Europe's most congested airports are London Heathrow, London
Gatwick and Frankfurt. There, demand for take-off and landing
slots has exceeded the available supply for many years. Yet, in
the face of persistent growth in air traffic activities, these
airports have achieved remarkable increases in their capacity
to handle flights, despite retaining the same basic infrastructures.
In this these airports share a broad similarity with the Horndal
iron works in Sweden, which were discovered to have increased
their output in the 1930s despite retaining the same basic technologies.
This phenomenon became known as the 'Horndal effect', and is at
origin of the debate started by Arrow on 'learning by doing'.
Thus, we ask, is there a 'Horndal effect' apparent at Heathrow,
Gatwick and Frankfurt?
In investigating this growth of capacity we draw on detailed
case study work, but are concerned with four theoretically oriented
matters. Firstly, exploring innovation in services - which innovation
studies have largely neglected in favour of manufacturing. The
difference is non-trivial, as services such as air traffic control
are 'co-produced' - that is they are produced jointly by the provider
and the users. This means operations are dependent on co-operation
in the literal sense of the term - operating jointly. This brings
to the fore the importance of rules and routines, both in defining
operating practices and to reduce conflict over the allocation
of scarce resources. But co-operation in the second sense of a
willingness to work together for mutual gain is also important,
particularly to the innovation process, as co-production means
neither the provider nor the user can introduce unilateral changes
to the operating process.
This leads us into exploring the second issue, which is innovation
as a distributed process (between agents) of learning by interacting
and co-operating (in both senses of the term). Because airport
(runway) operations are 'co-produced' mutually beneficial innovations
aimed at providing additional capacity from the existing infrastructure
(i.e., runways) must be negotiated to a much greater extent than
is common in manufacturing, where one agent (the manufacturer)
generally owns and controls the processes by which 'its goods'
are produced. This in turn influences the pattern of innovation
that is acceptable, and blocks innovations that are not acceptable
to all of the co-production partners.
Thirdly, we emphasise the importance of procedural change, or
'soft innovation', as a complement to innovation through the adoption
or incorporation of capital equipment. In this sense we are attempting
to break-open the 'black box' of 'learning by doing'. Although
procedural changes are rarely visible to the layman, they are
clearly significant in finding efficiencies in the operation of
service systems such as airports. Moreover, and although we accept
that some 'automatic' learning-by-doing occurs (as a product of
experience), by understanding the processes of seeking improved
efficiencies through procedural change we show that not all 'learning-by-doing'
is an automatic, sub-conscious, activity.
A fourth interesting feature is the gradual transformation in
the sources of knowledge used for innovation. Initially, innovations
for capacity enhancement were heavily dependent on context dependent
operating knowledge, combined with knowledge of (or seeking knowledge
of) what changes to the existing operating procedures would be
acceptable to all of the co-production agents. Gradually, however,
the scope for such change has been reduced (i.e., used up), but
as the pressure to increase capacity has remained, attention has
turned to investigating the scope for change to the basic institutions
(such as aircraft separation distances) that were originally based
on experience rather than scientific research. This has meant
the incorporation of more formal methods (such as mathematical
modelling, including simulation modelling, as well as the use
of scientific methods to measure such thing as aircraft wake turbulence)
into the process of seeking possible improvements to the efficiency
of operations. This change in the sources of knowledge has been
accompanied by a change in the agents participating in the 'system
of innovation'. Finally, we reconsider the nature of 'systems
of innovation'. Rather than being national or sectoral, we consider
that they can be constructed around a problem, which may be contingent
and evolving, but which acts as a focusing device around which
'the system of innovation' is constructed.
[View
Paper] [Back
to CRIC Papers]
Centre
for Research on Innovation and Competition
