Implementing Electronic Trading at the New Y ork Stock Exchange -
A Case of Organizational C hange
Nicole Zimmermann
University of Mannheim
Industrieseminar
Schloss
68131 Mannheim
Phone: (+49) 621 181-1578
E-mail: zimmermann.nicole@ gmail.com
Abstract
Over the course of the last few years, the New York Stock Exchange quickly changed its trading
mechanisms from floor-based to mainly electronic trading. Researchers did not expect this
change to happen because they assumed the organization to be too inert. Organization theory
also has different views about the adaptability of organizations to change and about the question
whether the organizational management of the environment is the trigger for change. In order to
investigate this question this paper analyzes the New York Stock Exchange’s move towards elec-
tronic trading from a rational adaptation view, from a perspective that takes endogenous feed-
back mechanisms of culture and resistance into consideration, and from the perspectives of man-
agement inertia and management choice. It becomes obvious that only the combination of per-
spectives is able to create the observed behavior, but that different trajectories according to other
organization theories would also have been possible.
Key Words
Organizational Change, Adaptation, Punctuated Equilibrium, Organizational Inertia, Determin-
ism, Choice, New Y ork Stock Exchange
1 Introduction
When thinking of the New Y ork Stock Exchange (NY SE) one may picture a loud and boisterous
place, a crowd clustered around the trading booths, and much shouting for an eighth of a dollar of
price improvement. If one were to visit the New Y ork Stock Exchange now, it would render a
surprise for most visitors. Today, since many stock transactions are matched by computers in-
stead of people, the same trading floor is empty and quiet.’ Three of the five trading floors closed
down permanently, and the remaining two convey an almost idle atmosphere. The lively picture
of manual floor trading dominated the exchange until the end of 2005. Then electronic trading
rapidly replaced the lively floor trading by which orders were still matched by people so that in
2007 already more than 95 percent of trades and more than 85 percent of volume were handled
by computer programs. A drastic change has taken place.
The lively process of manual trading is shown in Figure 1. A buyer of e.g. 1000 shares of
the Sony Corporation gets in contact with his broker or brokerage firm which can be his bank, for
example. This firm is in contact with a floor broker, a person who is physically situated on the
NYSE trading floor. The floor broker then goes to the specialist who ‘makes the market’ in the
Sony Corporation securities and brings together supply and demand.” The frequent bidding of
many floor brokers around the specialist booth created the loud and boisterous trading process
that was symbolic for the New Y ork Stock Exchange. The market making activity of the special-
ists provides value by fulfilling the following responsibilities: Specialists are supposed manage
the auction process by maintaining a fair and orderly market, they execute orders from floor bro-
kers, match orders and work as catalysts by bringing together buyers and sellers. Additionally,
they are supposed to step in with their own capital to minimize imbalances in buy and sell orders
and to stabilize prices.’
Buyer Seller
Broker Broker
rN -
v v
Floor Broker Specialist Floor Broker
The Floor
Figure 1: Trade participants and interactions
Despite these valuable qualities of the specialists, the last decades showed a tendency towards
more and more electronic or automated trading in the US and worldwide securities markets. Here
a computer replaces the whole trading floor and assumes the matching of orders between buying
and selling brokers. This form of trading developed from the technological possibilities of the last
decades as well as from the rise in institutional investors.
Since the 1950s, institutions like pension and investment funds held more and more capital.
From 1950 until now, securities held by institutions increased from more than 5-fold.* For the
LEE, Stevennan, Ben: NY SE: Hooray for Market Volatility, in: Business Week Online (2007), Issued 5. Nov. 2007,
p. 25.
? At the New York Stock Exchange, the ‘market maker’ is called a ‘specialist’. “Specialists on the trading floor are
charged with maintaining fair, orderly and continuous trading markets in specific stocks by bringing buyers and
sellers together and, when circumstances warrant, adding liquidit by buying and selling stock for their own ac-
count.” (NY SE Group, Inc.: Annual Report of 2006, New Y ork chy 007a, p. 4).
3 Cf. NYSE Euronext: Types of Members, [Home > Products & Services > NY SE Equities > Trading Licenses >
Overview > Types of |. Members| accessed 4. Oct. 2007:
http://www. nyse.com/productservices/nyseequities/1167954368183.html.
4 CE NYSE Euronext: Facts and Figures, [NYSE Time Series Data]:
http://www. nyxdata.com/nysedata/D efault.aspx?tabid=115: Facts and Figures > Institutional Investors > Holdings
of corporate equities in the U.S. by type of institution, accessed: Aug. 21, 2008 .
New Y ork Stock Exchange the institutional investors are the greatest source of order flow.° Since
these organizations trade professionally, they often take advantage of tiny price differences and
engage in what is called arbitrage. Here, they also make use of computer programs to place,
change, and cancel orders so as to participate in the trading process at maximum speed.
Due to the increase in available technology, in institutional customers and their growing
demand for automation and speed, many of the NY SE’s competitors implemented electronic trad-
ing much earlier. NASDAQ, for example, was founded in 1971 as an electronic communications
system and later purely electronic stock exchange and the NY SE’s largest competitor today. The
Cincinnati exchange implemented an electronic system already in the late 1970s.° Researchers as
well as industry experts predicted this trend to continue.’ The electronic brokerage effect sup-
ports this view because it states that direct exchange between agents will replace intermediaries,
and electronic markets will fulfill the intermediaries- functions.® Clemons and Weber expect
technology-triggered changes in the securities market, but they also admit that there will be resis-
tance from the exchanges’ most powerful members.’ Threatened intermediaries may also form
cooperative groups attempting to stop their disintermediation, like banks do with S.W.LF.T. to
retain their primacy in money transfer.’°
With a history of more than 200 years the New York Stock Exchange is a very long-
established organization of manual floor trading. The NY SE is often even regarded as a national
symbol. Its trading process was boisterous, but as an organization it portrayed both stability and
strength and the image of a conservative gentlemen’s club. Lucrative specialist positions used to
be passed on from fathers to sons, making it difficult for outsiders to get their feet in the door of
this ‘gentlemen’ s club’.!' With the image of conservatism and strength, the New Y ork Stock Ex-
change was for long a symbol for the US economy as the largest and most prestigious stock ex-
change of the world. With the rise of institutional investors and electronic trading there were
many voices affirming the New Y ork Stock Exchange was becoming outdated, because it was not
expected to change and adapt. The organization was even described as a non-innovative socialist
collective.’ Already in the 1970s, many researchers firmly believed in the decline of the New
Y ork Stock Exchange because they thought it to be both getting out-dated and unchangeable.'?
5 Cf. Blume, Marshall E., Jeremy J. Siegel and Dan Rottenberg: Revolution on Wall Street: The Rise and Decline of
the New Y ork Stock Exchange, New Y ork, London 1993, pp. 105 and 108.
® Cf. Seligman, Joel: The Transformation of Wall Street: A History of the Securities and Exchange Commission and
Modern Corporate Finance, rev. Ed., Boston 1995, p. 521.
" Cf. Feldman, Stuart: Electronic Marketplaces, in: Internet Computing, IEEE, Vol. 4 (2000), No. 4, a
Clemons, Eric K. and Bruce W. Weber: Information Technology and Screen-Based Securities Trading: Pricin
the Stock and Pricing the Trade, in: Management Science, Vol.'43 (1997), No. 12, p. 1695; and Picot, Amold,
Christine Bortenlanger and Heiner R6hri: The Automation of Capital Markets, in: Journal of Computer-Mediated
Communication, Vol. 1 (1995), No. 3, electronic article without page number.
