The Power of Norms and the Norms of Power:
Who Governs International Electrical and Electronic Technology?
Paper prepared for the 3rd Conference on "Who Are the Global Governors?"
George Washington University, Washington DC, 15-17 November 2007
DRAFT, 4 NOVEMBER 2007
PLEASE DO NOT CITE NOR CIRCULATE WITHOUT AUTHOR'S PERMISSION
© Tim Büthe*
Duke University and UC Berkeley
*
Assistant Professor of Political Science and Associate Director, Center for European Studies, Duke University (on
leave); 2007-2009: Robert Wood Johnson Foundation Scholar in Health Policy Research, University of California,
Berkeley. Email: buthe@duke.edu; http://www.buthe.info.
For very helpful comments on a previous draft, I am grateful to Deborah Avant, Martha Finnemore, Jim Goldgeier,
Susan Sell, and participants of the "Who are the Global Governors?" conference at George Washington University.
For sharing background information, I am grateful to participants in standards negotiations in IEC Technical
Committees from various national member bodies of the IEC.
I. Introduction
"Who governs?" Robert Dahl (1961) asked in his classic pluralist study of New Haven
politics. This volume takes his question to the international level, asking a series of related
questions about the actors of "global governance." I apply Dahl's deceptively simple core
question to the governance of terminology, measures, design and performance characteristics of
electrical and electronic phenomena and products, focusing on the standards of the International
Electrotechnical Commission (IEC). Here as elsewhere, governance involves the exercise of
power and hence warrants political analysis.
There are many reasons—at least 5,075 of them, as of the end of 2006—to concern
oneself with IEC standards. Three brief examples will illustrate the point. If you take a picture
with a digital camera, you can view the image on the camera's LCD screen or send the image to a
printer, almost anywhere in the world. While resolution, clarity, and quality might differ by
manufacturer and model, the screen and printer will both recognize and produce essentially the
same image, even though the LCD screen uses a mix of red, green, and blue light to produce any
color whereas the printer uses cyan, magenta, yellow and black ink or dyes. How is this
possible? Second, when a patient needs x-ray images, s/he usually can trust that the x-ray
machine will emit a sufficiently high dose of radiation to ensure that a usable x-ray image is
taken without exposing him/her to exceptionally dangerous doses, even though neither the
patient nor the physician has measured the radiation emitted from the x-ray machine (and in
fact, neither may fully understand the technology). Why are we willing to have such trust?
Third, as recently as twenty years ago, some vacuum cleaners, hair dryers, and other motorized
electrical appliances would invariably interfere with the reception of a nearby TV or some
radios; some microwaves and cordless phone sets would interfere with each other, etc. Today, a
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
2
manufacturer of these products can, usually truthfully, give a blanket assurance that such
interference will not occur (or can be easily fixed by the consumer), even though the product in
question has not been tested for interference with any specific TVs, radios, phones, etc. How is
this possible?
The answer to all three questions is, at least in part, IEC standards. IEC 61966-series
"color management" standards define colors in such a way that it allows reliable communication
of color data between a broad range of devices with very different ways of reproducing those
colors (see IEC 2007c for more details).1 These IEC standards thus ensure interoperability. IEC
standard 60580, "Medical Electrical Equipment - Dose Area Product Meters" specifies where
and how to measure the dosage emitted by radiological devices. This allows the manufacturers
of such devices, if they implement IEC 60580, to provide technical data about their products,
which can be meaningfully compared across competing products and against regulatory
measures that often specify maximum permissible radiation levels based on the same
measurement standard. 2 Finally, IEC 61000-series standards for electromagnetic compatibility
(EMC) specify thresholds for electromagnetic disturbances that may be emitted by electronic and
electrical products and the level of insulation/immunity a product must have from disturbances in
its environment, so as to ensure non-interference even when they are operated in close proximity
to each other—crucial for pacemakers, electronic components of ABS break systems on cars or
trucks, or laptop computers running essential applications.
1
Parts of these standards originated in the strictly private "ICC" standards consortium, founded and operated for
profit by Adobe, Agfa, Apple, Kodak, Fogra, Microsoft, Sun Microsystems, and Taligent, but 6196 is a series of
IEC standards.
2
As this example illustrates, these non-governmental technical standards are often used (explicitly or implicitly) by
governments and regulatory agencies; they also are used by accredited conformity assessment bodies (such as
Underwriters' Laboratories in the U.S.), which can certify a product's compliance with a particular standard.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
3
Since all of this sounds rather technical, some observers have concluded that international
standardization is a pure science and engineering optimization problem. For Loya and Boli
(1999), for instance, international technical standardization is evidence of the triumph of
universalistic/global technical rationality over the use of political or economic power to "settle"
conflicts of interest.
There are, however, many reasons to doubt this harmonious image.
Standards are prominent non-tariff barriers to trade; where their harmonization opens markets, it
benefits more competitive firms at the expense of less competitive ones. Standards also affect
the value of patents—the root cause of Thomas Edison's ruthless multi-year campaign to keep
the U.S. from adopting alternating rather than direct current as the standard for household
electricity (e.g., McNichol 2006). And although IEC standards as such are merely prescriptive
(that is, they are explicit norms in technical language), the increasing reliance of governments on
international standards as the technical basis for regulatory measures means that they often
effectively are mandatory rules for much of the global economy. Finally, wherever prior practice
differs, standardization (i.e., harmonization) entails switching costs and thus distributional
conflicts—as shown in Büthe and Mattli's survey about ISO and IEC standardization among
firms in five countries and five industries (e.g., Mattli and Büthe 2003). Despite the many
indications that standardization is often an intensely political as much as a technical process,
however, social science analyses of standardization and standards-developing organizations are
still far and few between. The IEC in particular has received hardly any attention, even though it
is, to my knowledge, the oldest still operating institution for transnational governance in the
international political economy.3
3
IEC might like it that way; the organization is much less transparent than the International Organization for
Standardization (ISO), with which it otherwise shares many characteristics. It makes less information public about
its standards-setting process and its internal debates, is more protective of information about the historical evolution
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
4
In the century since it was founded in 1906, the IEC has grown from 11 to 68 national
electrotechnical member bodies (51 with full membership, 17 with associate membership, one
per country) and affiliates in 76 further countries (IEC 2007b).4 It has vastly expanded the scope
of its activities from the international "standardization of the nomenclature and ratings of
electrical apparatus and machinery" (1904 declaration calling for the establishment of the IEC) to
setting international standards for measurement, compatibility, performance, design, engineering
development, and safety of industrial and consumer products in "all electrotechnologies
including
electronics,
magnetics
and
electromagnetics,
electroacoustics,
multimedia,
telecommunication, and energy production and distribution, as well as associated general
disciplines" (IEC Mission Statement, 2006). As a result, there are now 5,075 IEC standards,
most of them developed anew or updated/revised within the last two decades (IEC 2007d; figures
are for 12/31/2006).
As the predominant standards developing organization (SDO) for
international technical standards for electrical, electronic, and related technologies, the IEC plays
an important part in governing the international political economy. It derives its authority from
at least two of the sources identified by Avant, Finnemore and Sell (this volume): the
concentrated expertise assembled in its technical committees and the delegation of authority
from the member states of the WTO to this transnational non-governmental organization. How
did this happen? Who defined the vastly expanded range of issues over which the IEC came to
claim governance authority? Who is the IEC and who really "governs" electrical, electronic, and
related technologies?
And why, after more than a century of the ostensible international
of its membership, and practices tight message control by only allowing written inquiries. Many representatives of
IEC member bodies, however, have been very forthcoming with information about the standardization process in
particular, as reflected in section II of this paper.
4
Associate membership provides for more limited participation in exchange for greatly reduced membership fees
for small or developing countries and countries with a stake in only a limited number of electrotechnologies. The
Philippines' membership is suspended for non-payment of membership fees.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
5
standardization of electrical technology, do we still need to carry adapters or even transformers
to plug in our razors, hairdryers, and laptop computers when we travel between the Americas,
Europe, and Asia, or even between countries within any one of these regions?
I make three arguments, based on an analysis of the increasing scope of the IEC's
authority (section I), an analysis of who exercises power within the IEC5 (and beyond) during
each of the types of governance activities identified by Avant, Finnemore and Sell—agenda
setting, rule-making, implementation, monitoring, enforcement, and adjudication—(section II),
and a brief illustration of the international (non)standardization of power plugs and
sockets/outlets (section III). Most generally, I argue that (at least in the realm of electrical,
electronic, and related technologies) the cast of actors is diverse and varies greatly depending on
the specific governance activity considered.
Even at face value, Dahls' question of "who
governs" thus does not have a simple, context-independent answer. Second, I argue that formal
and informal institutions at the international and domestic level largely shape who governs and
what means they can use to do so. Third, I argue that functionalist explanations, though they go
a long way toward answering the questions above, are at best incomplete. We need to recognize
the diversity of actors at the various stages of governance—and analyze the ways in which
institutions and "institutional complementarity" (Mattli and Büthe 2003) constrain and empower
actors—in order to understand both the explosive growth in IEC international standardization
and specific gaps in global technological harmonization.
