Ruef, Markard 1 EASST 2006
EASST Conference 2006, August 23-26, University of Lausanne
Session: “Politics of Expectations: dynamics, production and practices of technological
expectations”, PoE II, Konrad et al.
What happens after a hype? Changing expectations and their effect
on innovation activities in the case of stationary fuel cells
Annette Ruef, Jochen Markard
Cirus - Innovation Research in Utility Sectors
Eawag, Ueberlandstrasse 133, 8600 Dübendorf, Switzerland
Keywords: expectation dynamics, levels of expectations, fuel cell
Innovation processes are influenced by the social dynamics of expectations, for instance by
hypes and subsequent disappointments of expectations about the future of a technology or
innovation. This paper takes a closer look at the expectation dynamics in the case of
stationary fuel cells and its impact on the development of the innovation field. We observed a
certain disappointment following a hype at the beginning of the millennium (2000-2001) but
no notable negative effects on the innovation activities. We stress the need to distinguish
different types of hype-disappointment dynamics according to different levels of expectations.
In the case of fuel cells, societal expectations on the technology (such as the vision of a
future hydrogen economy) remained intact and allowed for a mere scaling down of the high
risen expectations after the end of the hype. Together with the considerable degree of
institutionalisation which has taken place in the course of the hype period, this largely
explains why innovation activities continue almost unabated.
Ruef, Markard 2 EASST 2006
The development of new technologies or innovations is often accompanied by the occurence
of hypes. Such temporary peaks in public attention and high rising expectations about the
potential of an innovation do not pass without consequences. On the one hand, they are able
to add considerable momentum to an innovation process by mobilising interest and
resources and coordinating heterogeneous actors. On the other hand, hypes may also have
ultimately damaging effects on the innovation process if ‘overshooting’ promises and
expectations are not met.
The social dynamics of expectations with ups and downs of hypes and disappointments have
been examined in the field of science and technology studies with a focus on their impact on
technological developments (cf. van Lente 1993; van Lente & Rip 1998; Brown et al. 2000;
Konrad 2004). 1 Various case studies allowed the identification of mechanisms at play and
typical patterns of expectation dynamics. Concepts such as promise-requirement cycles (van
Lente 1993; van Lente 2000) or hype-disappointment cycles (Brown 2003) explain how
hypes are built up and how widely shared expectations can get a dynamic of their own. Such
collective expectations influencing a variety of actors and activities can be characterised as
institutional structures shaping technological development (Bender 2005).
More often than not, however, reality turns out to be different from expectations, which can
fail for a variety of reasons (Geels & Smit 2000). Disappointments of hyped expectations and
their effects on technological developments are not very broadly discussed in the literature,
where the focus has rather been on the emergence of hypes and on mechanisms with
positive effects on the innovation processes. Still, it has been suggested that the possible
‘failure arising from overheated expectations’ is linked to costs and that down-turns in
expectations can occur very rapidly (Brown 2003; Van Lente 1993). Furthermore, the
complex relationship between disappointments and a basically flexible interpretation of
project results has been identified as a potential mechanism at play (Konrad 2006). In the
business literature, hypes are even referred to as a rather common phenomenon in the
course of the development of new technologies. According to the Gartner hype cycle model,
it is quite usual that expectations about a new product or technology pass a phase of
disillusionment before they are eventually introduced into the market. 2 Obviously, this cannot
be the case for every innovation, i.e. the degree of disappointment may vary and the actual
impact on the innovation process may well be different. It remains thus a topic for further
research, what the underlying dynamics of disappointments are and under what
circumstances innovation process turn out to be robust enough to endure such
This paper analyses expectation dynamics and innovation activities in the case of stationary
fuel cells, a technology for which attention in the media strongly increased by the end of the
See also the recent special issue of Technology Analysis & Strategic Management, 18, 3/4 on the
See http://www.gartner.com/pages/story.php.id.8795.s.8.jsp [2006-06-14].
Ruef, Markard 3 EASST 2006
1990s and peaked in 2001 before it came to a clear drop of media interest. Despite several
changes at the expectations level and a certain kind of disappoint after the hype, innovation
activities did not show any decline but were even intensified. Our aim is to understand the
interaction of expectations and innovation processes in the case at hand and to identify the
mechanisms that may explain the development.
In conceptual terms, we draw on the literature on expectation dynamics and innovation
processes. With regard to the former, we particularly suggest to distinguish different levels of
expectations including expectations about societal forces and trends at the macro level,
expectations related to the technological field or sector (meso level) and expectations about
small scale, local events and entities at the micro level (van Lente 1993, 182ff.). And with
regard to the latter, we pay particular attention to institutionalisation processes that may lead
to the emergence of niches (e.g. Kemp et al., 1998) or even a technological innovation
system (e.g. Carlsson et al., 2002), in which innovation activities are embedded in and
sustained at the same time.
The remainder of this article is structured as follows. After an introduction of the methods
used, the findings of the case study on stationary fuel cells are presented in more detail.
Chapter 4 then provides an interpretation of expectation and innovation dynamics and the
interplay of what we may call discourse arena and innovation arena. Chapter 5, finally,
concludes on the explanatory strengths of a differentiation of at least two levels of
The study, which was conducted in 2004, includes two major elements of analysis: an in-
depth, qualitative discourse analysis and a review of innovation activities. The scope of our
survey was limited to the German speaking part of Europe (Germany, Switzerland, Austria)
and the period of 1993 to 2003/04.