® Cf. Malone, Thomas W., Joanne Y ates and Robert I. Benjamin: Electronic Markets and Electronic Hierarchies, in
Allen, Thomas J. and Michael S. Scott Morton (Ed.): Information Technology and the Corporation of the 1990s:
Research Studies, New Y ork, Oxford 1994, pp. 67-68.
* Cf. Clemons and Weber: Information Technology and Screen-Based Securities Trading, 1997, p. 1706.
1° Cf. Picot, Amold, Christine Bortenlanger and Heiner Rohrl: Organization of Electronic Markets: Contributions
from the New Institutional Economics, in: The Information Society, Vol. 13 (1997), No. 1, p. 115.
1 For an economic analysis of cooperative club membership see Buchanan, James M.: An Economic Theory of
Clubs, in: Economica, Vol. 32 (1965), No. 125.
? Cf Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 256.
'S Cf Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 253-259. oo
Abolafia, for exampl , describes how specialists perceived decline and the threat of extinction around 1990. (Cf.
Abolafia, Hee .: Making Markets: Opportunism and Restraint on Wall Street, Cambridge, MA, London
), pp. 1o0-Lol.
p. 95; and
Still in 2003, the then CEO Richard Grasso confirmed the idea of continuity, strong traditions,
stability, even inertia with regards to the future of the exchange:
“\.. change at the NY SE is likely to be incremental at best -- with the interests of his
seat holders remaining a matter of paramount importance. Elimination of the ex-
change's floor-trading system, as urged by some exchange critics, ... is not about to
happen. The specialists are the exchange... [...]. Specialists and floor brokers are
likely to continue to hold sway at the exchange, for the simple reason that they own it
and dominate its corporate culture."
This creates a puzzle. On the one hand researchers expected stock exchanges to automate and
adapt to technological possibilities and institutional investors’ demands. On the other hand, no-
body expected the New Y ork Stock Exchange to do the same; they expected it to remain inert and
finally fail. Apparently the New Y ork Stock Exchange radically transformed itself in the recent
past, proving wrong those who predict the gradual adaptation of electronic trading as well as
those who predicted the NY SE’s cessation. In the organizational theory literature this duality of
adaptability and inertia actually represents one of the most important fields of research. Organiza-
tional researchers heavily discuss whether organizations are adaptable at all, and if whether they
change incrementally or radically. A further question of high relevance is whether changes are
determined deterministically by environmental developments or whether there is room for choice
and management action.
In order to solve the puzzle, different views on organizational change will be presented in
more detail in the following. Developments in the environment of the New Y ork Stock Ex-
change, the US securities market, will also be elaborated as well as researchers’ opinions about
the future of this market. Within this context, the transformation of the NY SE will then be ana-
lyzed in a case study. In order to understand how this change came about, different aspects will
be taken into consideration— that of adaptation to environmental pressures, of culture and resis-
tance, as well as that of management. Neither of these is able to fully explain the transition that
the NY SE went through, but by their combination they give a more complete account of the dy-
namics of how the changes unfolded.
2 Deterministic and Deliberate Views on Organizational C hange
Researchers portray organizational change as an adaptation process vs. a radical transfor-
mation as well as a result from environmental forces vs. management action. The literature is
divided here; assumptions range from what is called the organizational imperative to the envi-
ronmental and technological imperative. The organizational imperative regards organizations as
transformable in accordance with management aims, other than the deterministic environmental
imperative which assumes that either organizational inertia renders the organization unchange-
able so that environmental selection determines its fate or that the environment forces the pace
and the timing of adaptations. Different theories thus postulate very different drivers of organiza-
tional change and different outcomes.
M Weiss, Gary: The $140,000,000 Man, in: Business Week (2003), No. 3849, Issued 15. Sep. 2003, pp. 90-92.
“The problem is that different schools of thought tend to focus only on single sides of
issues and use such different logics and vocabularies that they do not speak to each
other directly.”*°
As a consequence, Astley and Van de Ven regard it essential to create awareness of the assump-
tions and biases that underlie a single theory.” There is demand for consolidation and integration
of the different schools of thought for making progress in the organization theory field. In order
to reduce this either-or thinking of environmental, technological or organizational drivers of
transformations, multiple views will be described as well as attempts of their integration in order
to get a better understanding of organizational change. Child adopts the view that different or-
ganization theories or paradigms are irreconcilable in their philosophical underpinnings, i.e. in
their world view, school of thought, their assumptions, but they are not incommensurable when
they are applied to the study of organizations.” In order to utilize each theory’s contributions, the
deterministic views of rational adaptation as an adaptation to the environment, of the population
ecology approach as a non-adaptation to developments, the organizational imperative of the de-
liberate strategic choice approach as well as the mixed punctuated equilibrium view will be dis-
cussed.
In the first half and well into the second half of the 20" century organization theory was
dominated by views of rational adaptation which may also be the most prominent view of organ-
izational change. Many theories fall under the category: the classical theories of scientific man-
agement, Fordism, and Weber's bureaucracy, industrial organization economics which is an eco-
nomic theory of the rational behavior of organizations in markets, contingency theories, and re-
source dependence theories. In their core these theories assume human rationality and rely on
humans as decision makers according to the image of man of the self-interested and utility maxi-
mizing homo oeconomicus or economic man.'® Based on this cognitive and behavioral assump-
tion of humans and organizations, organizations are assumed to adapt to the environment. Han-
nan and Freeman describe rational adaptation as “designed changes in strategy and structure of
individual organizations in response to environmental changes, threats, and opportunities.” ” Due
to the environment-triggered adaptation Mellahi and Wilkinson as well as Child call these theo-
ries deterministic.’ Deliberate management choice is assumed to play a minor role in rational
adaptation theories.
'S Astley, W. Graham and Andrew H. Van de Ven: Central Perspectives and Debates in Organization Theory, in:
Administrative Science Quarterly, Vol. 28 (1983), No. 2, p. 246.
16 Cf. Astley and Van de Ven: Central Perspectives and Debates in Organization Theory, 1983, p. 270.
Cf, Child, John: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment: Retrospect
and Prospect, in: Organization Studies, Vol. 18 (1997), No. 1, p. 44.
*8 concept of man is the theory’s mental model about the nature of human beings. Already in the late 19th century,
a concept of man developed which describes humans as rational and self-interested utility maximizers. It is a de-
terministic concept of man. The term homo oeconomicus was formed by Wilfredo Pareto in the beginning of the
20" century, but according to Stengel, the idea goes back to Adam Smith, Thomas Malthus, David Ricardo, and
John Stuart Mill. Ge for example, Pareto, Vilfredo: Manual of Political Economy: Transl. from the French ed,
of 1927, London [u.a.] 1972, The original edtiion Manuale d economia polititica was published in 1906; and
Stengel, Martin: Psychologie der Arbeit, Weinheim 1997, p. 78.
1° Cf. Hannan, Michael T. and John Freeman: Structural Inertia and Organizational Change, in: American Sociologi-
cal Review, Vol. 49 (1984), No. 2, p. 150.