5
As I recognize explicitly below, the IEC has an organizational interest that makes it useful at times to treat the
institution as an actor in its own right (see also Büthe 2007b; Hawkins et al. 2006; Haas 1990), but scholars of
institutionalized non-state actors in world politics should not replicate at the level of inter-/transnational
organizations the analytical sleight of hand for which many of us criticize state-centric theories of IR, namely the
reification of structures as agents, such that the individuals and groups disappear who exercise power via the
institutional structure.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
6
II. Scope of Electrotechnical Governance
II. 1. Founding the IEC: Institutionalizing Cooperation
The rapid development of electrical engineering right on the heels of the growth of the
scientific understanding of electricity and related phenomena such as electromagnetism led in the
late 19th century to the establishment of "electrotechnical societies" in most of the 20th century's
"advanced industrialized" countries (see Table 1). Usually, these were professional associations
of physicists and electrical engineers, who sought to institutionalize their information exchange.
The historical documents of the early meetings, however, also convey a sense that having a
national electrotechnical association was at the time considered an increasingly necessary part of
being a modern, industrializing country (cf. Meyer 1980).
Table 1
Founding Dates of the First National Electrotechnical Societies
1871
Institution of Electrical Engineers (IEE), UK
(first founded as Society of Telegraph Engineers)
1883
Société Internationale des Electriciens, France
1883
Elektrotechnischer Verein, Austria-Hungary
1884
American Institute of Electrical Engineers (AIEE)
1891
Canadian Electrical Association
1893
Verband Deutscher Elektrotechniker (VDE), Germany
1897
Associazione Elettrotecnica Italiana
The push for electrotechnical standardization—at the domestic and at the international
level—arose largely out of these societies, often initially motivated by a desire to have common
measures in order to be able to replicate and build on each other's research, but increasingly also
out of commercial interest, as the development of electrical technology and machinery proceeded
at breakneck speed. This demand for international standardization led to a formal declaration at
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
7
the 1904 International Electrical Congress in St. Louis, calling on the "technical societies of the
world" to come together and create a "representative commission" to "consider questions of
standardization …," which in turn led to the founding of the IEC by representatives from
Austria-Hungary, Belgium, Canada, France, Germany, Holland, Italy, Spain, Switzerland, the
UK and the U.S., meeting 26-28 June 1906 in London. Why did they create a transnational
organization to deal with electrical and electrotechncical standardization—and who were "they?"
An important part of the explanation for the creation of the IEC and the initial scope of its
authority is functional. There was a wide-spread belief that, for basic research and technological
development to be fruitful and cumulative, common terminology, measures, and symbols were
needed not just at the national but, given the rapidly increasing transnational flow of ideas and
products, at the international level, too (Hughes 1983:esp.47ff, 79ff, 140ff) (see also O'Rourke
and Williamson 1999; Rodgers 1998). Market selection can create de facto standards (VHS
video format and the MS windows operating system for PCs are prominent examples), but it may
be unsuitable to achieve harmonization of basic terminology, measures, etc.6 More importantly,
as an empirical matter, most of the participants of the early international electrical congresses
saw that harmonization was simply not happening: The exhibits that occupied the "Palace of
Electricity" at the 1904 World's Fair in St. Louis, for instance, required not only electricity of
numerous different voltages, but differed in whether they needed direct current, or 1-, 2-, or 3phase alternating current, with frequency ranging from 25 to 60 cycles (see also AIEE 1904).7
Moreover, the slow, limited success of attempts to achieve harmonization informally in the
6
Notwithstanding my argument that institutionalized standards-setting is a political rather than just a scientific or
engineering optimization process, there is no evidence to suggest that market-driven de facto standardization leads
to superior standards and it often fails to lead to broad-based harmonization at all. There is, however, no indication
that these issues were considered by the proponents and participants of international standardization at this time.
7
See Erdmann 2007 for details.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
8
context of the international Electrical Congresses, which took place at irregular intervals,
suggested that ad hoc cooperation was insufficient: While the delegates of the first International
Electrical Congress in Paris in 1881 adopted Ampere, Volt, and Ohm as common "practical
units" from among no fewer than 10 different units of electric current, 12 different units of
electromotive force, and 15 different units of resistance previously in use, it took another twelve
years to agree on common definitions of these units at the 4th international congress in Chicago
(IEC 2006a:1). Institutionalizing this kind of cooperation by giving it some formal institutional
structure was hoped to facilitate and speed up standardization on a global level.
As noted in the introduction, however, harmonization of previous divergent standards
creates winners and losers. Devoid of actors, the functional explanation for the creation of the
IEC hides this distributional effect.
But who actually founded the IEC—to carry out
standardization in whose interest?
The official delegates at the 1904 Electrical Congress and the 1906 founding meeting in
London were mostly private individuals, often highly respected technical experts in physics or
engineering at universities or polytechnic institutes but also with an important role in industrial
application and usually selected by their national electrotechnical societies. It was this group
that initially defined the general issue area over which the IEC was to acquire global governance
authority. While scientific considerations and the desire for cumulative research contributions to
the understanding of electricity and related phenomena played an important role, commercial
interests were the driving force behind electrotechnical standardization at the national and
international level. Inventors and commercial developers of electric technology and machinery
realized in particular that the lack of standardization of such basics as the type of current (direct
vs. alternating) or even the way in which voltage was measured impeded their ability to achieve
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
9
economies of scale in the production of anything from incandescent light bulbs and telephones to
electrically powered machinery. As one of the participants noted at the time, a difference of
"1/10th of a Volt" in 110 Volts could make a difference of "large sums of money in regard to a
contract for incandescent lamps" (Erdmann 2007:4). Consequently, usefulness for technological
development was a key criterion for decisions of where to focus standardization efforts,8 and
protecting or enhancing the value of one's technology and patents often drove support and
opposition to the standardization of any particular standard.
The rapid development of
electrotechnology at the time, however, ensured that standardization would usually open many
more new, profitable opportunities than it foreclosed. And IEC standardization—carried out
through correspondence and regular plenary sessions every 2-3 years—was often close behind
the cutting edge of technological development, allowing the development of international
standards often before national ones had been firmly established and before a great many had
developed a stake in a particular standard or practice, which can impede shifting to a new
standard even if the new one is clearly superior (see David 1985). Drawn-out battles were
therefore rare.
Equally importantly, who was not sitting at the table?
Some governments saw
electrotechnology early on as holding great promise for economic and industrial development.
The Prussian/German government, for instance, set up the extensive basic research and
standardization facilities for the physical sciences at the Physikalisch-Technische Reichsanstalt,
founded in 1887, as an incubator of scientific knowledge to supercharge Germany's electricity-
8
Since the scientists and engineers who were the individual actors in the technical discussions in the early years
often had a personal financial stake in the commercial applications, the distinction between purely scientific
considerations and commercial interests should not be overdrawn.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
10
driven late industrialization (Cahan 1989; Hughes 1983:178) (Warburg 1916).9 Some of the
leading figures in national electrotechnical societies also had appointments in public universities
or as lead scientists in publicly funded research institutes—though they may have acted there as
much as representatives of industry—in which they often also had a stake—as they acted as
representatives of the state.10 Governments as such generally played little or no role in the IEC
(see below). More important for the question of who governs, however, were other exclusions.
Reflecting the prowess and social status afforded to many of the pioneers of the "electrical age,"
the founding meeting of the IEC in June 1906 was held in London's premier luxury hotel, the
Hotel Cecil, assuring in effect (if not necessarily intentionally) the exclusion of stakeholders
from poorer parts of the world.11 The organizational features of the IEC (some of them arguably
inevitable, given the technology at the time) also meant that only those able and willing to pay
the costs of international travel would be able to participate fully in IEC standardization. This in
turn affected who the national member bodies would appoint to be the country's official
representative on any given IEC technical committee (once they were formed). To what extent
those delegates would represent a broader set of interests at the international level depended
largely upon the institutions at the domestic level, i.e. the structure, rules, and norms of each
national electrotechnical society. Yet, the material costs of participation clearly created a bias in
9
Countries and empires since ancient times, such as the Babylonian and Roman empires, had thrived on trade
enabled by the availability of common "weights and measures," and economic historian had begun to understand
and advertise this in the late 19th and early 20th century in the context of the intellectual ascendancy of models of free
trade. Governments therefore supported and encouraged the development of "basic" standards (such as the U.S.
National Bureau of Standards (today NIST), founded in 1901—secure in the knowledge that protection of industry
was still available at all times and more profitably achievable (for the government's coffers) by raising tariffs than by
fragmenting markets through differing standards.