The discourse analysis had the aim to identify expectations and the change of expectations
over time. In the study, we concentrated on the public discourse on stationary fuel cells
expressed in newspapers. A selection of articles mentioning fuel cells in their lead or title was
analysed in full text. 3 Text passages with explicit or implicit references to the future of fuel
cells formed the primary units of analysis. To examine the textual data we adopted a
sociology of knowledge approach to discourse (in German: wissenssoziologische Diskurs-
analyse, cf. Keller 2004; Keller 2005). The way discourses are conceptualised by Keller as
“structured and structuring structures” (with reference to Bourdieu 1985) fits well the role of
widely shared or collective expectations in technological development. Both concepts share
the view of a socially constructed reality. Discourses are not the intentional outcomes of any
individual effort, but rather unintended results of heterogeneous practices performed by
Further selection criteria were the length (>200 words) and the topic of the article – not exclusively
dealing with mobile or portable applications. A sample of every second year from 1993 to 2003 was
analysed, a total of 46 articles (out of 96).
Ruef, Markard 4 EASST 2006
social actors. 4 Discourses, at the same time, may influence or structure actors’ practices by
offering normative orientations, rules of signification for meaning constitution or legitimisation
for action (Keller 2005).
Collective expectations have a similar emergent quality (Konrad 2006) as they are not
attributable to single actors. Furthermore, they may provide some sort of justification of
innovation activities and the strategic allocation of resources to a specific product or
technology. Finally, they function as orientation points for the coordination of heterogeneous
actors in the same way as discourses do. Thus, the expectations about fuel cells were
assumed to form a discourse about the future of this technology.
The analysis of the data corpus – each document representing a “fragment of discourse”
(Jäger 1999) – was carried out with different tools from grounded theory (Strauss & Corbin
1996) and concepts proposed by Keller (Keller 2005). In a computer-aided qualitative data
analysis with the software Atlas.ti, different steps of “coding”, writing “memos” and building
“code families” etc. were carried out in order to reconstruct the discourse. On the basis of the
empirical data, a classification of different types of expectations was elaborated including
specific expectations at the level of innovation projects and more general expectations at the
level of the innovation field. For these categories, expectation contents were further
characterised along two dimensions: potential applications and the time frame of
commercialisation targets. Furthermore, a third category of expectations was assigned to
interpretative schemes or frames used in the discourse (“Deutungsmuster”, see Keller 2005).
The latter are very broad expectations placing the technology in the context of generic
societal visions. These frames contain a sort of scenarios for the future – of society as a
whole and of the use of the technology.
In the second part of the analysis, the development of the innovation activities from 1993 to
2004 was reconstructed with the help of different publicly available data sources (event
calendars, websites, newsletters, newspapers and professional journals, reports, etc.). Data
were collected on the conference activities related to hydrogen and fuel cells and on the
activities of key players (active in Europe) in the field of stationary fuel cells in order to trace
major achievements and innovation activities at the firm level. In addition, the emergence of
fuel cell initiatives as well as the set up of research and development programmes at the
national and regional level in Germany, Switzerland and Austria was observed.
Analytically, we thus differentiate discourse and innovation activity, respectively discourse
and innovation arena as the levels the activities can be attributed to. Despite this
differentiation discourse activities can be closely linked to innovation activities, e.g. when a
fuel cell manufacturer launches a new prototype and communicates this event in a press
release that is picked up by the media. In some cases, events or activities may be quite
ambiguous in this respect. Conferences, for instance, can be regarded as innovation activity
in the sense that they contribute the diffusion of knowledge and as a discourse activity as
they represent a platform where actors deliberately influence the professional discourse.
Discourse activities, unlike the discourse itself, can also be expected to be intentional in most
cases (see below)
Ruef, Markard 5 EASST 2006
3 Expectation dynamics and innovation activities from 1993 to 2003
This chapter presents the key findings on how the expectations about stationary fuel cells
have evolved over the last decade and which way innovation activities have taken in the
3.1 Evolution of expectations: Results of the discourse analysis
In was said before, three different types of expectations were reconstructed including
product- or project-specific expectations, generalised expectations about the future of the
technological field and interpretative schemes or frames.
At all three levels we observed a change of expectations over time with regard to content
(e.g. commercialisation target) and quality (positive, neutral or negative expectation).
However, the degree of change was different at the three levels. Actually, expectations at the
micro- and meso-levels changed significantly while the framing of the technology as such
remained relatively stable. In terms of expectation dynamics, roughly three periods can be
distinguished (see Table 1): The rediscovery of fuel cells (1993-1997), the hype (1999-2001)
and the disenchantment (2003).
Table 1. Changes in the content of expectations at different levels over time
Rediscovery Hype Disenchantment
1993-1997 1999-2001 2003
Dominant frames Technological progress Strategy Decentralisation
Generalised Electricity production Cogeneration, heating Energy production
expectations (decentralised small power (residential units combined (decentralised small and
(applications, plants) with gas boilers) micro cogeneration units,
commercialisation possibly combined to
prospects) virtual power plants)
Longer-term Short-term commercialisation Medium- to long-term
Specific Decentralised energy Decentralised energy
expectations production: production:
(applications, ● Siemens, Sulzer (1993): ● Siemens (1999):
commercialisation 50 to 200 kW, > 6 years 250 kW, 2006
target) ● MTU (2001)
250 kW, 2005
FC heating systems: FC heating systems: FC heating systems:
● Sulzer Hexis (1997): ● Vaillant (1999/2001): ● Vaillant, Sulzer Hexis:
1 kW, 2001 1-4,5 kW, 2001/2004 1-4,5 kW, unclear
● Sulzer Hexis (1999/2001): ● Viessmann:
1 kW, 2001/2005 4,5 kW, 2006
● Others (2001):
1,5-4,5 kW, 2003/04 or
In brackets: year of the article in which the statement was made; the change of frames is depicted in detail in
Ruef, Markard 6 EASST 2006
At the level of firms and projects, only a few statements with specific expectations were cited
in the media during this period. 5 In 1993/94, Siemens and Sulzer, two well established
industrial companies, communicated their intention to develop stationary fuel cells of 50 to
200 kW (SOFC). Both had been working on (small) prototypes for several years and were
prepared for longer development periods (at least six years) to achieve their goal of
commercial stationary fuel cells. At the same time, a demonstration project of a fuel cell from
ONSI (USA) carried out by a Swiss utility (SIG) was started in big optimism about the short
term-availability of competitive fuel cells: The project manager expected the serial production
of fuel cells in three to four years, which would allow for electricity prices similar to
conventional (nuclear) power. A few years later, in 1997, no news about such developments
was available though. Only Sulzer was present in the media, announcing the spin-off of the
fuel cell development into the new firm Sulzer-Hexis, which aimed now at a new product:
very small, 1 kW-fuel cells combined with a gas boiler for residential cogeneration. The
commercialisation of the product was targeted for 2001 in collaboration with partners from
the heating sector.