20 Cf, Mellahi, Kamel and Adrian Wilkinson: Organizational failure: a critique of recent research and a proposed
integrative framework, in: International Jounal of Management Research, Vol. 5/6 (2004), No. 1, p. 23; Child:
Strai gic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, Rg 45; and Child,
{oan Tganizational structure, environment and performance: The role of strategic choice, in: ociology, Vol.
, pp. 8 and 10.
The assumption of organizational inertia
Coming back to the New Y ork Stock Exchange, most researchers predicted the failure of
this organization because they expected it to be inert and unable to undergo necessary organiza-
tional change. Inertia expresses the idea that organizations do not change as quickly or com-
pletely as some groups want them to change in order to be adequately adapted to the environ-
ment. In the view of Kieser, Hegele and Klimmer as well as Hannan and Freeman, inertia sym-
bolizes a lack of adaptation ability what means that an organization is not able to react to a
change in circumstances, even if it tries.”4
Opinions about the usefulness of inertia differ. Larsen and Lomi as well as Hannan and
Freeman regard inertia as an important outcome of daily operations which gives the organization
reliability and continuity. They admit that inertia prohibits organizational change, but they also
emphasize that continuity is advantageous for experience and performance.” Change attempts
render the formerly experience obsolete.*? But in the change management literature, inertia is
also often referred to as a cause of the failure of innovation initiatives.” Adequate change is de-
sired. Already Darwin proposed that surviving species are those who are the most adaptive to
change.”
Population ecology
Organizational inertia is a central aspect of the population ecology perspective. Hannan and
Freeman question organizational rationality and they also regard analyzing the whole organiza-
tional population more useful than the rational adaptation theory’s focus on the organization only.
They assume an ecological-evolutionary process of the manifestation of strong inertial forces
within populations that prevent major changes.”° Due to this process the investigation at the
whole population level is crucial for understanding organizational change. Following the ecologi-
cal-evolutionary perspective on whole populations Hannan and Freeman call their point of view
population or organizational ecology. Remembering that for the New Y ork Stock Exchange most
researchers predicted the organization to be so unchangeable that it would fail, this organization
theory seems worth looking at.
Population ecology addresses the question of whether organizations change through adapta-
tion or replacement.”’ The theory links to evolutionary biology by its assumption is that a popula-
tion evolves by Darwinist selection. This also means that changes cannot be planned and enacted
by organizational members. The whole population becomes the unit of analysis in the evolution-
21 Cf. Kieser, Alfred, Comelia Hegele and Matthias Klimmer; Kommunikation im omanisatorischen Wandel, Stutt-
gart 1998, p. 123 and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 151.
2 Cf. Amburgey, Teny L., Dawn Kelly and William P. Bamett: Resetting the Clock: The Dynamics of Organiza-
tional Change and Failure, in: Administrative Science Quarterly, Vol. 38 (1993), No. 1, p. 52; and Hannan and
Freeman: Structural Inertia and Organizational Change, 1984, p. 149.
?3 Cf. Larsen, Erik R. and Alessandro Lomi: Representing change: a system model of organizational inertia and ca-
pabilities as dynamic accumulation process, in: Simulation Modelling Practice and Theory, Vol. 10 (2002), No. 5-
, p. 276; Larsen, Erik R. and Alessandro Lomi: Resetting the clock: a feedback approac! to the dynamics of or-
ganisational inertia, survival and change, tn: Joumal of the Operational Research Society, Vol. 50°(1999), No. 4,
p. 408; and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 153.
4 See for example: Tushman, Michael L. and Elaine Romanelli: Organizational Evolution: A Metamorphosis Model
of Convery nce and Reorientation, in Cummings, Larry L. and Barry M. Staw (Ed.): Research in Organizational
Behavior, 7. Vol., Greenwich, CT 1985, p. 177
°5 Cf. Darwin, Charles: The Origin of Species, London 1971.
26 Cf. Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 149.
27 Cf. Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 150.
ary theories, and an organization is regarded as a “cohesive organism”.** The seminal papers of
Hannan and Freeman were crucial for the development of population ecology.”° The theory fo-
cuses on the organization-environment relations with particular focus on the effects of environ-
mental change on an organization or population. Perrow emphasizes that an evolutionary per-
spective and the ecology point of view often go hand in hand. This has the effect that it strength-
ens organizations which are adapted best and makes maladapted organizations disappear.*” Due
to significant organizational inertia, this theory has little room for adaptive processes within or-
ganizations.*’ In contrast to the approach of rational adaptation which was described earlier, the
evolutionary population ecology explains changes mainly by the failure of maladjusted and the
creation of new organizations.” Inertia can be stronger than decision makers' attention and can
block change completely.
Population ecology is a point of view that is also part of what could be called the environ-
mental imperative. Astley and Van de Ven argue that population ecologists conclude that the en-
vironment has primacy. Since population ecology focuses its analysis on the environmental level
only, it also assumes internal dynamics and strategic choice, and an organization’s effect on its
environment to be of minor importance. Later Hannan and Freeman attenuate their earlier the-
ory by acknowledging that minor organizational changes
“occur frequently and that organizations sometimes even manage to make radical
changes in strategies and structures. Nevertheless, we argue that selection processes
tend to favor organizations whose structures are difficult to change. That is, we claim
that high levels of structural inertia in organizational populations can be explained as
an outcome of an ecological-evolutionary process.”
Strategic choice
Child challenged both the rational adaptation theory as well as population ecology for their
deterministic explanation of change as either an adaptation process or for the assumption that
inertia renders organization immutable. Knowing that the New Y ork Stock Exchange finally un-
derwent a major change a new explanation seems worth looking into. Child criticized that ra-
tional adaptation theories and population ecology explained change deterministically as a result
only of environmental constituencies.*° This criticism addressed methods and content of re-
search. Child pointed out that the common analysis leaves underlying processes to be inferred
and clouds the complex network of direct and indirect interrelationships. The deterministic envi-
ronmental and technological theories do not see a sphere of influence of those with power in the
28 Cf. Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 66-67.
29 Cf. Hannan and Freeman: Structural Inertia and Organizational Change, 1984; and Hannan, Michael T. and John
Freeman: The Population Ecology of Organizations, in: The American Journal of Sociology, Vol. 82 (1977), No.
ioe Cf. Perrow, Charles: Complex Organizations: A Critical Essay, New Y ork [u.a.] 1986, pp. 208-210
31 Cf, Adler, Paul S. and Bryan Borys: Materialism and Idealism in Organizational Research, in: Organization Stud-
ies, Vol. 14 (1993), No. 5, p. 663; and Hannan and Freeman: The Population Ecology of Organizations, 1977.
°? Cf. Mellahi and Wilkinson: Organizational failure: a critique of recent research and a proposed integrative frame-
work, pp. 21-41; and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 150.
33 Cf, Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 155.
34 Cf. Astley and Van de Ven: Central Perspectives and Debates in Organization Theory, 1983, pp. 257-258.
55 Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 149.
3° Cf. Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 45.
organization.°” Many theories give too much weight to constraints and too little to choice. They
underestimate the dominant coalition’s power in choosing organizational structure and criteria of
performance.*® Regarding technology Child applies the same criticism that technology may re-
strict work in the short-run, but that decisions will finally be carried out by those in control of the
organization.** In the deterministic theories the distinction between environmental reality and its
perception and evaluation is often overlooked so that perceptions and interpretations of the envi-
ronment are frequently mixed with environmental qualities and characteristics. Child states that
the exercise of choice requires a prior evaluation of the environment. “” Actions do not follow
from environmental conditions per se.