10
Even the German Reichsanstalt owed its existence and success at least as much to Werner von Siemens as "first
man of German industry" as to Hermann von Helmholtz as "first man of German science" (Cahan 1989).
11
It is not clear whether the exclusion of what we would today call developing countries would even have occurred
to many observers at the time, given the prevailing ideology of colonialism. Notice, however, that even from among
the many already independent countries of the Americas, only the US, Canada, and (rich and highly developed)
Argentina were represented in the IEC initially (and Argentina not even at the 1906 meeting).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
11
favor of commercially successful stakeholders from rich countries, though a neat distinction
between scientific and commercial interests is nearly impossible for this early history of electrotechnical standardization, since most of the key participants personally had both of these
interests.12
In sum, IEC in the early years focused on measurement and nomenclature, establishing
electrical and magnetic units such as Hertz and Gauss, as well as developing a unified
"International Electrotechnical Vocabulary" (now the 13-language IEC Multilingual Dictionary,
first published in 1938). These were essential building blocks in the industrialization process,
the formation of international markets, and the development—one might say the social
construction—of electrical engineering as a universal (global) scientific profession (see Ruppert
1956:4f) (see also Herrigel 1996). IEC standardization in the early years thus had public goods
characteristic with the promise of substantial benefits for most participants (Kindleberger
1983)—and was effectively limited to a small set of countries whose technical and socioeconomic elites shared a common understanding of modern science. Moreover, those who
sought electrotechnical standardization had learned from the difficulties of establishing common
basic measures of length and volume in the 19th century. That experience suggested that the
adoption of another country's or ruler's standard generates far more resistance than the adoption
of a standard derived by a transnational body of experts from "universal" scientific principles or
12
Elihu Thomson, for instance, the President of the U.S. organizing committee for the 1904 International Electrical
Congress that called for the establishment of the IEC, was scientist, inventor—and founder of General Electric.
Lord Kelvin, pathbreaking Irish/British scholar of thermodynamics and the IEC's first president, held 70 patents,
including for the adjustable compass and several of the technologies used in the first transatlantic cables, though he
was not directly involved in their commercial exploitation. Eugen Wüster, the Austrian founder of terminological
lexicography and driving force behind electrotechnical terminological standardization for several decades beginning
in the 1920s, started out as a successful industrialist.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
12
from nature (e.g. Trapp and Wallerus 2006:28).13 From the start, the IEC therefore adopted a
strong norm that arguments over the specific provisions of a standard or even its desirability had
to have a scientific rationale (that would be open to investigation) and be presented in terms of
scientific or engineering optimization.14
II.2. Expanding Scope of IEC Governance, 1906 - 2006
While terminology, symbols, and measurement remain important areas of its
standardization work, the IEC successively broadened its (claimed) global governance authority
over the years. In the 1920s, the IEC added developing standards for consumer products to its
portfolio, starting with the (US-chaired) technical committee (TC6) for "lamp sockets and caps."
The IEC's committee structure expanded as technology changed, with committees for computing
and information processing standards added in the 1960s, laser equipment in 1970s, fiber optics,
superconductivity, and wind turbines in the 1980s, fuel cell technology in the 1990s, etc.
In addition to broadening, the authority the IEC as a global governor has deepened in
recent years.
Probably most important for this development has been the Agreement on
Technical Barriers to Trade (TBT-Agreement), which is an integral part of the Uruguay Round
Final Act and hence binding on all WTO member states. In this treaty, the member states
committed themselves and their subunits, effective in 1994, to use international standards as the
technical basis for laws and regulations whenever international standards exist that can achieve
the legitimate objectives of such laws and regulations, such as health, safety, or consumer
protection. In an implicit (and some would argue explicit) act of delegation, the IEC was one of
13
Thus, the meter, which was defined as 1/40,000,000 of the (assumed to be constant) circumference of the earth,
was far more palatable as a base metric for length for the kingdoms and principalities of Europe in the 19th century
than some neighboring ruler's (ancestor's) foot…
14
These norms, to be sure, were grounded in the broader professional norms of the physicists and engineers that
populated the organization.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
13
only three standards-developing organizations recognized in the text of the Agreement as a
source of such standards (Marceau and Trachtman 2002).15 Partly as a consequence, many
developing countries (including large, fast-growing ones like Brazil and India) often simply
adopt IEC standards as domestic standards and require conformity with current IEC standards for
market access. Moreover, some newly industrialized countries, such as South Korea, have
refrained from setting up a full-fledged standards-development process at the national level.
Instead, their national IEC member bodies focus on organizing domestic input into the
international standardization process. Similarly, some smaller advanced industrialized countries
essentially abandoned national-level standards in the 1990s to concentrate their energies on the
regional and international standards-setting processes, allowing Sweden, for instance, to be a
very important player within the collective global governor, IEC. But even for Germany's DKE,
traditionally one of the biggest sources of national electrotechnical standards, domestic
standards-setting had shrunk to 5% of its standards-setting work by 2004 (DKE 2004). Finally,
rapid technological development and increasing specialization in recent decades led to the shift
of most technical work from broad technical committees to more specialized sub-committees,
each with a secretariat (with resources and staff volunteered by a national member body) and a
chairman, by custom from a different country (for details see IEC 2006a, 2006b).
What explains this evolution of the IEC and specifically the broadening of the range of
issues over which it exercises a global governance role?16 Technological development is clearly
part of the story and offers again a functional explanation. I propose, however, to think of
technological change as merely a necessary condition: Setting up a committee to develop
15
This legal recognition is non-exclusive, but safeguards the IEC against the controversies encountered by some
other (mostly US-based) SDOs that have sought to be considered international standards-setters under the terms of
the TBT-Agreement.
16
The specific ways in which IEC and other actors "govern" via IEC standards are discussed in the next section.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
14
technical standards that safeguard against magnetic interference from the operation of electrical
devices would have made little sense until such interference occurred; a technical committee on
fiber optics would have been literally unthinkable until after fiber optics had been invented.
Technological developments created the demand for governance.
But as noted by Avant,
Finnemore, and Sell (this volume), the demand for governance is rarely universal (some would
likely prefer the continued absence of governance), and even if it is widespread, demand for
governance does not automatically lead to its provision. We therefore must explain why some
actors who seek inter-/transnational governance win out over those who oppose it—and supply
as well as demand for governance must be explained (Mattli 1999: esp.41ff).
I submit that, at a general level, the institutional structure of the IEC provides an
explanation both for the predominance of those who seek global governance and for the supply
of such governance. Transnational or global technical standardization should always be in the
interest of those who are competitive producers and would benefit from increased market size.
They are therefore likely to push for the IEC to expand its issue space to cover the technical issue
that impedes their global market access by requesting that their national IEC committee propose
new/revised standards or the establishment of new technical committees or sub-committees.
Moreover, whenever there are network effects or other incentives to settle on a single standard in
the end, standards-setting resembles a non-symmetric coordination ("Battle-of-the-sexes") game,
shown in Figure 1 with ordinal payoffs.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
15
Player Column: Adopt Standard …
Player Row:
Adopt Standard …
X
Y
3
X
4
2
2
1
Y
1
4
3
Figu re 1
Co or dinat io n Ga me w| Distr ibut iona l Co nflict
Note: Ordinal payoffs; 2 equilibria: XX, YY.
Under these conditions, those who have the technical expertise to develop a particular
standard (i.e. make the rules and in that sense supply governance) should be willing to do so as
long as the difference between the payoffs a given "player" would obtain at the possible
coordination equilibria (i.e., in Figure 1, the difference between obtaining ordinal payoff 3 in the
one equilibrium and obtaining 4 in the other) exceeds the costs of supplying governance,
conditional on the likelihood of achieving coordination on one's preferred standard.
Those who oppose such a broadening of the IEC's issue scope have a number of options,
but none of them is particularly promising given the institutional structure of the IEC. They can,
to be sure, organize at the national level to keep the national IEC committee from submitting
such a proposal to the IEC. Such political opposition may well succeed in any given country, but
if those who seek to expand the realm of international standardization can make such proposals
in numerous countries, then opponents have to defeat these proposals domestically in each
country to keep it from coming up in the IEC—increasingly unlikely as ever more industries
have become every more internationally oriented (Milner 1988).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
16
Once the proposal is on the table at the IEC, forestalling the extension of IEC authority
requires either building a coalition of national IEC committees opposed to IEC governance or
challenging IEC governance in the given realm by establishing competing standards through
market mechanisms or through alternative institutional mechanisms. None of these possible
courses of action is likely to succeed. While it is certainly possible to build a transnational
political coalitions of domestic actors opposed to transnational politics (Bob, this volume), doing
so in the IEC would require convincing technical experts who by virtue of being representatives
on IEC technical committees devote substantial amounts of time to providing global governance
that global governance is a bad thing, at least in the issue area at hand.17 Given the broad-based
membership of the IEC, challenging it by seeking competing standards through other
organizations is unlikely to succeed, given that creating inter-/transnational institutions is
difficult and costly, that a number of other IEC participants would also have to participate for
such competing standards to be a serious challenge to IEC standards, and given that the IEC—as
an actor interested in its own perpetuation—has struck cooperation and coordination agreements
with many other organizations, such as the ISO (joint technical committee 1 and its by now 36
sub-committees). Alternatively, opponents can try to challenge IEC standardization by getting
competing standards established through the market mechanism.