The rediscovery period was characterised by rather diffuse generalised expectations about
the future of fuel cells. The technology as such was thought to be promising for energy
conversion, mainly for electricity generation, thanks to its exceptionally high (future) degree
of efficiency. A differentiation of application fields was found in some of the statements (i.e.
examples of mobile and stationary applications), mainly to express the variety of uses for fuel
cells. “Small” or “compact” fuel cell power plants were usually mentioned as future stationary
applications whereas the fuel cell car represented a typical mobile application.
Commercialisation prospects were not discussed in most of the expectation statements. It
seemed to be clear, however, that the technical development was well advanced (at least for
certain types of fuel cells) while commercial products were not to be expected very soon.
During the whole rediscovery-period, the development of fuel cells was mainly framed as part
of technological progress which is intrinsically positive and does not need any justification. 6
In the context of this frame, there was no doubt about the need to invest in the development
of fuel cells, nor about the innovation achieving a product status sooner or later. In addition to
the progress frame, the innovation was also presented in ecological frames (see Table 2),
such as the need to fight pollution and climate change (low emissions, no CO2 in the ideal
case), and to find alternative energy sources (fuel cells run without fossil fuels). Another
interpretative scheme was used only occasionally, mostly in 1997, setting fuel cells in the
context of strategic interests of firms. Manufacturing firms were argued to be interested in
The period is termed rediscovery because the actual discovery of the fuel cell principle goes back
to the end of the 19th century.
This interpretative scheme is characterised as an “ideograph” by Van Lente (van Lente 2000, 47-
48): „The ideograph ‚technical progress’ thus contains an element of ongoing evolution, an
unfolding logic that is captured in the notion of a ‚next generation’. The notion is part of the rhetoric
of a progress that should not be stopped – because not to have the next generation is to commit
Ruef, Markard 7 EASST 2006
being pioneers in the development of fuel cells, and some utilities were thought to be keen
on testing fuel cells as one of the technologies for decentralised energy production.
At the level of project- and product-specific expectations, the development of fuel cells for
residential cogeneration (< 10 kW) became dominant in the hype-period. Projects for bigger
power plants were only discussed for two manufacturers (Siemens and MTU), both aiming at
units of about 250 kW available in serial production by 2005 or 2006/07. After Sulzer-Hexis
had launched a project for residential cogeneration in 1997, the development of fuel cell-
“heating systems” was announced by the heating manufacturer Vaillant in 1999, who then
planned their introduction into the market by the end of 2001. Two years later, in February
2001, its prospects for serial production had shifted to 2004. By the time, however, a number
of other firms from the heating sector had entered the competition for the development of
residential cogeneration units. All of them were manifesting great optimism about a
commercialisation within a few years – serial production was generally expected by 2004 and
the achievement of competitive prices by 2007. The “first mover” of the field, Sulzer Hexis
(which had announced the development of 1 kW-units for residential cogeneration in 1997)
was still lying ahead, executing field tests all over Europe while the other manufacturers were
still developing prototypes. Even if it had not achieved its original goal of serial production,
the start of a “pre-series” in 2001 was largely publicised in the press, and a production of
bigger series (10’000 units/a at competitive prices) was announced for 2005.
In the hype-period, diffuse generalised expectations about fuel cell technology with vague
commercialisation prospects were still present, increasing in number and optimism. Apart
from that, generalised expectations about a particular application prevailed in this period,
namely about small residential cogeneration units (< 10 kW). The focus had shifted from
electricity production to cogeneration, and the relatively new idea of heating public and
private buildings with fuel cells (combined with gas boilers) had moved into the centre of
interest. Indeed, such micro-power plants for residential co-generation, referred to as “fuel
cell heating systems”, were seen in competition to conventional heating systems. They were
seen as the first promising field of commercialisation for stationary fuel cells, and the first
serial productions were expected to be started within a few years. A majority of the discourse
was highly optimistic, expecting serial production of residential cogeneration units already by
2003 or 2004, whereas more reluctant voices warned from too much euphoria. In spite of the
argument that the technology would need more time to be introduced into the market, their
time horizon for commercialisation was not much longer – i.e. expecting serial production “at
the earliest in the second half of the century” or by 2005 or 2006. The general optimism of
the hype period was also highlighted by the fact that technical or economic problems which
remained to be solved to achieve this goal were hardly discussed.
In 1999 and 2001, the strategy-frame had become the dominant interpretation scheme in the
discourse about the future of fuel cells. The development of the innovation was seen as
driven by firms’ strategic interest in new products and markets. On the one hand, first mover
advantages were (implicitly) assumed to be an important motivation for “pionieer” firms to
invest in the development of fuel cells. On the other hand, the competition to “stay in the
Ruef, Markard 8 EASST 2006
race” (Eames & McDowall 2006) was seen as a justification for the commitment of others.
