Power Supply Industry Facts,”http://www.epsa.org/Competition/quick_facts_mp.cfm.

Electric Utility Consultants. (2004). The Installed Base of U.S. Distributed Generation. Greenwood,CO: EUC Publishing.

Fallek, Mark. (2004). “Distributed Generation: Electric Utility Perspective,”http://uschpa.admgt.com/Rdmap03Fallek.pdf.

Friedman, S. Julio and Thomas Homer-Dixon. (2004). “Out of the Energy Box.” Foreign Affairs 83(6)(November/December): 71-82.

Garman, David. (2005). “Diversification of Power Generation.” Hearing Before the House Committeeon Energy and Commerce (March 8). Washington, DC: Government Printing Services.

Halverson, Matthew. (2002). “Downturn in Demand Gives Contractors Time to get in the DG Game.”ECandM (February), http://www.ecmweb.com/mag/electric_downturn_demand_gives/.

Herzog, Antonio, Timothy Lipman, and Jennifer Edwards. (2001). “Renewable Energy: A ViableChoice,” Environment (December): 1-34.

Hinrichs, Doug, Susan Conbere, and Mike Lobash. (2002). “Taking Control of Power Supplies.” InBuilding Operating Management (July),http://www.findarticles.com/p/articles/mi_qa3922/is_200207/ai_n9110155.

Th

eJ

ou

rna

lo

fTe

ch

no

log

yS

tud

ies

84

Hirsh, Richard F. (1989). Technology and Transformation in the American Electric Utility Industry.New York: Cambridge University Press.

Hirsh, Richard F. and Adam H. Serchuk. (1996). “Momentum Shifts in the American Electric UtilitySystem: Catastrophic Change—Or No Change at All?” Technology and Culture 37: 280-311.

Hirsh, Richard F. (1999). Power Loss: The Origins of Deregulation and Restructuring in the AmericanElectric Utility System. Cambridge, MA: MIT Press.

Hughes, Thomas P. (1983). Networks of Power: Electrification in Western Society, 1880-1930.Baltimore: Johns Hopkins University Press.

Hughes, Thomas P. (1987). “The Evolution of Large Technological Systems.” In Wiebe E. Bijker,Hughes and Trevor J. Pinch (Eds.) The Social Construction of Technological Systems: NewDirections in the Sociology and History of Technology. Cambridge, MA: MIT Press.

Hughes, Thomas P. (1989). American Genesis: A Century of Invention and Technological Enthusiasm1870-1970. New York: Penguin Books.

Insull, Samuel. (1915). Central-Station Electric Service: Its Commercial Development and EconomicSignificance as Set Forth in the Public Addresses (1897-1914) of Samuel Insull. William EugeneKeily, (Eds.). Chicago: Privately Printed.

International Energy Agency. (2002). Distributed Generation in Liberalized Electricity Markets. Paris:International Energy Agency.

Kendrick, John W. (1961). Productivity Trends in the United States. Princeton, NJ: PrincetonUniversity Press.

Kendrick, John W. (1973). Postwar Productivity Trends in the United States, 1948-1969. New York:Columbia University Press.

Kleinbach, Paul and Henry Hinrichs. (2002). Energy: Its Use and the Environment. New York:Harcourt College Publishers.

Landsberg, H. H. and S. H. Schurr. (1968). Energy in the United States: Sources, Uses, and PolicyIssues. New York: Random House.

Lin, Jackie. (2004). “Power Outage Hits Industrial Park Hard.” Taipei Times (April 11), p. 10.

Liss, William E. (1999). GRI Technical Paper: Natural Gas Power Systems for the DistributedGeneration Market. New Orleans: Power-Gen International.

Lovins, Amory et al. (2002). Small is Profitable: The Hidden Benefits of Making Electrical Resourcesthe Right Size. Snowmass, CO: Rocky Mountain Institute.

McCraw, Thomas K. (1984). Prophets of Regulation: Charles Francis Adams, Louis D. Brandeis,James M. Landis, Alfred E. Kahn. Cambridge, MA: Belknap Press of Harvard University Press.

Melosi, Martin V. (2000). The Sanitary City: Urban Infrastructure in America from Colonial Times tothe Present. Baltimore, MD: Johns Hopkins University Press.

Meyer, Niels I. (2003). “Distributed Generation and the Problematic Deregulation of Energy Marketsin Europe.” International Journal of Sustainable Energy 23(4): 217-221.

Noble, David F. (1999). “Social Choice in Machine Design: The Case of Automatically ControlledMachine Tools.” In Donald MacKenzie and Judy Wajcman (Eds.), The Social Shaping ofTechnology. New York: Open University Press.

Ostergaard, Poul Alberg. (2003). “Heat Savings in Energy Systems With Substantial DistributedGeneration.” International Journal of Sustainable Energy 23(4): 169-176.

Pepermans, G. et al. (2003). “Distributed Generation: Definition, Benefits, and Issues.” Energy Policy(August): 21-29.

Rendon, James. (2003). “In Search of Savings, Companies Turn to the Sun.” New York Times (October12): B12.

Rigby, Peter. (2003). “Deregulation’s dysfunctional markets strike back; The blackout of ‘03 intensi-fies political and regulatory risk for U.S. transmission.” Platts Energy Business and Technology5(8): 28-35.

Th

eJ

ou

r na

lo

fT

ec

hn

olo

gy

St u

die

s

85

Sheer, Hermann. (2001). A Solar Manifesto: The Need for a Total Energy Supply and How to AchieveIt. New York: Earthscan Publications.

Silberglitt, Richard, Emile Ettedgui, and Anders Hove. (2002). “Strengthening the Grid: Effect ofHigh-Temperature Superconducting Power Technologies on Reliability, Power Transfer Capacity,and Energy Use,” at http://www.rand.org/publications/MR/MR1531/.

Smith, Rebecca. (2004). “Not Just Tilting Anymore.” Wall Street Journal (October 14), C1.

Stigler, George J. (1971). “The Theory of Economic Regulation.” Bell Journal of Economics andManagement Science 2 (Spring): 3-21.

U.S. Department of Energy, Energy Information Administration. (2000). Modeling DistributedElectricity Generation in the NEMS Buildings Models. Washington, DC: Department of Energy.

U.S. Department of Energy, Energy Information Administration. (2003) Form EIA-860 Database,Annual Electric Generator Report, at http://www.eia.doe.gov/cneaf/electricity/page/eia860.html.

U.S. Department of Energy, Energy Information Administration, Annual Energy Review 2003 (2004),at http://www.eia.doe.gov/emeu/aer/txt/ptb1306.html

U.S. Department of Energy. (2005a). Energy Information Administration Statistics,athttp://www.eia.doe.gov/emeu/mer/pdf/pages/sec9_14.pdf.

U.S. Department of Energy. (2005b). “Electric Power Annual 2003,” athttp://www.eia.doe.gov/cneaf/electricity/epa/epa.pdf.

Wiser, Ryan and Ole Langniss. (2001). “The Renewables Portfolio Standard in Texas: An EarlyAssessment.” Lawrence Berkeley National Laboratories Report LBNL-49107.

Yacobucci, Brent D. and Aimee E. Curtright. (2004). “A Hydrogen Economy and Fuel Cells: AnOverview.” CRS Report for Congress (January 14). Library of Congress: Congressional ResearchService.