25 years after the development of the strategic choice theory Child slightly amends his earlier
theory and puts greater emphasis on the mutuality of deliberate action and environmental con-
straint, i.e. on voluntarism combined with determinism. He focuses on the role of agency and
choice in organizational analysis, the nature of the organizational environment, and the relation-
ship between organizational agents and the environment." Child admits that also in the strategic
choice perspective the organization-environment fit or the performance indicators represent an
important input for decision-makers.”
Overall, the strategic choice perspective reveals the importance of choice as a tool of the
organization's decision makers. It concentrates on the actions of leading groups in organizations
and their role in shaping the organization. The role of agency and choice, the nature of the organ-
izational environment, and the relationship between organizational agents and the environment
are central.* Strategic choice makes no restrictions on the pace and extent of transformation; it
leaves open how changes take place: incrementally and adaptively, radically, or else.
Punctuated Equilibrium
Tushman and Romanelli propose a holistic theory of organizational evolution that combines ele-
ments of population ecology and strategic choice and that has a clear suggestion to how changes
take place.
“Patterns of organizational evolution are characterized by periods of convergence
punctuated by reorientations leading to the next convergent period. These cycles are
driven by the emergence of tension between organizational and institutional forces for
inertia and competitive, technological and legal pressures on performance which are
mediated by the perceptions and decisions of executive leadership.” “*
Due to the assumption of convergence periods punctuated with reorientations this theory is also
called punctuated equilibrium. It links to all of the aforementioned theories and combines aspects
of each of them.
3” See Child: Organizational structure, environment and performance, 1972, pp. 1-2.
38 Cf. Child: Organizational structure, environment and performance, 1972, p. 19.
°° Cf. Child: Organizational structure, environment and performance, 1972, p. 6.
4° Child: Organizational structure, environment and performance, 1972, pp. 4-5.
a See pid Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 43 and
“ Cf. Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 48.
8 Cf. Child: Strategic Choice in the Analysis of Action, Structure, Gryanizations and Environment, 1997, p. 43; and
Child: Organizational structure, environment and performance, 1972.
“ Tushman and Romanelli: Organizational Evolution, 1985, p. 181.
“Our punctuated equilibrium model of organization evolution borrows from ecologi-
cal, [rational] adaptation and transformational approaches to evolution. Environments
do actively select out firms which do not align themselves with environmental con-
straints. [... Those organizations that evolve over a product class life cycle are those
that initiate and successfully implement strategic reorientations. Environments select
out those firms which either do not reorient, choose inappropriate reorientations
and/or can not implement strategic reorientations. Finally, for successful organiza-
tions, the period between strategic reorientations is characterized by incremental,
adaptive change, as structures, systems and processes are more finely tailored to the
firm's strategic orientation.”
In Tushman and Romanelli’s view population ecology, industrial organization economics, strate-
gic management and organization theory literatures all regard organizational competence as re-
lated to the fit between the organization’s strategic orientation and its internal and external envi-
ronmental conditions.*° There are strong linkages between structure, processes, and values. The
organization needs to be both adapted to its environment and internally consistent. Periods of
convergence bolster normative developments in organizations, e.g. the development of “values,
beliefs, and ideologies at individual, group and organization levels of analysis” *”. This is a rein-
forcing mechanism which is supported by respective recruiting, socialization, training, and lead-
ership behavior.*®
In internally consistent organizations reorientations face political challenges. “Coalitions of
interests in large organizations are made up of stable, self-perpetuating groups who have a vested
interest in the status quo, and who make consequential decisions slowly and with frequently bi-
ased and distorted information.”“’ This means that interest groups may make change attempts
difficult and that information processing may be biased.
The punctuated equilibrium theory of Tushman and Romanelli states that in convergent pe-
riods pressure for change accumulates in the environment. But due to the accumulation of inertia
within the organization, the perception of and reaction to these pressures is biased. The executive
team has great importance in this process. Its homogeneity creates bias and hinders the percep-
tion of outside pressures. Executive leadership mediates the two opposing forces which are iner-
tia and pressures for change in the environment. Due to the biased perceptions of executive lead-
ership, reorientations will most frequently occur after a sustained performance decline or with an
outside successor in leadership.°° The performance decline as well as when an outsider takes of-
fice reduce the commitment to the current strategic orientation which makes receptive for outside
pressures and often triggers radical change.
“Given the pervasiveness of inertial forces, both perception and action are usually
triggered only by sustained low performance, a major shift in the distribution of
45 Tushman and Romanelli: Organizational Evolution, 1985, p. 214.
46 See Tushman and Romanelli: Organizational Evolution, 1985, p. 188.
“” Tushman and Romanelli: Organizational Evolution, 1985, p. 192.
“8 Cf. Tushman and Romanelli: Organizational Evolution, 1985, p. 192; referring to Argyris, Chris and Donald A.
Schon: Organizational Leaming: A Theory of Action Perspective, Vol. 1. Reading, Mass. [u.a.] 1978
“9 Tushman and Romanelli: Organizational Evolution, 1985, p. 192. Tushman and Romanelli refer to Downs (1967),
Olson (1982), and Wilensky (1967).
5° Cf. Tushman and Romanelli: Organizational Evolution, 1985, p. 211.
power, and/or organizational crises. These perceptions and the response to changing
strategic requirements are shaped by the characteristics of executive leadership”
Reconciliation
Overall organization theorists do not agree on the question of how organizations change—
adaptively and continuously or seldom and then radically— and the question whether the envi-
ronment determines transformations or whether there is room for deliberate management choice.
Rational adaptation theories expect a deterministic adaptation of organizations to environments;
population ecology as a further deterministic theory expects the environment to select out non-
adapted organizations. Strategic choice theory then also allows for deliberate management action.
Then Tushman and Romanelli as well as Astley and Van de Ven as well as Adler and Borys
called for a reconciliation of change theories. Researchers have similarly different views concem-
ing the evolution of stock exchanges. Research in the development of stock exchanges predicted
and predicts a strong prevalence of environmental forces. The electronic market hypothesis ex-
pects a strong decline of intermediaries; and stock exchanges are expected to automate. The pic-
ture that researchers portray is that of rational adaptation to the environmental and technological
forces in the market. For the New Y ork Stock Exchange in particular predictions were different.
The NY SE was described as inert, traditional, and it was expected to remain unchangeable and
eventually die. A case study of the New Y ork Stock Exchange's transformation will be analyzed
with respect to organization theories, to the prevalence of adaptation vs. inertia, as well as deter-
minism and choice.
3 Method Mix
Method
The case study of the New Y ork Stock Exchange will be analyzed with a mix of methods. Here,
journal as well as newspaper articles, NY SE time series data, an ethnographical analysis of the
NYSE blog, four interviews as well as system dynamics modeling of the issue will help the clari-
fication of the case as well as its analysis. There has been no study of the transformation of the
New Y ork Stock Exchange from an organizational perspective so far. Research on the change is
sparse and the existing articles rather focus on finance aspects. There is popular science research
by Gasparino who studies the years preceding the NY SE’s change. His book offers valuable in-
sights, but he explained the observed behavior or incidents by the personalities of important indi-
viduals, on the individual's play of politics as well as on events.*” My approach radically differs
as I analyze the underlying forces of the same behavior. The focus is on how the behavior
emerges from these forces, and how dynamic forces allow for the observed events. Behavior is
often looked at as a course of events, but here it will be important what are the factors allowing
for certain types of events rather than the details of the events themselves.