Both version of the frame – opportunity vs. challenge – were found for manufacturers as well
as for utilities, but by 2001, fuel cells (and not to miss out on their market potential) were
more and more often presented as a competitive challenge. The ecological frame was also
referred to from time to time during this period. The vision of a “hydrogen economy”,
however, appeared as a frame only in 2001. This interpretative scheme places fuel cells at
the heart of an emission free energy economy based on ‘renewable’ hydrogen (be it called
hydrogen economy or not) which is supposed to solve environmental as well as resource-
and energy-economic problems. This vision had been alluded to only once in the rediscovery
period (1993-1997), but in 2001 it appeared several times, expressing the high expectations
connected to fuel cells as a future key technology.
The few statements of specific expectations at the micro-level still concentrated on
residential cogeneration in 2003. By contrast to 2001, however, the manufacturers cited
refrained from indicating new deadlines for a commercialisation of their products. 7 On the
contrary, they cautiously avoided precise forecasts or conceded that they did not know when
the targeted price level (about 1000 €/kW) would be achieved. As a matter of fact, the goals
uttered two years before had not been achieved or seemed not realistic any more, even if
this was not expressed directly.
Similar reservation was observed at the level of generalised expectations. Hurdles on the
way to commercialisation were broadly discussed (for the first time since 1993), emphasising
above all the necessity of cost reduction, but also technical requirements (such as a long
enough lifetime) still to be reached. A wider introduction into the market was now generally
expected to require about ten more years, at least it seemed not realistic anymore before
2010. Apart from residential cogeneration (still prominent in the discourse), the range of
future applications considered was widened to decentralised energy supply in general.
Occasionally, the idea of virtual (fuel cell-) power plants was referred to, which had not been
present before. 8 In other words, the expectations about stationary fuel cells had become
more diffuse again (even if residential cogeneration was still at the centre of interest) – as
well in terms of application fields and functionalities, as in terms of their time to market. This
“scaling down” of the (quantified) expectations on applications for residential cogeneration
and their commercialisation prospects can be read as an expression or effect of non-
fulfilled/disappointed expectations. That no explicit disappointment was expressed in the
discourse and that no negative results of the techno-economic development were reported
can be mainly ascribed to the logic of the media discourse. 9
At least Sulzer Hexis and Vaillant. An exception is formed by heating manufacturer Viessmann who
seems to keep to the targets set at the beginning of its development program (2000).
The term “virtual power plant” appeared for the first time in 2003 in the daily press while it had
already been mentioned occasionally in the weekly press (Zeit and Spiegel) from 2000 on.
The media need occasions for the press coverage of a technology which are often “produced” by
the (lead) actors of the field, for instance with press releases about the start of a demonstration
project, a fair about fuel cells or a new industrial partnership (scientific findings published in
Ruef, Markard 9 EASST 2006
More remarkably though, the general optimism about the future potential of stationary fuel
cells remained intact. The justification of further investments into fuel cell-development was
therefore not put into question. Rather, the need of continued efforts was stressed. Although
the frame of firms’ strategic interest in future market potentials was less prominently used
than in the hype phase, it was brought into play to emphasise the competitive challenge
imminent for the heating and for the utility sector. Moreover, the trend towards a higher
degree of decentralisation in the energy sector gained some importance as a context for
future applications of stationary fuel cells. While increasing decentralisation was primarily
seen as an opportunity for the market introduction of stationary fuel cells, these were also
assigned the potential to revolutionise the whole energy supply. Other frames such as the
scenarios of climate change and a hydrogen economy played a rather marginal role in this
Table 2 provides an overview of the frames identified in the discourse (mostly used in
parallel). The dominant frames for the three periods are highlighted.
Table 2. Overview of frames identified in the discourse analysis
Context Frame Scenario/trend Role assigned to fuel cells Occurence
Progress Technological Process of ongoing evolution part of the progress as a mainly
(society) progress towards better technologies „new“ technology 1993-1997
Revolutionary Substitution of technologies substitute conventional 2003
potential heating systems and change
decentralised energy supply
Ecology Protection of Environmental degradation prevention of pollution and 1993-2003,
climate and and climate change going on greenhouse gas emissions always in
New energies End of fossil fuels provide (transition to) non- 1993-2003,
approaching fossil energy always in
Hydrogen- New energy „era“, based on key technology, in 2001, (2003)
economy 11 hydrogen combination with hydrogen
Energy Energy supply National economies need (one of the alternatives to) 1993-2003,
policy new power sources secure energy supply always in
Decentralisation Decentralisation of energy part of the trend towards 2003
supply/generation more decentralised energy
Science or Nature form another – rare – occasion for articles about fuel cells). As long as nobody
is interested in reporting “bad results” or having a negative press (i.e. in the absence of
opponents/critics of the technology), only “positive” reports can be found. xxxREFxxx
Cf. Van Lentes description of the ideograph of technological progress (van Lente 2000) and the
narrative of „Inevitability and technological progress“ in Eames & McDowall 2006.
Cf. The vision of a Hydrogen economy with six overarching and competing narratives (Eames &
Ruef, Markard 10 EASST 2006
Strategy 12 Market potential New markets offer promise new markets and 1997, mainly
opportunities for first- opportunities 1999-2001,
movers/challenges for others 2003
Utilities Different trends challenge opportunity for differentiation 1999-2001,
utilities and decentralised energy 2003
3.2 Development of innovation activities
Different indicators show an increasing trend in the innovation activities in the field of
stationary fuel cells with a considerable rise at the end of the 1990s and no notable break-in
after the end of the hype in 2001. This section briefly outlines the development of the
indicators collected to retrace the innovation activities: conference activities, development-
projects of major firms present in Europe as well as research and development programmes
(and -initiatives) started in the observation period.