In particular so-called
standards consortia, collaborative ventures of several firms in a given industry, which are usually
created to reserve for the participating firms the intellectual property rights to the innovations
that arise in the process of developing a particular (set of) standard(s), can be quite successful in
establishing de facto standards, sometimes leaving the IEC with little choice but to endorse as an
17
If the argument against IEC standardization were based on specific characteristics of the standard likely to be
adopted then the same coalition could presumably ensure the adoption of a different/better standard, so technical
arguments are unlikely to be compelling to keep the IEC from developing a standard as such, though they certainly
can be successful in changing the specific provisions of that standard.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
17
international standard a technical specification for which some players already hold a patent.
But while this may work for any given product, it seems unlikely to succeed for an entire issue
area.18 This leaves opponents with the final option of blocking adoption/implementation at the
domestic level. As discussed below, this strategy can be lastingly successful, but it does not stop
IEC governance for an issue area as such.
The institutional structure of the IEC and of
international standards-setting thus favors the expansion of the scope of IEC governance by
empowering those who (for usually entirely self-interested reasons) seek such an expansion (see
also Büthe 2007a).
II.3. The IEC as a Non-Governmental Organization
IEC has its origin in the decision by the organizing committee for the sixth International
Electrical Congress to invite governments to appoint official delegates for a special meeting at
the 1904 Congress. Those delegates in turn passed the resolution calling on the technical
societies of the world to come together in the "appointment of a representative commission to
consider the question of the standardization of the nomenclature and ratings of electrical
apparatus and machinery," which led to the 1906 founding of the IEC as sketched above. While
the resolution had no legal force and in that sense was simply an assertion of authority, the
governmental authorization of the delegates lent it some added legitimacy. Yet, despite this
initial seeking of governmental appointment, the IEC was founded and has resolutely remained a
non-governmental organization.19 So as to have no "bureaucratic influence imported into the
Commission" (IEC 1906:10), national technical societies (i.e., professional societies of electrical
18
Moreover, this course of action would only keep the development of the international standard out of the IEC; it
would still lead to the development of an international standard and thus offer little to uncompetitive producers
which rely upon standards-induced fragmentation of markets to remain in business.
19
Note, however, that member bodies may be public agencies, hybrid public-private entities, or private bodies with
public financial support and regulatory oversight.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
18
engineers, with input from industry as each saw fit) rather than governments were to constitute
each country's "local [IEC] committee" and appoint the national representatives to the IEC.20
Governments were to be asked to appoint a local committee only in countries "having no
Electrotechnical Institution," and those countries' technical societies, if subsequently founded,
could appoint a new local committee once they had been in existence for at least three years (IEC
1906:18, 20). These national "local" committees constitute the IEC's member bodies to this day
(see IEC 2007a). Moreover, the founding delegates agreed that "every country joining the
Commission [should be placed] on an absolutely equal footing" (IEC 1906:10, 34), with one vote
per country whenever decisions were to be taken by vote. A central office staff (financed by
similarly equal contributions from by each national member organization) was to coordinate
information flow and standardization activities.
Why did the founders put such emphasis on IEC being a non-governmental organization,
which has persisted to this day? Institutional mimicry may have played some role, as the (nongovernmental) British Institution of Electrical Engineers served as the explicit inspiration for the
IEC. But the commitment to the non-governmental character of the IEC had broad support,
including from many continental European delegates, because it could serve quite diverse
interests. For electrical engineers from Britain and the U.S., self-regulation was among the
hallmarks of the traditional liberal profession (law and medicine), which they (along with many
other white-collar occupations) sought to emulate (Brint 1994:5).21 In continental Europe with
its different political traditions22 and lack of the liberal Anglo-Saxon notion of the "profession"
(Jarausch 1990), the explicit recognition of autonomy and self-government in professional and
20
The U.S., today represented by ANSI, was represented initially by the American Institute of Electrical Engineers.
And engineering was just starting to become a distinct occupation at the time, see Gispen (1990).
22
See, however, Philip Nord's (1995) historical analysis of France.
21
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
19
industry associations (even if sometimes more apparent than real) compensated for the lack of
democracy in political life (Blackbourn and Evans 1991; Sciulli 2005).
The sometimes adamantly non-governmental character of the IEC has two important
consequences 'til this day. First, it has restricted the fungibility of power resources. Whereas
economically or militarily powerful states have in international governmental organizations, such
as ITU and WTO, often been able to use seemingly unrelated power resources to influence
outcomes—allowing economic interests from these country to sometimes dominate the
preferences of smaller/weaker states (through issue linkage and other means, see e.g., Davis
2004; Krasner 1991; Steinberg 2002)—attempts at direct governmental interference in IEC
standardization are considered illegitimate and rarely even tried (Mattli and Büthe 2003). Here,
norms really have power: Consensus procedures in the standards-setting process guarantee that
all arguments for or against specific provisions of a standard or even its desirability have to be
considered and accommodated, if possible—except when they fail to provide a scientific or
engineering rationale that can be investigated, in which case the technical committee may simply
dismiss them for lack of a "technical basis."23 Second and relatedly, the non-governmental
nature of the IEC has caused the organization to operate "under the radar" of IPE scholars that
take a statist analytical approach to questions of global governance (e.g., Drezner 2007). Its
important role in global governance has therefore been overlooked.
23
None of this is to say that all countries are equal in IEC standardization, notwithstanding the one-country-one-vote
rule, which has so far only half-heartedly and unsuccessfully been challenged, most notably by the U.S. Countries
with greater scientific and engineering expertise and an ability and willingness to devote more of those resources to
the IEC standardization process clearly are in a better position to influence IEC standards than poor, developing
countries, ceteris paribus, see Büthe and Witte (2004).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
20
III. IEC Standards as Instruments of Governance
Standards are (written) norms. An IEC standard (e.g. IEC 61000) specifies in often
highly technical language the characteristics that a phenomenon or device should exhibit in order
to be considered a device of the particular type that is defined through the standard (e.g. an
electromagnetically compatible TV or vacuum cleaner).
Standards as such are "merely"
prescriptive; they are not binding rules. They become instruments of governance only if those
whose behavior they target have legal, material, normative, or other incentives to adopt,
implement, or comply with the standards. To understand the role of IEC standards in global
governance and properly answer the question of who governs, we must consider five stages in
what I call the governance sequence (see also Avant, Finnemore, and Sell, this volume): first,
agenda setting in the IEC, focused on the decision whether an international standard for a
particular product or process should be developed; second, rule making, that is the process of
developing any specific IEC standard; third, the adoption or implementation of the IEC standard
by manufacturers and other potential users; fourth, the monitoring of compliance with the IEC
standard; fifth, enforcement of compliance in cases where compliance is mandatory.24 For each
of these activities, I seek to identify the key groups with a stake in IEC governance, their
motivations and resources, and why they do or do not play a role.
III.1. Agenda Setting in the IEC
Proposals for the IEC to begin work on developing a standard can be made only by an
IEC member. The members, as noted above, are the "national committees" of the IEC, i.e., the
24
I do not here consider adjudication, which takes place, with respect to IEC standards, at three levels. The IEC's
SMB (described below) adjudicates between competing claims of authority between different technical committees.
Second, the IEC has set up various arrangements with other international organizations with which a conflict of
responsibility might arise (ISO, ITU, etc.). Third, national and supranational judicial institutions, such as the WTO
Dispute Settlement Mechanism, deal with conflicts of interest over the (ab)use of IEC vs. differing domestic
standards (e.g. as non-tariff barriers to trade).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
21
dues-paying, national-level electrotechnical societies or official bodies (see IEC 2007b). How
any given national IEC committee decides whether or not to make a proposal to the IEC if any of
its domestic constituents (scientists, firms, government regulators, consumer advocates, etc.)
requests that it do so, depends upon the decisionmaking procedures of each national committee,
that is, on the domestic institutions.
If it is a proposal to revise and existing standard, the proposal is made in the technical
committee with responsibility for the standard to be changed; the members of the committee,
only, decided whether to act on the proposal. If the proposal is for a new standard, the IEC
central office circulates it among all the members of the IEC for discussion and a vote. This is
the main agenda-setting stage, which should be analytically distinguished from the standardsdeveloping process (below). At this point, the member organizations decide "only" whether to
put the development of an IEC standard for a particular issue/product/part on the agenda of work
to be done in the near future, often without yet having at hand any specific ideas regarding the
content of that potential standard.