In order to test the impact of these effects on the implementation of electronic trading, the
system dynamics method is used. Here, a formal model reveals the structure of the underlying
system. At the same time it aims at reproducing real world behavior as well as well as at illumi-
nating behavior in different scenarios.
5! Tushman and Romanelli: Organizational Evolution, 1985, p. 204.
ne oe Gag arin0, Charles: King of the Club: Richard Grasso and the Survival of the New Y ork Stock Exchange, New
on i
10
Validation
System dynamics models should be validated for (1) plausibility and relevance, (2) behav-
ioral consistency, and (3) dominant structure analysis.** First, plausibility analysis tries to gain
face validity for causal relationships, equations, and model structure in general by reference to
data. This data can be numerical, qualitative or it can be experts’ mental models. This is done to
build a model that includes the areas which plausibly explain the reference behavior. Second, by
a behavioral consistency analysis one makes sure that the behavior of the model matches behav-
ior or schematic patterns of behavior observed in the real world.” It is particularly concerned
with the relevance of parameters and non-linear functions in the model. Third, the loop domi-
nance analysis helps gain confidence in the model by providing a causally coherent story and
explanation.» The author of this paper will follow these steps and combine the plausibility analy-
sis— the search for face validity of causal relationships and model structure— with the case de-
scription. Dominant structure analysis of the important loops will be done continuously, and be-
havioral consistency will be tested both continuously as well as in a separate analysis of model
sensitivity.
The three-step organization of the case study with the extemal environment perspective, the
endogenous perspective of stakeholder pressures and the management perspective serve to fulfill
the requirements of the relevance and behavior analysis. It leads to the relevant model boundary,
revealing which pressures are relevant, but also how they endogenously create reactions and how
feedback is created in the system.
4 Change at the New Y ork Stock Exchange
From Oct. 6, 2006 to April 2, 2007 the New Y ork Stock Exchange completed the rollout of
the Hybrid Market that is a mostly automated market with elements of both automation and floor
trading. It combines specialist obligations and floor broker expertise with the speed of electronic
markets in order to be liquid and provide better market quality than purely electronic ex-
changes.*° In June 2008 the New York Stock Exchange introduced further changes. It even
turned away from the “old” specialist system which it had protected for so long, not completely
abandoning it, but abandoning some of the specialist responsibilities and advantages and making
them designated market makers.°’ This happened in response to falling market share and the ac-
knowledgement of customer demands.*®
53 This tripartite division of model validation was expressed by Mohammad Mojtahedzadeh (Mojtahedzadeh,
Mohammad: Personal Communication 2008, Apr. 2008).
a cr ptema, John D.: Business Dynamics: Systems Thinking and Modeling for a Complex World, Boston [u.a.]
, p- 860.
55 Mojtahedzadeh: Personal Communication, 2008.
3 Cry SE Regulation, Inc.: Information Memo: Hybrid Market Implementation - Phase IV, No. 07-12, 29 Jan.
://apps.nyse.com/commdata/PubInfoMemos.nsf/A llPublishedInfoMemosNyseC om/85256FCB 005E19E88525
icrosoft%20W ord%20-%20D ocument% 20in%. -pdf, accesse ug.
57 According to NY SE Group, the Designated Market Maker has similar responsibilities to the specialist and will:
-- Be rewarded for quoting, which benefits other traders, the DMM, and NY SE as a whole;
-- Be obliged to maintain an orderly market in assigned stocks;
-- Not have an advance "look" at incoming order information;
-- Be required to quote at the national best bid or offer for specified percentages of the time;
-- Also be required to facilitate pice discovery at the open, close and in periods of significant imbalances;
-- Provide liquidity based on a Capital Commitment Schedule that will be programmed into the Display Book but
will receive no order information;
-- Have economic incentives that will be transparent and based on performance, which will be reviewed periodi-
11
Change at the New Y ork Stock Exchange started when the former CEO of the Exchange,
Richard Grasso, got involved in a scandal about his outrageous pay and retirement package which
was considered inappropriate for a CEO of a non-profit company. The scandal broke after Grasso
tried to cash in his 140 million retirement money. When he left the Exchange, Grasso said that
changes at the NY SE would be incremental, major changes would be unlikely due to the corpo-
rate culture that was committed to the specialists and the floor trading system. °° Grasso finally
was replaced by Thain, the former co-president of Goldman Sachs. He brought change into the
management team and the organization. With him came a major shift in the way the New Y ork
Stock Exchange does business. In 2001, as Figure 2 shows, the NY SE under Grasso had already
enabled purely electronic trading for small orders. This meant that these small orders were
matched by a computer instead of having the human intervention of the specialist. Since it con-
cerned small orders only, the way of doing trading changed incrementally only. In 2004, how-
ever, the change was somewhat more substantial when the limit on the size of orders that could
be handled electronically got removed. Then, over the period of 2006 to 2007 the Exchange im-
plemented the Hybrid Market. This is a form of trading that allows the customers to trade elec-
tronically, and it enables them to choose whether their order gets executed by a computer or
whether their order is routed to the specialist.
Fraction of NYSE Competitors Enabling Some E-Trade
Fraction of —
largest foreign 0.75 | :
NYSE el lectronic
competitors that °5 bey at the
enable some = g 95,
e-trade
0 — E-order size limit removed
1960 1975 1990 2005 2020 2035 2050
Time (Year)
“fraction of 22 largest foreign NYSE competitors that enable some e-trade"
Figure 2: Reference mode
Institutional customers make use of the electronic trading possibilities." As a consequence, the
trading floor is quiet now. Additionally, the trading floor space was reduced.°! Most surprisingly,
the NY SE inflicted damage on its floor brokers and specialists who until quite recently had great
influence in the Exchange’ s board and were the symbol for the institution.
The automation of trading was not the only change that Thain introduced. He also arranged
a merger of the New Y ork Stock Exchange with Archipelago, a purely electronic trading plat-
form, and one with the European Euronext exchanges. Additionally, the NY SE demutualized and
cally;
-- Be able to trade on parity with other orders.
sd aagiocnett Aaron: NYSE Plans to Revise Specialist-Trader Rules, in: Wall Street Joumal - Eastem Edition
, P.O.
5° Cf. Weiss: The $140,000,000 Man, 2003, pp. 90-92.
®° Cf. McGeehan, Patrick: Next to Downsize On Wall Street? The Exchange Floor, in: New York Times (2007),
Issued 23. Sep. 2007, p. 37.
©! Cf, McGeehan: Next to Downsize On Wall Street, 2007, p. 37.
12
is a publicly traded corporation now. Y et, since trading represents the NY SE’s core activity, only
the transformation of its way of trading will be analyzed.
How did this change away from purely floor trading towards mostly electronic trading
come about? The reasons for this change seem obvious. Many of the NY SE’s competitors im-
plemented electronic trading much earlier, e.g. the Cincinnati Exchange in the late 1970s. Never-
theless, concerning the implementation of automated trading, it took the NY SE 25 years to follow
suit. It thus seems interesting to analyze what let to the NY SE’s adoption of electronic trading
and why it happened at this point of time. It will also be interesting to see whether the exchange
could have followed a different path and implemented electronic trading earlier or maybe not at
all.