During the 1990s, the number of conferences 13 about hydrogen and fuel cells remained
relatively low. However, a notable increase in conference activities took place from 2000 on,
peaking in 2002 and decreasing somewhat since then (see Figure 1). The development is
shaped mainly by single conferences on the subject, while conference series show a more
stable development. As a matter of fact, many new conference series have been started over
time – quite a number of them in 2002, while only three out of twenty do not exist anymore in
Number of Conference (series) per year
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Figure 1. Development of conference activities from 1994 to 2004
Cf. The narrative of “staying in the race” in Eames & McDowall 2006.
As a „conference“ were counted events of minimum half a day aiming at a professional public,
some of them combined with fairs (such as the Hannover Messe); the number of conference-days
shows a similar development.
Ruef, Markard 11 EASST 2006
Industrial innovation activities showed an increasing number of firms engaging in fuel cell
product-developments, above all in the heating sector at the end of the 1990s. The field of
small fuel cell power plants (200-250 kW) remained highly stable over the whole period with
four major companies active in development-projects: IFC/ONSI, Siemens, MTU and Ballard.
In 1997, Siemens and MTU started their field tests in Europe while ONSI-fuel cells were
tested already in 1993 and Ballard achieved the test stage only in 2000. 14 In the field of
residential cogeneration units (< 10 kW) many development projects were started only at the
end of the 1990s. Sulzer had been the first company (in Europe) active in the development of
micro-fuel cells, starting a fuel cell project in 1988 which was sourced out into the spin-off
Sulzer-Hexis in 1997, developing 1 kW fuel cells for residential cogeneration. In 1999,
Vaillant announced the development of fuel cell-heating systems as the first heating
manufacturer. The partnership of Vaillant with the US manufacturer Plug Power dated back
to 1997, and from 1999 on Vaillant took a prominent role. A number of other firms, mainly
from the heating sector, then started development projects with different partners in 1999,
2000 or even later, some of them incited by public funding programmes: Hamburg Gas
Consult/European fuel cells, Viessmann, Nuvera and Buderus/BBT Thermotechnik. Even if
the intensitiy of the development activities of single firms (i.e. the resources spent for the
projects) cannot be judged from the available data, they show that all projects were carried
on after 2001 (as retraced until 2004).
Tab 1. Overview of development-projects of major European firms
The engagement of utilities in development activities of fuel cells were more difficult to trace,
but judging from the available data (mainly from firm’s websites and the discourse analysis),
they have increased as well since the end of the 1990s. Many of the demonstration projects
with ONSI-fuel cells (200 kW) carried out at the beginning of the 1990s have seen no follow
up and utilities involved in these early tests adopted a rather passive observation strategy
after the stagnation of ONSI-fuel cells. However, some of the utilities have taken up fuel cell
activities again in recent years together with the bunch of others starting fuel cell projects
since the year 2000. Most prominently, big German utilities such as EnBW, EWE, RWE or
Ruhrgas/E.ON (most of them dealing with fuel cells for some time already) have intensified
their fuel cell activities in these years, adopting different strategies regarding the capacities of
the units tested. In most of their test- and demonstration-programmes – often in partnerships
with other utilities and gas suppliers – small fuel cell power plants (about 250 kW) play a
major role, but residential cogeneration has also been widely tested since the end of the
1990s. Field tests and public fuel cell “pavillons” were generally occasions for press releases,
but fuel cell heating systems were even the object of big advertisement campaigns of RWE
and EnBW in 2001 and 2002. More recently, RWE Fuel Cells (a spin-off created in 2002)
entered partnerships with manufacturing firms: a joint venture with MTU in 2003 and a
Ballard as the exclusive fuel cell manufacturer had started the development of stationary fuel cells
only in 1993 after concentrating on mobile fuel cells since its foundation in 1988. The other fuel cell
projects were all part of traditional industrial companies.
Ruef, Markard 12 EASST 2006
collaboration with BBT Thermotechnik (and IdaTech) in 2004. RWE thus adopted a clear
leader strategy (Markard, 2004) and was the dominating discourse actor since 1999. With
one exception, no failures or abandonment of demonstration projects are known. The joint
project of Siemens and several utilities (with EU and US public funding) for a 1 MW-fuel cell
combined with a microgasturbine had to be given up in 2002, because no microgasturbine of
the required dimension was available on the market.
Besides joint ventures and inter-firm partnerships (between suppliers, manufacturers and
utilities), industrial networks and fuel cell initiatives (with members from different sectors and
public funding) have been initiated in the last years of the observation period. In 2001 for
instance, a number of utilities (EWE AG, MVV Energie AG, E.ON Ruhrgas AG, and VNG AG)
have founded the “Initiative Brennstoffzelle” with the aim of coordinating the activities in the
field of residential cogeneration. A growing number of German federal states initiated their
own fuel cell or hydrogen initiatives together with industrial partners since 2001, only the
programmes of Bayern, Baden-Württemberg and Nordrhein-Westfalen dating back to the
end of the 1990s. The year 2004 has seen the foundation of the “Brennstoffzellen-Bündnis
Deutschland BZW” bringing together the 20 leading associations and initiatives on fuel cells
with another 300 members. The strategy of market introduction for fuel cells elaborated by
the BZB mainly aims at mobilising political support for the technology. Its slogan “Fuel cells
are coming – if not from Germany, they will come to Germany” alludes to the “staying in the
race”-narrative (Eames & McDowall 2006), this time referring not to firms’ strategies but to
the competition between nations.