If the national member bodies vote in favor of developing an IEC standard for a given
issue or product, then an important next step in IEC governance is carried out by a special
committee of the organization, the "Standardization Management Board" (SMB). The SMB
today consists of 15 members and a non-voting chairman. 25 Six of the 15 seats on the SMB are
reserved for the six member bodies that make the largest contributions to the IEC budget and
provide the staff support for the largest number of technical committee secretariats.
The
remaining nine seats are filled through elections in the assembly of the member body presidents
25
The IEC general secretary also sits on the committee, ex officio (i.e. s/he has no voting power).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
22
(the "Council") for three-year terms (IEC 2005, Art.10). 26 The SMB wields considerable power,
coordinating the work of 178 IEC technical committees and subcommittees (TCs) with their 449
working groups and the 1357 standards projects underway, most importantly by appointing TC
secretariats and chairmanships, assigning individual standards projects to particular TCs, and
adjudicating jurisdictional conflicts between the TCs (IEC 2005, 2007d).27
In sum, at the agenda-setting stage, the key actors are the national member bodies of the
IEC (particularly those active in the TCs), with special powers reserved for the six largest
contributors to the IEC's work and the nine elected members of the SMB. Within theses national
IEC member bodies, the primary actors tend to be firms, especially manufacturers of electrical
and electronic products and energy/power suppliers, and secondarily electrical/electronics
engineers as a profession. A full analysis of who exercises how much influence through the
member bodies is beyond the scope of this chapter, since they differ considerably in their legal
status, organizational structure, and voting rules. But since agenda-setting is conducted via the
IEC member bodies, it is clear that domestic institutions play an important role in this realm of
global governance.
III.2. Rule Making: The Development of an IEC Standard
The actual technical work of standards setting is carried out in IEC's technical
committees or subcommittees (TCs) and their "working groups" and "project teams." These
committees and groups are staffed by (often highly specialized) technical experts, mostly from
industry but also from government regulatory agencies, academia, and research laboratories.
26
For the six appointed seats, budget contributions and committee secretariats are each calculated as a percentage of
the total and the percentages are added together. Should this lead to a tie for the sixth seat, committee secretariats
are given greater weight (IEC 2005, Appendix 1). For the election of the remaining nine seats, the Rules of
Procedure stipulate that member bodies aim for a "balanced geographical distribution" among qualified candidates.
27
Number of projects underway is for 12/31/2006.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
23
IEC also encourages the participation of labor union representatives and consumer groups,
though they constitute only a small portion of the ca. 10,000 technical experts directly involved
in IEC standards setting worldwide.28 The participating experts (or their public or private sector
employers) volunteer their time, and the resulting technical documents become the common
property of the IEC. As far as the rule-making function of global governance is concerned, it is
these technical experts who collectively govern.
Individual technical experts, however, cannot claim this power on their own.
Participation in a technical committee (TC) is limited to those nominated for such a role by their
respective national member bodies of the IEC (each of which may nominate several individuals
to a given committee, albeit with only one vote collectively). Accordingly, the experts on any
TC are to represent the interests of their country's stakeholders as a whole, as established (if so)
through the national institutions. Every full member body has the right to nominate participants
with voting rights to every committee (which makes the member body a "P-member" of that
committee); associate members can become P-members of up to four technical committees only
(though they may send technical experts without committee-level voting rights to any number of
additional committees).
Moreover, standards setting takes place in five stages, as illustrated in Figure 2 and
discussed in greater detail in Büthe and Witte (2004:37ff). After the "Planning" or "Proposal"
stage, during which preliminary, often informal discussions within and among member bodies
leads to the submission of a proposal to develop or revise a standard as discussed above, a first
working draft of the new standard is developed during the Preparatory Stage (if the initial
proposal is successful). This draft takes into account comments and specific submissions from
28
Many member bodies create "shadow committees" at the domestic level to coordinate input from stakeholders in
their respective countries.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
24
the national member bodies represented on the TC that is charged with developing the standard,
but is usually drawn up by the leader of a "project team" within that committee. The work then
moves to the committee level, where a full-fledged draft standard is developed with participation
from all P-members of the TC during the third "Committee" stage (the standard usually goes
through several drafts during this stage, with increasing specificity).
During the fourth
"Enquiry" stage, the bilingual Committee Draft is circulated among all of the national member
bodies (regardless of committee representation) for comment and an initial vote.
If the
Committee Draft passes this vote, the TC may proceed to revising the standard one last time to
take into account technical comments that may have been submitted along with or in lieu of a
vote. The resulting revised standard is then during the "Approval" stage circulated among all
national member bodies as a Final Draft International Standards (FDIS) for a two-months up or
down vote (member bodies may vote "abstention"). If approved, the standard is then published
by the central office within two months.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
25
1
2
3
4
5
Figure 2
5-Stage IEC Standardization Process
Since the specificity of the standard increases as it moves through the five-stage process,
early involvement in the process is key to exercising influence (Mattli and Büthe 2003; Büthe
and Witte 2004). This makes national member bodies that are involved at the TC level—and
able to formulate a coherent position on behalf of their domestic constituencies in a timely
fashion—the key actors for the rule-making process. That said, the Enquiry Stage procedures
(and to a lesser extent the final Approval Stage vote) ensure that stakeholders from a broader
spectrum of countries will also have an opportunity to make their voice heard: For both stages
that involve formal voting, every member body has one vote, effectively retaining the onecountry-one-vote system that was set up in 1906. At both stages, moving forward requires that
two-thirds of the P-members of the committee that developed the standard vote in favor of the
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
26
draft and that no more than 25% of all vote-casting member bodies vote against it. At either
stage, if the draft or final draft does not gather the required level of support, it is referred back to
the TC, which may either drop the standardization project or re-start the process from stage 2 or
3.
Consensus procedures require (at stages 2 trough 4) that all criticisms, objections, or
suggestions for improvement of a standard are taken up for substantive discussion and
accommodated if possible, though only if underpinned by specific technical arguments. This
procedure ensures that sometimes intense conflicts of political-economic interest are blunted, if
only in appearance, because they are carried out in terms of scientific/engineering optimization
and highly technical language (see Mattli and Büthe 2003).
III.3. Adoption—Implementation—Compliance
As noted in the introduction, a standard as such is merely an explicit norm. Setting
standards by no means guarantees that those whose behavior the norm seeks to influence will in
fact adopt or implement the standard, i.e. act accordingly. And the actors at the implementation
stage differ from those during the two prior stages except in an ideal-typical case of selfregulation.
Reasons for compliance vary for technical standards, just as for other norms. Electrical
engineers use the standard international nomenclature and units of measurement, developed by
the IEC in its early decades—Ampere, Volt, Hertz, Ohm, etc.—in part simply because they are
practical: The internal logic and the relationship between the basic units used in the IEC
definitions, along with the reliance on the decimal system, allow simple calculations of various
properties of electric phenomena and electronic devices (see Figure 3). But even more important
are the positive network externalities obtainable from using these standards: since they are
familiar and have an agreed-upon definition around the globe, using them facilitates basic and
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
27
applied scientific communication, ensures that product specifications and information about
energy use, interoperability, etc. will be easily understood by suppliers and customers, and thus
stimulates commerce.
Figure 3
International System of Units, incl. Electro-technical Units
Note: Source: NIST, "Relationships Among the SI Units", http://physics.nist.gov/cuu/Units/SIdiagram.html
The flipside is that standards (and technologies) with strong network effects are hard to
dislodge even by technically superior (more rational or more efficient) ones—the persistence of
an institution may have very different causes from its creation (Büthe 2002; 2006; 2007a;
Pierson 2000; 2004; Thelen 2000). For the same reason, if prior standards differ, harmonization
on a single standards entails switching costs at least for some, which may be sufficiently high to
ensure that standard nomenclatures or technologies are not adopted—as illustrated by decades of
resistance to the introduction of metric units for distance, volume, and weight in the United
States.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
28
Essentially the same logic drives decisions about the adoption/implementation of
electrotechnical standards by firms. Positive network externalities and the technical quality of
standardized solutions to a practical problem create incentives for implementing IEC standards in
one's production processes; switching costs create disincentives. But for commercial actors,
additional competitive and political considerations come into play. Consumer demand may
create pressure for implementation;29 conversely, if deviating from a standard provides an
opportunity for product differentiation, producers may consciously decide not to implement the
standard.30 Divergent producer-specific standards, especially when they are proprietary, also
provide an opportunity to create a captive consumer base: Having unique shapes and placements
of connectors between battery and cell phone, for instance, is what allows cell phone service
providers to sell replacement batteries at a substantial markup over technically equivalent, but
not interoperable batteries.
In the implementation decisions discussed so far, the actors are economic agents such as
firms, each of which is acts independently, even though network effects mean that many are
faced with the decision to implement even if they had no involvement at the agenda-setting or
standards-setting stages.