Thus, there are three areas of interest for the analysis of the NY SE’s shift towards elec-
tronic trading. First, it will be important to know about the forces that pressured for the adoption
of electronic trading following the rational adaptation point of view. Second, the pressures
against this adaptation process are vital. This may reveal a picture of organizational inertia fol-
lowing the population ecology point of view. Here, analyzing the impact of the specialists’ cul-
ture and their cohesiveness and resistance will be important since they are the symbol of the Ex-
change. Third, the influence of the management on the evolution of the New Y ork Stock Ex-
change may be important as well. The effects of inertia as well as the effect of market share will
be analyzed.
5 Model Structure and Basic Behavior
Figure 3 shows an overview of the system dynamics model, and outlines its scope and boundary.
While the characteristics of the market, like its time to execution (trading speed) and extent of e-
trade are external influences, the following mechanisms work endogenously. The management’ s
decision on the extent of electronic trade is central. Then, depending on the relative characteris-
tics of the market, the effects of customers’ accumulated dissatisfaction with the NY SE’s relative
speed is considered in the pressure that this creates on the decision about e-trade. On the other
hand, non-institutional customers value the specialist system, and the two forces work against
each other. This would be a view of rational adaptation to environmental forces. Floor firms
themselves value the floor system, too. The pressure for the floor system that this valuation cre-
ates is part of the analysis as well. Taken together these forces may represent what Hannan and
Freeman call the population ecology model. Finally, it will be considered how the management’ s
commitment to the current strategic orientation influences the management's decision about the
extent of e-trade.
13
PRESSUR FOR E-TRADE
- From institutional customers.
Deriving from relative NYSE speed
CHARACTERISTICS
OF MARKET
- Market speed
- Extent of e-trade in market
(-) Pressure by customers
Market
MGMT DECISION ON share
EXTENT OF E-TRADE (4) Inertia and nn
- font Pa
Implementation time Customer! <RATEGIC ORIENTATION
4 th orientation
- Customer orientation
- Pcvd relative pressure
- Inertia
- Market share
(-) Resistance by stakeholders
(+) Cultural pressure by stakeholders
PRESSURE FOR FLOOR TRADE
- From non-institutional customers
Deriving from market quality
- From floor firms
As resistance by stakeholders
As cultural pressure by stakeholders
Figure 3: Sector diagram
In order to understand these relations in more detail, causal loop diagrams of the system dynam-
ics model will be presented in the following section.
5.1 Causal Details and Behavior
5.1.1 Adaptation View
Figure 4 shows an excerpt of the model, and it points at an important driver of change. Since the
1950s, more and more securities were held by institutions, over the last 60 years, this number
increased from about 7% to well more than 50%.°2 According to Blume et al., “[bly the 1970s,
institutional investors had replaced individual investors as the dominant force in the market.”
The percentage of trading that is done by institutions instead of individual investors rose from
20% in 1950 to 75% in 1975." These large institutional customers are mutual and pension funds
and investment banks, for example. The Exchange's largest two customers— the leading invest-
ment and wealth management banks Goldman Sachs Group and UBS AG—each account for
about 10% of the trading.
of institution (Oct. I,
® Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 108.
° Cf. Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 105.
® Cf. NY SE Group, Inc.: Annual Report of 2005, New Y ork City 2006, Note 5— Related Parties and other Relation-
ships, p. 139, http: fir nyse.com/phoenix.zhtmi2c=129145&-p-trol-reportsA nnual (15. Nov. 07).
© Cf. NYSE website: Facts and Figures > Institutional Investors > Holdings of corporate equities in the U.S. by type
14
information technology wt. on speed
(and resulting tmarket time + vs. price
to execution)
securities held by
institutions
+
No. of
Institutional
Customers
Figure 4: External changes and the valuation of speed
It was and is crucial for these large institutional customers to gain direct access to the New Y ork
Stock Exchange’s trading floor because it gives them faster and less costly access.
Furthermore, as a second influence from outside, computer technology developed and
paved the way for electronic trading. This led to the formation of Electronic Communication
Networks (ECNs) where NY SE-listed stocks can be traded as well as to the foundation of
NASDAQ, which from then on has tried to take away market share in NY SE-listed stocks. For
example, in the first page of the “NASDAQ Facts’-webpage, NASDAQ states to be the mem-
ber® routing more order flow to the NY SE than any other member. They also maintain that their
“trading systems are the faster, more efficient, transparent, better way to trade NY SE-listed secu-
tities.”
Since large institutional customers trade professionally, they often make use of tiny price
differences and engage in what is called arbitrage. Here, they also make use of computers and
computer programs which issue the orders in order to automate the trading and to issue orders
faster. This kind of trading is called algorithmic trading. A bolafia describes a similar mechanism
for the bond market and its new forms of trading.” Since algorithmic trading on the side of the
customer is done by computers and since it uses minimal price differences in arbitrage, the trade
execution time becomes highly important. As Figure 4 reveals, customers put more weight on the
speed of order execution than on the spread which represents the price difference between the bid
and offer. The spread is a measure for the quality of the price that the customer receives or pays,
and small spreads are desired.
How this creates a balancing feedback loop can be seen in Figure 5. One major factor de-
termining the time to execution is the fraction of electronic trading.” With all possible technical
°° A “member” is an entity owning at least one “seat” of the NYSE. A “seat” is an expression for the right to directly
trade on the NY SE floor. They are still referred to as “seats” because in the early days of the NY SE, members sat
in assigned chairs. (NY SE Glossary: “Seat”, www.nyse.com/glossary/Glossary.html (Dec. 7, 2007).
6’ NASDAQ website, NASDA' Facts, Home > Nasdaq Corporate > Nasdaq Facts,
http://www.nasdaq.com/reference/nasdaq_facts.stm (Dec. 7, 2007)
°§ Cf. Abolafia: Making Markets, 1996, pp. 5-6.
For 2 detailed stock and flow diagram of the above-mentioned processes please refer to A ppendix I: Institutional
‘ustomers.
°° Tt needs to be said here that the fraction of e-trade in the market follows the access to information technology. This
is the variable which drives e-trade in the remaining market and the markets time to execution.
More details can be seen in
15
support, manual floor trading involves a specialist decision and cannot be done in less than a few
seconds. Electronic trading is fast and a market classifies as a fast market if the time to execution
is one second or less. Depending on the relative fraction of e-trade and resulting time to execution
of the NYSE and the market, institutional customers build up dissatisfaction with the trading
speed. This creates a balancing loop (B) which Appendix II reveals in full detail in a full stock
and flow diagram. Depending on their level of dissatisfaction, institutional customers exert pres-
sure for the implementation of electronic trading because going electronic is a way to particularly
increase the speed of trading. When e-trade gets implemented the relative time vs. the NY SE’s
competitors improve and dissatisfaction decreases.
information technology —— wt. on speed
(and resulting tmarket time + vs. price
to execution)
+
i +| Dissatisfaction
Gore ® with Time to
execution Execute
B
pressure for +
capital held by fraction of] + — e-trade
institutions e-trade
+
No. of
Instiutional
ustomers
Figure 5: Customer pressure for e-trade
The move towards electronic trading has consequences which the balancing feedback mechanism
in Figure 6 reveals. Here non-institutional customers like individuals who directly invest in secu-
rities pressure for the retention of the specialist system.”’. They have no interest in fast execu-
tions since they do not engage in arbitrage and algorithmic trading, but they have a high prefer-
ence for receiving the best price. With more and more trades being executed electronically the
specialist is not needed any more, and the market quality from specialist participation is lessened
in line with the reduction in this activity that has occurred. Market quality is a measure for the
specialists’ tasks of price improvement, stabilizing prices, reducing volatility, and providing li-
quidity. Specialists participate in floor trades and provide liquidity particularly for lower-volume
stocks.