Hydrogen and fuel cells has indeed become a prominent subject on the national and
European political agendas, especially since 2002. Public funding programmes for hydrogen
and fuel cells already existed at the beginning of the 1990s (in Germany, Switzerland and the
EU) and saw a notable increase in the public funding since 1999/2000. Germany has
launched extra programmes in 2000 and 2001, 15 and a law on cogeneration (Gesetz zur
Kraft-Wärme-Kopplung) of 2002, defines a a bonus for small cogeneration units (< 50 kW, for
ten years) to support fuel cells among other technologies. The European Union launched a
High Level Group in 2002 in order to elaborate a coherent strategy for the support of
hydrogen and fuel cells in Europe. The engagement of the EU for hydrogen and fuel cells
has especially increased since 2003 when Romano Prodi declared a hydrogen economy as
a goal to reach by 2050. The European Initiative for Growth assigned the projects Hycom
and Hypogen extra funds of 2.8 billion Euro for the support of hydrogen and fuel cell
technologies for 2005-2015. 16
In sum, the indicators described above illustrate the trend towards a notable increase in
innovation activities for stationary fuel cells at the end of the 1990s. Most remarkable is the
A programme for residential fuel cell power plants for about 10 Mio. Euro (Weider et al. 2003, p.6)
and the Zukunfts-Investitions-Programm, with about 60 Mio. Euro for fuel cells (for three years),
more than half of which were flowing into stationary application projects.
See report of the European Commission of November 11, 2003: COM(2003) 690 final,
Ruef, Markard 13 EASST 2006
growing number of firms engaging in the development of fuel cell heating systems since
1999. The developments after 2001 did not show any decrease in innovation activities,
except for the number of (single) conferences. On the contrary, new partnerships and
initiatives were founded, new conference series on hydrogen and fuel cells initiated and extra
public funding programmes started from 2002 on.
The findings presented above can be interpreted in the context of expectation dynamics and
their influence on innovation processes known from the literature. Here we focus on
expectation and innovation dynamics as such and on the interaction of expectations and
innovation activities, which will turn out to explain much of our empirical observations.
The expectation dynamics observed in the case of stationary fuel cells can be interpreted as
some kind of hype- and disappointment-cycle (e.g. Brown 2003) which went along with a
conversion of promises into requirements (cf. van Lente 1993; van Lente 2000).
Different aspects point to the building up of a hype in 2000/01. These include the increasing
focus on one particular application of fuel cells, i.e. fuel cells for residential buildings, the
quasi-absence of a critical public discussion on technical or economic problems and the
emphasis on advantages and technology potentials together with a sudden belief in short-
term commercialisation prospects. The less optimistic and downscaled expectations of 2003
can be read as some sort of disappointment. Now, problems and requirements of the techno-
economic development were highlighted in contrast to the uncritical optimism of the previous
period and the short-term perspective on commercialisation had vanished. Instead, more
vague or medium- or long-term perspectives were taken on commercialisation. Finally, the
strong focus on residential cogeneration had widened (again) to fuel cell applications for
decentralised energy supply in general.
Furthermore, changes in the framing of fuel cells may be regarded as an indicator that
promises have transformed into requirements in the course of the hype period. The frame of
strategic interest, which clearly represents requirements for the actors in the field, has been
used increasingly since 1997 and even became the dominant one during the hype period. Its
use had switched from “opportunity to challenge” between 1999 and 2001, i.e. the
development of fuel cell heating systems was increasingly presented as a competitive
challenge, a requirement to fulfil for manufacturers as well as for utilities in order to stay in
the race. Moreover, the two formerly independent sectors of electricity supply and heating
were now perceived to compete for the future market for residential cogeneration. Thus, a
certain pressure to fulfil the promise – or at least “staying in the race” – had created a
requirement for other actors to become active.
The review of innovation activities points to the emergence of a technological niche (cf.
Kemp et al., 1998) around fuel cells for residential cogeneration and can even be interpreted
Ruef, Markard 14 EASST 2006
as the development of a technological system (cf. Carlsson et al., 2002, Jacobsson and
Bergek, 2004) for stationary fuel cells in general.
At the beginning of the observation period, innovation activities were largely driven by
industrial players and political support was rather low. Still, the number of firms involved in
stationary fuel cells increased steadily and so did pilot projects, development activities etc.
Innovation networks not only grew larger but also showed an increasing number of
formalised relationships such as joint-ventures, officially announced co-operations or delivery
contracts. First technological niches at the firm level and early innovation networks, in other
words, became more and more stable and well established.
From 2001/02 on, political support for the development and commercialisation of fuel cells
gained importance and processes of institutionalisation were observed place at different
levels including the formation of fuel cell associations or committees concerned with
standardisation and the set up of public research programs and funding schemes. As a
result, established niches and networks were increasingly complemented by a set of
institutions that supported key functions like the creation and diffusion of knowledge, the
provision of resources and the guidance of search (cf. Johnson, 2001). Innovation dynamics
in the case of stationary fuel cells can thus be interpreted as the development process of a
technological innovation system.
Interplay of expectations and innovation activities
Although neither expectation nor innovation dynamics can be explained just by their mutual
interplay, our findings point to a certain degree of coherence of the developments in the
discourse arena and the innovation arena. At the end of the 1990s, the emergence of a hype
about residential fuel cell systems went along with a striking increase of innovation activities
in the field. Innovation projects at that time were most likely co-motivated by the promises
and shared expectations, which had been adopted at the meso-level and transformed into
requirements for action. Public discourse, on the other hand, was deliberately influenced by
the discourse activities of prominent actors such as RWE, EnBW or EWE who sketched the
future of fuel cells in bright colours and played a dominant role in the media as they reported
on their projects or even ran advertising campaigns for fuel cells. Innovation activities in
combination with discourse activities, in other words, resulted in a feedback loop that fuelled
Furthermore, the orientation of strategic actors towards a common perspective of the future
(i.e. fuel cell heating systems are going play an important role in the future and will be
commercialised soon) contributed to the coordination of individual strategies and innovation
activities thus mobilising and guiding resource allocation. The shared perspective at the field-
level was strengthened by the frames available for the fuel cell technology at the macro-level.