29
Since many IEC standards are primarily used for industrial goods sold from one firm to another, this makes those
who make decisions about inputs, such as purchasing managers, key actors in this stage of the standards governance
process.
30
Such non-compliance with international standards for product differentiation purposes may—but need not—lead
to lower performance: At one extreme is quality-reducig non-compliance in order to cut production costs. While
the IEC urges technical committees to specify in international standards the most cost-effective way to address a
given problem (if they specify a particular solution at all, rather than simply specifying performance requirements
that any given manufacturer's solution must meet), IEC standards generally reflect what is technologically feasible
to meet actual or imagined consumer demand in advanced industrialized countries. As a consequence, an electric
appliance with metal exterior parts may have double-insulation, auto-shut-off to guard against overheating, and
numerous other safety features—which make producing it so expensive as to price it out of the market of many
developing countries. Producing for these markets may then require using less stringent standards. At the other
extreme, the invention of a new, superior technical solution to a product safety problem might be incompatible with
the design specifications of an existing international standard; a firm may nonetheless use it precisely to be able to
say that its products are even better than those manufactured to the current international standard.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
29
Instead, local or national governments may be the key actors at this stage if they decide to
incorporate IEC standards directly into laws or regulations, or if they reference IEC standards in
those legal documents.31 Strictly speaking, economic actors (most often: firms) then still face the
individual decision whether to implement the stipulations of the standard, but compliance is no
longer voluntary but required. This in turn increases the incentives for firms not to remain
atomistic actors but to organize—via industry organizations, issue-specific lobby groups, or via
the domestic standards developing organization that is the country's representative in the IEC—
to oppose such regulatory measures, if they do not like the content of the international standard,
or possibly to ask that governments require compliance with the given standard, so as to level the
playing field (Vogel 1995). And as the literature on the faces of power reminds us, politicians'
responsiveness to the preferences of key social or economic groups might not require explicit
demands but may take place in anticipation.
A final motivation for adoption/implementation is firms' anticipation of the consequences
of non-compliance in the event of civil litigation, most importantly product liability lawsuits.
Since these standards often define "best practice" implementing them often provides one of the
few available safeguard against being sued (or against being in a weak position if sued). Firms
may therefore try to make sure that their products comply with the applicable IEC standards even
if they do not advertise that fact or seek certification for it. This is in fact a key reason why
larger manufacturing firms often have a standards manager.
31
The IEC now has a page on its website to guide government agencies in referencing IEC standards, so as to
facilitate such referencing in such a way as to make it compatible with continuous updating of the standard as
technology develops rather than having governments reference a specific (and possibly soon outdated) version of the
standard. Several IEC standards in recent years have been developed because the European Commission has asked
the European regional standards developing organization for electrotechnical standards, CENELEC, to develop a
standard for a given product or issue so as to allow the EU to reference it in a directive or regulation, prompting
CENELEC to request a joint IEC-CENELEC standardization process, so as to have standardization take place at the
broadest possible level. In such cases, governments or the EU already play a leading role at the agenda-setting
stage, though it is of course possible that these are cases of industry-capture.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
30
III.4. Monitoring
As the IEC became increasingly involved in setting standards for industrial and consumer
products in the post-WWII period, the question of how to safeguard the value of its trademark
(i.e., the value of firms being able to say that their product is manufactured to any particular IEC
standard) became more prominent. As the IEC website puts it: "How can the industrial user and
the final consumer be sure that the product they buy conforms to the criteria of an IEC
standard?" (IEC 2007e). Conformity Assessment has been the IEC's answer. IEC conformity
assessment operates largely like a certification scheme (Cashore, Auld, and Newsom 2004;
Meidinger, Elliott, and Oesten 2003; Prakash and Potoski 2006):
Producers submit their
products to a conformity assessment body, which uses a series of IEC-standard-specific but
otherwise general procedures to test it. If the product conforms to the specifications of the
particular standard, the conformity assessment body certifies this finding and grants the producer
a license to use its name in conjunction with statements about the product's compliance with that
IEC standard.
National member bodies of the IEC can either conduct conformity assessment themselves
or accredit independent laboratories within their respective countries to carry out IEC conformity
assessment.32 In either case, the IEC does not itself accredit the conformity assessors, but instead
oversees a system of "peer assessment" among the different national member bodies to verify the
competence and veracity of the conformity assessment bodies within each participating member
country.
In many countries, conformity assessment is delegated to independent labs and
agencies, such as Underwriter Laboratories (UL) in the U.S.
32
There are in fact three distinct IEC conformity assessment schemes, though their general logic is the same: IECEE
(-CB or -FCS) for electrotechnical equipment and components, IECQ for quality assessment of electronic
components, and IECEx for "electrical equipment for explosive atmospheres."
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
31
Moreover, each national member body decides for each standard whether to participate
in the IEC-sponsored conformity assessment scheme for that standard or not. This decision is
completely independent of any decision whether to participate in the standards-setting work for
the standard. A decision to participate in IEC conformity assessment for a given standard,
however, entails a commitment to mutual recognition of conformity assessments for this
standard by any accredited conformity assessment body in any of the countries whose member
body participates in IEC conformity assessment for the standard.
In monitoring adoption, implementation or compliance, the national member bodies thus
again play a central role, but the main actors are the conformity assessment bodies (CAB). A
specific CAB may organizationally be a part or subsidiary of a national IEC member body or it
may be accredited by the IEC member body, but for purposes of fulfilling the monitoring
governance function, it is acting independently. Alternatively or in addition, anyone can become
an agent for monitoring by testing products for compliance and publicizing the results; and when
governments write IEC standards into laws and regulations or incorporate them by reference and
thus make compliance mandatory, they may empower public regulatory agencies to monitor (and
possibly enforce) compliance. Strictly speaking, such government agencies monitor compliance
with the law or regulation, but whenever IEC standards constitute the technical basis for these
measures, governments directly become actors in the governance of electric or electronic
technology at the monitoring and enforcement stage, even when they played no direct role at the
rule-making or implementation stage.
III.5. Enforcement
Firms making capital investments or buying inputs (or, more rarely, individual consumers
making purchase decision) can provide "enforcement" if they demand compliance of a product
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
32
with certain standards and refrain from purchasing it (or demand a lower price) if it fails to do so.
Such enforcement works in part through the market mechanism, relying on a mix of
Hirschmannian exit and voice, and is likely to have its greatest effect because its anticipation
motivates firms to adapt, or comply with, the standard. Actual "enforcement" action by these
actors (in the sense of ex post punishment for non-compliance) should be rarely observable,
making it difficult to assess the magnitude of the effect. Notice, however, that the key actors
here are third parties—economic agents, including firms other than those whose products or
actions are the direct target of the standards—who have played no significant role at prior stages
of the governance process.33
The other key actors for enforcement are governments or state bureaucracies. States that
use IEC standards in, for instance, consumer and workplace safety regulations tend to put in
place various enforcement mechanisms. Regulatory agencies are often empowered to levy fines
for non-compliant behavior; private actors may be empowered to bring suits against noncompliant firms, thus involving the judiciary in enforcement. A comprehensive analysis of the
full range of ways to bring public authority to bear to enforce compliance with laws and
regulations that (may) make norm-consistent behavior mandatory is beyond the scope of this
chapter but not necessary to make the main point for purposes of the current analysis:
governmental actors that had little role during previous stages, especially during agenda-setting
and rule-making, may play a key role at the enforcement stage. By contrast, the IEC itself—
including its national member bodies—plays hardly any significant role at this stage.
33
Except through anticipation of their reaction at the adoption/implementation stage.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
33
IV. No Plug for Globetrotters?
Power plugs are fundamentally connectors: They (should) allow the user of electrical
devices, such as household appliances and portable light fixtures, to connect those devices to any
source of sufficient and safe electrical power.
A sketch of the history of their
(non)standardization provides a telling illustration of the effects of the changing cast of actors
during the stages of governance outlined in section IIII.
From a functional perspective, plugs and their counterparts ("outlets" or "sockets")
should be designed such as to allow any safe connection and inhibit unsafe ones. Given that, at
the household level on which I will focus here, there are two common voltages in the world
(110-120V in the U.S. and (consequently) large parts of the Western Hemisphere, Japan,
Madagascar, Libera, and Saudia Arabia; 220-240V everywhere else)34 and given that supplying
household devices with the "wrong" voltage can damage the device and endanger the user, there
should be two sets of mutually incompatible plug and socket designs. Instead there are two
(unidirectionally compatible) plug types for the 110-120V system and at least eleven different
plug types for the 220-240V system, a few of which are mutually or unidirectionally compatible,
but most of which are incompatible and sometimes (such as the Israeli 3-pin plug and socket)
unique to a single country (e.g., McGregor 2007).35
The result (beyond inconvenience to
travelers) is the fragmentation of markets, the loss of economies of scale, and the creation of
incentives to resort to often unsafe multi-standard solutions (see, e.g., IDC 2007).