According to NY SE Group, volatility increased only slightly since the implementation of
the hybrid market because they were able to increase liquidity simultaneously.” But the analysis
of the NY SE weblog revealed that a group of customers is extremely dissatisfied with the missing
price improvement and demands more specialist participation.
“\.. We need the specialists to add liquidity, and stabilize the price's. The value the
specialists add is what makes the NY SE the best place to trade and its what makes the
7 A ppendix III reveals in more detail what pressure for the specialist system by customers depends on.
” Cf. Ellul, Andrew, et al.: Order dynamics: Recent evidence from the NY SE, in: Journal of Empirical Finance, Vol.
14 (2007 1 No. 5,pp. 637 and 658. See also Appendix III: Customer Valuation of , which reveals this mechanism
in more detail.
” NYSE Group, Inc.: NY SE Completes Hybrid Market Phase III Activation, NY SE Group, Inc., [News Release],
accessed 11. Dec. 2007: http://www.nyse.com/pdfs/HybridPhaselV 1.24.07.pdf, p. 1-2.
16
exchange different from the other exchanges. The new CEO has to get them involved
if the NY SE has any chance of getting better for their customers.”
“Y our customers want the ability to execute blocks, be price improved, and not be
charged an arm and a leg to do it. We also want the specialists to create a better qual-
ity market [... (pe
A perceived inadequacy of market quality thus creates customer pressure for more floor trade.
This closes the balancing feedback loop.
information technology
(and resulting tmarket time
to execution)
wt. on speed.
+ vs. price
+
+|Dissatisfaction
rel, time to a
execution with Time
B
pressure for 4
capital held by fraction of|_+ e-trade
institutions e-trade
pressure for
" - floor
0. of specialist
insight participation e #
Customers 7
market quality + customer
from specialist > pressure for floor
Figure 6: Customer pressure for floor trade
This model represents an adaptation view of organizational change since the NY SE can be re-
garded as adapting to the external forces of the market. The customers exert pressure for more
electronic trading if the speed is worse than in the market so that it closely follows the trend in
the market. Figure 7 shows that only when change starts to get implemented, non-institutional
customers perceive the disadvantages in market quality and start pressuring for the retention of
the old system. This creates a slow down of the otherwise simultaneous implementation of e-
trade in the market and the NY SE.
73 NY SE Exchanges Blog: Comment by pony, Dey, owner of a day trading firm,
http://exchanges.nyse.com/archives/2007/11/men_at_work.php (25. March 2608).
™ NY SE Exchanges Blog: Comment by Tony Dey, owner of a day trading firm,
http://exchanges.nyse.com/archives/2007/06/duncan.php (25. March 2008).
7
Selected Variables
1 Dmni
1 Dmni
0,75 Dmnl
0,75 Dmni
0,5 Dmnl
0,5 Dmnl
0,25 Dmnl
0,25 Dmnl
0 Dmnl
0 Dmnl
1960 “1968~ 1976 1984 1992 2000 2008 2016 2024 2032 2040
Time (Year)
“fraction of e-trade in remaining market" : s adaptation Dmni
"NYSE Fraction of E-Trade" : s adaptation Dmnl
Figure 7: Adaptation view
Yet, this adaptation does not yet give a realistic picture of reality. Change happened later and
more radically. As Figure 6 indicated already and as the following figures will show, there is not
only pressure for e-trade but also pressure for the floor system.
5.1.2 Culture and Resistance View
When specialists and floor brokers noticed that a shift to electronic trading was imminent, some
specialists started to protest. According to an informant, one of them still wears around 20 stick-
ers on his suit showing his opposition to electronic trading and to the way it was implemented.
Although he used to be a member of the board, his opposition is rather powerless now. Special-
ists make money from brokerage commissions, from the spread between the bid and ask price for
their stocks, and from trading for their own account.”° It is thus not surprising that specialists and
floor brokers showed resistance against the implementation of e-trade which would reduce the
spread. It is not only the profitability but also the specialist’ participation and employability that
decreases with electronic trading, and thus both lead to resistance. Thus, based on the lower par-
ticipation (employability) and profitability some resistance occurred, and its balancing loop is
shown in Figure 8.”°
* Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 40.
ae Appendix IV: Resistance Pressure shows a detailed stock and flow diagram of the two resistance effects, profit-
ility and employability.
18
information technology §=____———— wt. on speed.
(and resulting tmarket time + vs. price
to execution)
+
i + Dissatisfaction
rel. time to
execution win Time t °
pressure -.
capital held by fraction of — e-trade
institutions e-trade
io ae for
4 - floor
0. oF specialist
ins itghat participation B ie
Customers #
market quality + customer
from specialist > pressure for floor
B resistance
:s > pressure for floor
. profitability of:
floor
Figure 8: Resistance pressure from floor
The specialists themselves plus other companies which mainly receive their revenue from trading
on the floor enforce this pressure. In Figure 8 a reinforcing feedback loop shows how specialists
value their own, culture depending on their ability to make profits and on their merits in providing
market quality.”’ Their profession used to be a “license to make money”. Specialists and floor
brokers could maintain their cartel for fixed commissions until the first half of the 1970s.” But
afterwards commissions diminished.”* This creates a reinforcing loop, but since the specialists’
valuation of themselves increases faster than it decreases, specialists in general only slowly let
loose of their culture; and for a long time they still pressure for their system, even if profitability
and their ability to provide market quality has long been diminished.
7] Further details can be found in A ppendix V: Floor Valuation of Floor Culture.
8 Ch lata Making Markets, 1996, p. 109, and Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993,
p. 25.
7° Cf. Gasparino: King of the Club, 2007, p. 48.
19
information technology = __—_—____—> wt. on speed.
(and resulting tmarket time + vs. price
to execution)
4
rel, time to
*)
Dissatisfaction
execution with Time 0
B
pressure for 2
capital held by fraction of|_+ e-trade
institutions e-trade
pressure for
% - floor
0. of specialist +
insWeighat participation B x
Customers “
market quality + customer
from specialist > Pressure for floor
B resistance
zi > pressure for floor
. " OR
profitability of- Floor Valuation off
floor Floor
t
Figure 9: Cultural pressure from floor
A remarkable factor that helped shape the pressure for the floor is the power that floor firms
have. Already in the beginning of the 1990s a specialist commented: “It was always a great busi-
ness because we eamed a commission on whatever we acted as agent on. There is tremendous
pressure on us to give up that business now.”®° In order to stay profitable and be able to provide
sufficient liquidity many specialist firms consolidated. Others went out of business. Since they
have to put their own or the company’s capital at risk when they work as specialists, many of
them exited with the capital they had accumulated earlier. The number of specialist firms at the
New Y ork Stock exchange decreased from about 150 in 1960 to 40 in 1990 and to six firms to-
day, making the whole specialist community less powerful. This decline in power was is trig-
gered by the growth of institutional investors. Additionally, floor firms’ power depends on the
market quality that specialists provide. Picot, Bortenlanger, and Rohri state that their power is
defined by the volume they attract to the exchange.” Volume is influenced by market quality. A
decrease in market quality from the introduction of electronic trading thus also diminishes power.