In interaction with this cognitive prospective structure, social, material and institutional
structures such as partnerships and innovation networks, fuel cell stacks and prototypes or
research programmes were established that form the core of the emerging innovation system
and are thus crucial determinants of the future developments of the technology (Bender
Ruef, Markard 15 EASST 2006
2005). In our case, the newly established structures proved to be stable enough to endure
the phase of re-orientation and disappointment when the hyped expectations were not met.
The case of stationary fuel cells is an illustrative example for expectation dynamics that
basically resemble the models described in the literature but are still different. So far, we
have labelled our observation as a hype-disappointment dynamic characterised by a
moderate form of disappointment. In order to arrive at a more elaborated differentiation of
disappointments, the distinction of different levels of expectations, again, seems to be a
promising anchor point.
Stable and changing expectations at different levels
The three types of expectations identified in the discourse analysis correspond rather well to
the levels of expectations found by Van Lente (van Lente 1993, 182ff.). Indeed, expectations
at the micro-level are represented by the specific expectations in the case of fuel cells while
the so-called generalised expectations concern the meso-level of the field. The frames of the
fuel cell discourse then correspond to broad and diffuse expectations at the macro-level
which “are helpful to justify work on the technology ‘as such’” (ibid.).
Changes in the expectations reconstructed for the observation period (1993 to 2003) took
place at all three levels. However, the expectation dynamic described above - the coupled
hype-disappointment and promise-requirement-dynamic - was limited to the meso-level.
While the expectations at the micro-level developed in tight interaction with the meso-level, 17
the constantly positive framing of the technology at the macro-level served as some kind of
stabiliser for the lower levels of expectations. The frames reconstructed at the macro-level of
expectations show a broad range of interpretative schemes into which fuel cells were placed
(see Table 2). Note that none of them is used in a “negative” way, i.e. to criticise fuel cells or
to question their development. On the contrary, fuel cells were always ascribed a positive
role in the different scenarios used more or less frequently in the three periods. In other
words, a variety of frames that remained intact justified the promises and requirements at the
meso- and micro-level. In sum, we may argue that the broad and constant legitimisation of
the technology at the macro-level might have alleviated the effect of the non-fulfilled
expectations thus being responsible for the moderate form of disappointment which has
followed the hype.
Different types of disappointment
The change of expectations (see Table 1) at the micro-level illustrates the interaction with the
meso-level of expectations. Promises of firms and projects had been taken up at the field-level,
and furthermore, non-fulfilled expectations at the micro-level had led to a disenchantment at the
meso-level. Conversely, the meso-expectations formed an important orientation for actors of the
field, and the interpretation schemes of the macro-level were used as a resource by some actors of
the micro level (van Lente 1993).
Ruef, Markard 16 EASST 2006
On the basis of these findings, we suggest to not just regard and interpret some kind of
aggregated expectations and their potential impacts on innovation processes but to
differentiate expectations and their dynamics at the meso-level from those at the macro-level.
Expectations at these levels can develop largely independently because frames are rather
subject to external influences than to the innovation and discourse activities in the context of
the innovation field. Meso- and micro-level expectations, on the contrary, are expected to be
coupled more closely in most cases.
As a result, we may arrive at a classification like the one presented in Figure 2, where
positive and negative expectations are differentiated at each of the two levels. A hype is then
characterised by a combination of positive expectations at both levels. It is likely to support or
reinforce niche building processes in the innovation arena and the conversion of promises
into requirements as was observed in our case and elsewhere. A change of expectations
from positive to negative at the meso-level combined with a stable, positive framing at the
macro-level may lead to some kind of moderate disappointment, which can be attributed
mainly to the meso-level. This dynamic, which we labelled disenchantment, has been
observed for stationary fuel cells. It may affect the innovation arena in a way that causes a
re-orientation or displacement of innovation activities to other technological sub-fields or
application contexts. Established niches may be maintained, though not reinforced and new
ones may emerge.
If on the other hand positive expectations collapse at the macro level but remain intact at the
meso-level, the change is characterised by a lost of positive visions or the emergence of
negative ones. Such kind of discourse dynamic may be labelled as disillusionment.
Nanotechnology or ‘gene food’ might serve as empirical examples for such a development
where mostly positive framings at first became more and more challenged by negative
expectations about the unintended consequences of these innovations. Successful
technological developments in the innovation arena, at the same time, may keep the meso-
level expectations high or positive. As a result, established niches may remain or even grow
but new niches, or application context respectively, have a harder time to develop.
Finally, if positive expectations collapse at both levels the discourse may either be
abandoned or, even worse, continued with a general tone that is mainly negative. Such a
constellation can certainly be expected to have the most negative impact on innovation
activities. Promises may be put into question and niches won’t receive further support from
the discourse arena. Even more, established niches may be weakened and eventually break
down in the end.
Macro-level expectations (framing)
positive (switch to) negative
Ruef, Markard 17 EASST 2006
(inflated expectations, (lost of visions or emergence of
positive conversion of promises into negative ones)
**** Established niches may remain,
Support of niche creation or but development of new niches is
reinforcement of niches more challenging
(switch to) negative
(meso-level disappointment) (promises questioned,
Displacement of innovation Weakening or break-down of
activities, established niches may niches
Figure 2. Different types of hype-disappointment dynamics and potential
effects in the innovation arena
Of course, none of these fields necessarily represents a final state of development – neither
in the discourse nor in the innovation arena. Discourse and thus expectation dynamics can
well be influenced by deliberate discourse activities of actors in the respective field. In the
case of disillusionment, for example, it is most likely that actors who have a strategic interest
that established niches are maintained and grow will try to arrange their discourse activities
in a way that leads to a positive re-framing of the discourse. Similarly, actors in the field of
stationary fuel cells may work towards some kind of “re-enchantment”.