34
A discussion of the reasons for this divergence and further distinctions, such as between 50Hz and 60Hz systems,
are beyond the scope of this brief overview.
35
Unidirectional compatibility here refers to a case where plug type 1 is compatible with socket type 1 and socket
type 2, but plug type 2 (compatible with socket type 2) is incompatible with socket type 1. The U.S. 2-pin and 3-pin
plugs and outlets are an example of such unidirectional compatibility.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
34
A pre-WWII initiative of Dutch and other European electricity suppliers to standardize
plugs and sockets among European countries was cut short by the outbreak of the war, and early
postwar work in the revived IEC TC 23 produced little more than an inventory of diversity. But
it was supposedly first and foremost frustration with inconsistencies in standards for electrical
installations (wiring) that resulted in part from the diversity of plugs and sockets (and intensified
concern about consumer safety, which is poorly served by some plug-socket combinations) that
let to the formation of a special IEC subcommittee in 1970 to develop a "worldwide plug and
socket outlet system" (see IEC 2007f).
The initiative received some support from major,
internationally competitive manufacturers of electrical devices and a small number of individual
scientists and electrical engineers committed to improving the safety of power plugs and sockets.
These discussions led to a Committee Draft that might have been technically optimal but had
been developed sufficiently in isolation that it ran into vehement opposition at the Enquiry Stage
of the standards development ("rule-making") process: Its flat pins design would have allowed
for no backwards compatibility in any of the 220-240V countries and thus would have required a
sudden switch at massive cost.
After several additional years' work, the IEC committee developed a dual standard with
round pins for 220-240V systems in 1986 (IEC 60906-1), and a new standard with flat pins for
110-120V systems in 1992 (IEC 60906-2). The three-pin plug of IEC 60906-1 was technically
unambiguously superior to virtually all plugs previously used in 220-240V countries; it was
combined (and perfectly compatible) with a two-pin design that had many of the benefits of the
primary, three-pin design while allowing for backwards compatibility with almost all of the
sockets in use in European countries and some non-European ones. This should have made the
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
35
transition to the new global standard plug for 220-240V systems relatively easy, if led by
manufacturers of electrical devices, with building codes and socket manufacturers following suit.
Yet, twenty years later, not a single country has IEC 60906-1-type sockets—concerns
over switching costs and the loss of a technical barrier to trade that was providing economically
significant protection for uncompetitive local industries drove the sometimes intense opposition
to any legislative or regulatory commitment to the new standard. In the absence of that route to
implementation, any breakthrough for the new standard would have required adoption by a large
number of economic actors that lacked an institutional mechanism to coordinate their actions,
making change very unlikely, given network effects. As a consequence, the global standard has
not taken hold, although Brazil has announced plans for 2009 to phase in a variant on IEC
60906-1 type plugs and sockets.36
In sum, contrary to what the international traveler might reasonably think, there is an
international plug standard, which offers a real technical improvement under a variety of
conditions.37 But due to the multitude of actors and domestic interests in avoiding switching
costs and preserving protectionist barriers, the standard has never made it past the rule-making
stage.
36
Note, however, that the two-pin plug standardized in IEC subcommittee 23 as part of 60906-1 became the basis
for what is today known as the "Europlug," developed subsequently by CENELEC and almost universally used on
any appliances sold in continental Europe and some parts of Africa and Asia.
37
Given the different voltages, this standard was never going to help the transatlantic traveler. Ironically, most of
the typical devices used by travels—such as hair dryers, electric shavers, and laptop computers—are today designed
to auto-adjust to voltages between 110 and 240 V and frequencies between 50 and 60 Hz, thus making it safe to use
them in both systems, but since this is not true for all appliances, a single global standard would not even be
desirable as long as the difference in voltage persist.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
36
V. Concluding Discussion
As a corollary of the rationalization and ever greater division of labor that are the
hallmarks of Weberian modernity, we take for granted the functioning of infrastructures and
technologies that most of us do not understand. We want sewer systems to carry our wastewater
without emitting unpleasant odors or feedback into the fresh water supply; we think we should
be able to plug in electrical devices without worrying about them getting damaged or burning
down the house; we are delighted if x-ray machines can wirelessly transmit the image of our
bone structure onto the physician's computer screen if that speeds up our visit to the hospital or
improves the quality of care. In short, we desire and expect all kinds of technology to be safe,
durable, and reliably deliver a certain level of performance without any need for us to worry
about how this result is achieved. Technical standards allow for this to happen. And since they
succeed, by and large, there is a strong temptation to think that technical standards are just about
science. But science is not monolithic. Because science is a social process (Weber 1991
(1919)), it is far from certain that a unique technically optimal solution will be reached, even
when (or if) it exists. And because standards are instruments of governance and standardization
is therefore a political process, it is far from certain that any technical maximum/optimum
constitutes a socio-political equilibrium. Understanding global governance of electro-technology
therefore requires political analysis of the actors in the process, the global governors.
The cast of actors is diverse and changes across the five stages of the governance process
examined here. Outcomes arise from the interactions between these agents and are therefore
unlikely to be simply a manifestation of the preferences of any one actor. Taking the actors and
institutional structures at the domestic and international level into account allows us to
understand both the evolution of global governance of electrical and electronic technology (why
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
37
the scope of IEC governance has so greatly increased over the past century) and the limitations
of global standards (why having global standards does not guarantee global uniformity of
behavior).
References
AIEE (American Institute of Electrical Engineers), Local Reception Committee, Boston Mass. 1904. The Niagara
Falls Electrical Handbook. http://ia331337.us.archive.org/2/items/niagarafallselec00ameriala/
niagarafallselec00ameriala.pdf (last accessed 10/30/2007).
ANSI (American National Standards Institute). 2004. Presentation for "Session 3, Part 1: International Operations"
at the Conference on U.S. Leadership in ISO and IEC Technical Committees, December 1-2, 2004.
Blackbourn, David, and Richard J. Evans, eds. 1991. The German Bourgeoisie: Essays on the Social History of the
German Middle Class From the Eighteenth To the Early Twentieth Century. London–New York:
Routledge.
Brint, Steven. 1994. In an Age of Experts: The Changing Role of Professionals in Politics and Public Life.
Princeton: Princeton University Press.
Büthe, Tim. 2002. "Taking Temporality Seriously: Modeling History and the Use of Narratives as Evidence."
American Political Science Review vol.96 no.3 (September 2002): 481-494.
———. 2006. The Dynamics of Principals and Agents: Institutional Persistence and Change in Setting U.S.
Standards for Accounting. Duke University.
———. 2007a. "Institutional Change in the European Union: Two Narratives of European Commission Merger
Control Authority, 1955-2004." (Manuscript: Duke University, March 2007).
———. 2007b. "Review of Hawkins et al, Delegation and Agency in International Organizations." Perspectives
on Politics vol.5 no.4 (December 2007).
Büthe, Tim, and Jan Martin Witte. 2004. Product Standards in Transatlantic Trade and Investment: Domestic and
International Practices and Institutions. Washington, D.C.: American Institute for Contemporary German
Studies.
Cahan, David. 1989. An Institute for an Empire: The Physikalisch-Technische Reichsanstalt, 1871-1918.
Cambridge: Cambridge University Press.
Cashore, Benjamin, Graeme Auld, and Deanna Newsom. 2004. Governing Through Markets: Forest Certification
and the Emergence of Non-State Authority. New Haven: Yale University Press.
Dahl, Robert A. 1961. Who Governs? Democracy and Power in an American City. New Haven: Yale University
Press.
David, Paul. 1985. "Clio and the Economics of QWERTY." American Economic Review vol.75 no.2 (Paper and
Proceedings of the 97th Annual Meeting of the American Economics Association, May 1985): 332-337.
Davis, Christina L. 2004. "International Institutions and Issue Linkage: Building Support for Agricultural Trade
Liberalization." American Political Science Review vol.98 no.1 (February 2004): 153-169.
DKE (Deutsche Kommission Elektrotechnik). 2004. "Internationales Arbeitsumfeld."
www.vde.com/DKE/Normen+erarbeiten/Internationales+Arbeitsumfeld.htm (last accessed 10/30/2007).
Drezner, Daniel W. 2007. All Politics Is Global: Explaining International Regulatory Regimes. Princeton:
Princeton University Press.
Erdmann, Jeanne. 2007. "The Appointment of a Representative Commission" Geneva: IEC.
http://www.iec.ch/about/history/articles/appointment_commission.htm (last accessed 10/30/2007).
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
38
Gispen, Kees. 1990. "Engineers in Wilhelmian Germany: Professionalization, Deprofessionalization, and the
Development of Nonacademic Technical Education." In German Professions, 1800-1950, edited by
Geoffrey Cocks and Konrad H. Jarausch. New York: Oxford University Press, 1990: 104-122.
Haas, Ernst B. 1990. When Knowledge is Power: Three Models of Change in International Organizations.
Berkeley: University of California Press.
Hawkins, Darren G., et al., eds. 2006. Delegation and Agency in International Organizations. New York:
Cambridge University Press.
Herrigel, Gary. 1996. Industrial Constructions: The Sources of German Industrial Power. New York: Cambridge
University Press.
Hughes, Thomas P. 1983. Networks of Power: Electrification in Western Society, 1880-1930. Baltimore: Johns
Hopkins University Press.
IDC Plugs. 2007. "IDC Plugs - Standards." Alton Hampshire (England): IDC.
http://www.powercords.co.uk/standard.htm (last accesed 10/30/2007).
IEC. 1906. Report of Preliminary Meeting. London: International Electrotechnical Commission.
IEC. 2005. IEC Statutes and Rules of Procedure (2001 edition, as amended on 2004-01-02, 2005-01-07 and 200509-02, endorsed by the National Committees). Geneva: IEC.
IEC. 2006a. "IEC Technical Committee Creation: The First Half-Century." 1906-2006: The Electric Century.
Geneva: IEC. www.iec.ch/online_news/etech/arch_2006/etech_0106/focus.htm (last accessed 5/6/2006).
IEC. 2006b. "Development and Growth of IEC Technical Committees, 1950-2006."
www.iec.ch/online_news/etech/arch_2006/etech_0206/focus.htm (last accessed 1/17/2007)
IEC. 2007a. "IEC Membership." www.iec.ch/about/members-e.htm (last accessed 10/30/2007).
IEC. 2007b. "Members" and "IEC Affiliate Country Programme Participants." www.iec.ch/cgibin/procgi.pl/www/iecwww.p?wwwlang=e&wwwprog=membrs3.p and www.iec.ch/cgibin/procgi.pl/www/iecwww.p?wwwlang=E&wwwprog=membrs32.p (last accessed 10/30/2007).
IEC. 2007c. "Colour Management" www.iec.ch/zone/colooumgmt/cm_intro.htm (and linked pages; last accessed
10/29/2007).
IEC. 2007d. "The IEC in Figures [as of 207-06-18]" Geneva: IEC. www.iec.ch/news_centre/iec_figures/ (last
accessed 10/30/2007).
IEC. 2007e. "Conformity Assessment." Geneva: IEC. www.iec.ch/conformity/ (last accessed 10/30/2007).
IEC. 2007f. "Plugs and Sockets: A Brief History." Geneva: IEC. www.iec.ch/zone/plugsockets/ps_history.htm
(last accessed 10/30/2007).
INCITS (InterNational Committee for Information Technology Standards, Washington D.C.). 2003. "Minutes of
the INCITS Standards Policy Board, April 9-10, 2003."
www.incits.org/minutes/0304incits/0304spb/in031005.htm (last accessed, 5/6/2006).
India, Embassy of. n.d. "International Harmonization of SPS Standards." Paper submitted by India in the WTO
Committee on Sanitary and Phytosanitary (SPS) Measures.
ITA (International Trade Administration). "Department of Commerce Standards Roundtable, Machinery
Equipment, August 5, 2003." Mimeo, at www.ita.doc.gov/ (last accessed 5/6/2006).
Jarausch, Konrad H. 1990. The Unfree Professions: German Lawyers, Teachers, and Engineers, 1900-1950. New
York: Oxford University Press.
Kennelly, Arthur E. 1933. "Conference of the Symbols, Units and Nomenclature (S.U.N) Commission of the
International Union of Pure and Applied Physics (I.P.U.) in Paris in July, 1932 and Its Results"
Proceedings of the National Academy of Sciences of the United States of America vol.19 no.1 (15 Jan
1933): 144-149.
Kindleberger, Charles P. 1983. "Standards as Public, Collective and Private Goods." Kyklos vol.36 no.3: 377-396.
Krasner, Stephen D. 1991. "Global Communications and National Power: Life on the Pareto Frontier." World
Politics vol.43 no.3 (April 1991): 336-366.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
39
Loya, Thomas, and John Boli. 1999. "Standardization in the World Polity: Technical Rationality over Power." In
Constructing WOrld Culture: International Non-Governmental Organizations since 1875, edited by John
Boli and George Thomas. Stanford, CA: Stanford University Press, 1999: 169-197.
Marceau, Gabrielle, and Joel P. Trachtman. 2002. "TBT, SPS, and GATT: A Map of the WTO Law of Domestic
Regulation." Journal of World Trade vol.36 no.5 (October 2002): 811-881.
Mattli, Walter. 1999. The Logic of Regional Integration: Europe and Beyond. New York: Cambridge University
Press.
Mattli, Walter, and Tim Büthe. 2003. "Setting International Standards: Technological Rationality or Primacy of
Power?" World Politics vol.56 no.1 (October 2003): 1-42.
McGregor, Conrad H. 2007. "Electricity Around the World." http://users.pandora.be/worldstandards/electricity.htm
(last accessed 10/30/2007).
McNichol, Tom. 2006. AC/DC: The Savage Tale of the First Standards War. San Francisco: Jossey-Bass.
Meidinger, Errol, Chris Elliott, and Gerhard Oesten, eds. 2003. Social and Political Dimensions of Forest
Certification. Remagen-Oberwinter: Forstbuch Verlag.
Meyer, John W. 1980. "The World Polity and the Authority of the Nation-State." In Studies of the Modern WorldSystem, edited by Albert Bergesen. New York: Academic Press, 1980: 109-137.
Milner, Helen V. 1988. Resisting Protectionism: Global Industries and the Politics of International Trade.
Princeton: Princeton University Press.
N.N. 1905. Transactions of the International Electrical Congress, St. Louis, 1904. Albany, NY: J. B. Lyon Co.
Nord, Philip. 1995. The Republican Moment: Struggles for Democracy in Nineteenth-Century France.
Cambridge, MA: Harvard University Press.
O'Rourke, Kevin H., and Jeffrey G. Williamson. 1999. Globalization and History: The Evolution of a NineteenthCentury Atlantic Economy. Cambridge, MA: MIT Press.
Pierson, Paul. 2000. "Increasing Returns, Path Dependence, and the Study of Politics." American Political Science
Review vol.94 no.2 (June 2000): 251-267.
———. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton: Princeton University Press.
Prakash, Aseem, and Matthew Potoski. 2006. The Voluntary Environmentalists: Green Clubs, ISO 14001, and
Voluntary Environmental Regulations. New York: Cambridge University Press.
Rodgers, Daniel T. 1998. Atlantic Crossings: Social Politics in a Progressive Age. Cambridge, MA: Belknap
Press of Harvard University Press.
Ruppert, L. 1956. Brief History of the International Electrotechnical Commission. Geneva: IEC.
Sciulli, David. 2005. "Continental Sociology of Professions Today: Conceptual Contributions." Current Sociology
vol.53 no.6 (November 2005): 915-942.
Seeker, Steven A. 2002. "Report of the U.S. National Committee of the IEC." Washington, D.C.: ANSI, 2002.
Steinberg, Richard H. 2002. "In the Shadow of Law or Power? Consensus-Based Bargaining and Outcomes in the
GATT/WTO." International Organization vol.56 no.2 (Spring 2002): 339-374.
Thelen, Kathleen. 2000. "Why German Employers Cannot Bring Themselves to Dismantle the German Model." In
Unions, Employers, and Central Banks: Macroeconomic Coordination and Institutional Change in Social
Market Economies, edited by Torben Iversen, Jonas Pontusson and David Soskice. New York: Cambridge
University Press, 2000: 138-169.
Trapp, Wolfgang, and Heinz Wallerus. 2006. Handbuch der Maße, Zahlen, Gewichte und der Zeitrechnung.
5. durchgesehene und erweiterte Auflage. Stuttgart: Philipp Reclam junior.
Vogel, David. 1995. Trading Up: Consumer and Environmental Regulation in a Global Economy. Cambridge,
MA: Harvard University Press.
Warburg, E. 1916. "Werner Siemens und die Physikalisch-Technische Reichsanstalt" Naturwissenschaften vol.4
no.50 (December 1916): 793-797.
Büthe
Power of Norms, Norms of Power (Draft 11/4/07)
40
Weber, Max. 1991 (1919). "Wissenschaft als Beruf." In Max Weber: Schriften zur Wissenschaftslehre, edited by
Michael Sukale. Stuttgart: Philipp Reclam junior, 1991 (1919): 237-273.
Williams, Bob. 2001. "Driving the Standardization of ITS forward." ISO Bulletin (September 2001): 6-9.
Zuckerman, Amy. 1999. "International Standards on Front Burner." Electronic News online 1/4/1999. reedelectronics.com/electronicnews/article/CA71595.html