Methodological reflection and outlook
From a theoretical point of view, the differentiation of levels of expectations as suggested by
Van Lente (1993) was very fruitful for the explanation of the case as the expectations
reconstructed in the discourse analysis corresponded rather well to the three levels
described by the author. Even more, the differentiation of expectations at the macro- and
meso-level has turned out as a promising approach facilitating a more advanced analysis of
expectation dynamics as well. In further studies, it seems thus worth differentiating micro-,
meso- and macro-levels when examining expectations and speaking of expectation
From a methodological point of view, the conceptualisation of expectations as a discourse
turned out to be useful. Discourses exhibit similar characteristics as widely shared
expectations and offer thus an appropriate way to make the study of expectation dynamics
operational. The hermeneutic method of discourse analysis allows the reconstruction of
different types or levels of expectations with their contents and evolutions.
Moreover, media texts offer relatively simple access to the historical evolution of
expectations which are rather difficult to reconstruct from other data sources. Problems such
as the linearization of events by interviewed persons can thus be avoided. It would be useful,
however, to complement the discourse analysis by expert interviews. In the present case,
Ruef, Markard 18 EASST 2006
these could have provided in depth information on the effects of changing expectations at the
meso-level on expectations as well as innovation activities in firms or other organisations.
This is all the more valuable in empirical fields characterised by a low degree of controversy.
In such cases, media reports tend to be biased by positive events and negative results as
well as disappointments are not well covered and therefore more difficult to identify.
The relatively simple indicators used for innovation activities, however, provided a general
idea of the amount and participants of development activities taking place at a given
moment. It is clear, however, that a lot of innovation activities are missed if only publicly
available data are considered.
Further research will be needed to investigate how far the proposed differentiation of hype-
disappointment dynamics and the respective effects in the innovation arena can be
generalised. Do innovations systematically survive a hype if it is coupled with a promise-
requirement-dynamic and thus going along with institutionalisation processes? Can the
crucial role of positive frames at the macro-level for the emergence and protection of a niche
in case of meso-level disappointments be corroborated? Against the background of these
questions, cross-case comparisons of successful and abandoned innovations could therefore
further our understanding of the interplay of disappointed expectations and innovation
Ruef, Markard 19 EASST 2006
Bender, G., 2005. Technologieentwicklung als Institutionalisierungsprozess. Zeitschrift für
Soziologie 34 (3).
Bourdieu, P., 1985. Sozialer Raum und "Klassen". Suhrkamp, Frankfurt/M.
Brown, N., 2003. Hope Against Hype - Accountability in Biopasts, Presents and Futures.
Science Studies 16 (2), 3-21.
Brown, N., Rappert, B., Webster, A., Eds., 2000. Contested Futures - a sociology of
prospective techno-science, Aldershot.
Carlsson, B., Jacobsson, S., Holmén, M., Rickne, A., 2002. Innovation systems: analytical
and methodological issues. Research Policy, 31 (2), 233-245.
Eames, M., McDowall, W., 2006. Negotiating Contested Visions and Place-Specific
Expectations of the Hydrogen Economy. Technology Analysis & Strategic
Management 18 (3/4), in press.
Geels, F., Smit, W., 2000. Lessons from Failed Technology Futures: Potholes in the Road to
the Future. In: N. Brown, B. Rappert and A. Webster (Ed.), Contested Futures - A
Sociology of Prospective Techno-Science, Aldershot.
Jacobsson, S., Bergek, A., 2004. Transforming the energy sector: the evolution of
technological systems in renewable energy technology. Industrial and Corporate
Change, 13 (5), 815-849.
Jäger, S., 1999. Kritische Diskursanalyse - Eine Einführung. Duisburger Institut für Sprach-
und Sozialforschung, Duisburg.
Johnson, A., 2001. Functions in Innovation System Approaches. Nelson and Winter
Conference, June 12-15, Aalborg.
Keller, R., 2004. Diskursforschung - Eine Einführung für SozialwissenschaftlerInnen. Leske +
Keller, R., 2005. Analysing Discourse - An Approach From the Sociology of Knowledge.
Forum Qualitative Social Research 6 (3), 32.
Kemp, R., Schot, J., Hoogma, R., 1998. Regime Shifts to Sustainability Through Processes
of Niche Formation: The Approach of Strategic Niche Management. Technology
Analysis & Strategic Management, 10 (2), 175-195.
Konrad, K., 2004. Prägende Erwartungen: Szenarien als Schrittmacher der
Technikentwicklung. Edition Sigma, Berlin.
Konrad, K., 2006. The Social Dynamics of Expectations: the Interaction of Collective and
Actor-Specific Expectations on Electronic Commerce and Interactive Television.
Technology Analysis & Strategic Management 18 (3/4).
Markard, J., 2004. Strommarkt im Wandel - Veränderung von Innovationsprozessen am
Beispiel von Ökostrom und Brennstoffzelle, vdf Hochschulverlag, Zürich.
Roelofs, E., 1996. Vol Verwachting van de Brandstofcel. Universiteit Twente, Twente.
Strauss, A., Corbin, J., 1996. Grounded Theory - Grundlagen qualitativer Sozialforschung.
van Lente, H., 1993. Promising Technology, Enschede.
van Lente, H., 2000. Forceful Futures: From Promise to Requirement. In: N. Brown, B.
Rappert and A. Webster (Ed.), Contested Futures: a sociology of prospective techno-
van Lente, H., Rip, A., 1998. The Rise of Membrane Technology: From Rhetorics to Social
Reality. Social Studies of Science 28 (2), 221-254.
Weider, M., Metzner, A., Rammler, S., 2003. Die Brennstoffzelle zwischen Umwelt-, Energie-
und Wirtschaftspolitik. WZB, Berlin.