Figure 10 reveals the reinforcing mechanism which this creates.®? The power of floor firms is
multiplied with the resistance and cultural pressure and thus mediates the floor’s pressure for the
old system.
jae Quoted in 4 in fbolatia: Making Markets, 1996, p. 131.Similar statements of specialists can be found on pp. 135, 141-
and
®1 Cf, Picot, | eaters and Rohn: The Automation of Capital Markets, 1995
®2 A detailed view is provided in Appendix VI: Power of Floor Firms.
20
information technology
(and resulting tmarket time
to execution)
wt. on speed
+ vs. price
rel. time to
execution
capital held by fraction o'
institutions e-trade
+
No. of specialist B
Institutional participation
ustomers 4
market quality +
B
pressure for 4
+,
e-trade
Dissatisfaction
Time to
Execute
pressure for
customer
from specialist > pressure for floor
floor
#
resistance
e for floor
R
Floor Valuation of
Floor
Power of Floor
Firms
Figure 10: Power of floor firms
The model now incorporates pressures for e-trade as well as pressures for the manual floor sys-
tem. When it is simulated, the implementation of electronic trading is shaped by these pressures
for and against e-trade. Figure 11 reveals that the pressure that comes from the valuation of the
floor culture and from resistance is influential and that it can defer and diminish the implementa-
tion of electronic trading. Nevertheless, it still is not able to demonstrate a behavior similar to the
real course of events. At the New Y ork Stock Exchange e-trade was not implemented until the
early 21st century. In the model which currently includes an adaptation view plus culture and
resistance this happens much earlier.
NYSE Fraction of E-Trade
1960 1968 1976 1984 1992 2000 2008 2016
Time (Year)
"NYSE Fraction of €-Trade" : s adaptation
2024
2032
2040
"NYSE Fraction of E-Trade" : s ex. + end. pressures
Figure 11: Culture and resistance
So far it was assumed that the management implemented everything that the joint pressures of
customers and stakeholders called for. Although institutional investors had called for the automa-
tion of trading, the NY SE did not automate order execution until the year 2000 when it intro-
duced e-trade for small orders, Before, there had been no signs for a full rollout of e-trade within
the next few years. Since the radical shift did not base on sudden changes in the forces for floor
and e-trade, the role that management had for electronic trading is also worth consideration.
5.1.3 Management View
The New Y ork Stock Exchange itself regards its own success, market share and complacency as
drivers for its decision making and attention to customers.
“Part of being very successful for a very long time and having a large market share [is
that] the New Y ork Stock Exchange did become complacent. We have to be receptive
to change. We have to give our customers what they're looking for."
Being successful and not having to change created complacency which is inertia on the one hand
and a bias in attentiveness. The organizational change literature discusses their influence as well.
Hannan and Freeman as well as Tushman and Romanelli address how inertia renders organiza-
tions inalterable. Long periods of convergence strengthen the development of values, beliefs and
ideologies.™. This reinforcing mechanism is supported by respective recruiting, socialization,
training, and leadership behavior.®
The New Y ork Stock Exchange did not change the basic principles of its way of trading for
more than 100 years. During this convergence period there was much opportunity for the institu-
tionalization of routines and the growth of consistency. New employees seldom came with new
ideas from outside, instead they were “grown from within the organization”. The resulting inertia
led to the observed complacency. For Van de Ven and Poole inertia and persistence maintain old
organizational forms, and they are a major force that antagonizes the reinforcing variation loop.
This reinforcing loop is shown in Figure 12. Inertia enhances the commitment to the current stra-
tegic orientation and antagonizes the move towards electronic trading. But once some e-trade gets
implemented the commitment to the strategic orientation declines and more change becomes pos-
sible.
® Ewing, Jack: When CEOs Talk to Each Other, World Economic Forum in Davos, Switzerland, 2005, in: Business
Weak Gal tne from 31. Jan. 2005, http://www. husinessweek. com/bwdaily/dnflash/jan2005/nf20050131 5572.htm
pril 2
® Tushman Le — Organizational Evolution, 1985, p. 192.
®5 Cf. Tushman and Romanelli: Organizational Evolution, 1985, p. 192; referring to Argyris and Schén: Organiza-
tional Learning, 1978
5° Cf. Van de Ven, Andrew H. and Marshall Poole: Explaining development and change in organizations, in: The
Academy of Management Review, Vol. 20 (1995), No. 3, pps 514 and 518.
The full stock and flow diagram is show in A ppendix VII: Management Decision-Making.
22
information technology
(and resulting tmarket time
to execution)
wt. on speed
vs. price
+
rel, time to Dissatisfaction
Pre
+,
with Time to
execution Execute
B
‘a a pressure for 4
commitmen i 7
capital held by to strategic R fraction of] + e-trade
institutions orientation Jy C-trade
pressure for
‘4 - floor +
specialist B
\ ugitenal participation *
ustomers
Es
market quality + customer
from specialist pressure for floor
resistance
e for floor
Floor Valuation of Power of Floor
Floor : Firms
Figure 12: Management and organizational inertia
The above mechanism only explains a path-dependent behavior, but no change of path, and does
thus not explain the sudden NY SE transformation. Child suggests how a performance crisis may
trigger a change in the dominant coalition, i.e. in the inert decision-making group itself or its be-
liefs.
“For example, conditions of crisis occasioned by the failure of an organization to
achieve a level of performance in its environment that can meet the expectations of its
member groups are likely, according to strategic choice analysis, to give rise to a
radical change in the composition of the ‘dominant coalition’ [...] with a probable
consequent change in the policies and practices instituted for the organization.”*”
A performance crisis diminishes the confidence in the current strategic orientation and. Tushman
and Romanelli as well explain that organizational evolution is driven by the tension between or-
ganizational inertia on the one hand and competitive, technological and legal pressures on per-
formance on the other. Additionally this struggle of forces is mediated by the perceptions of ex-
ecutive leadership. ®
5.1.4 Market Share
From late 2006 to early 2007 the New Y ork Stock Exchange completed the rollout of the
Hybrid Market with both manual and automated elements. It combines floor broker access, spe-
cialist and now designated market maker obligations and automation in order to have a fast and
high quality market." In June 2008 the New Y ork Stock Exchange introduced further changes.
87 Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 63.
88 Cf, Tushman and Romanelli: Organizational Evolution, 1985, p. 181.
80. cone SE Regulation, Inc.: Information Memo: Hybrid Market Implementation - Phase IV, No. 07-12, 29 Jan.
23
They have not gone into effect yet, but the NY SE will tum its specialists into designated market
makers with similar obligations. This happened i in response to falling market share and the ac-
knowledgement of customer demands."”
The NYSE market share used to range between 75 and 80 percent of trading in NY SE-
listed securities. Trading volume and market share fell; in September 2007 the NY SE had an av-
erage trading volume of 1.4 billion shares per day with a market share of trading in NY SE-listed
securities of 44 percent. In June 2008 the NY SE on average traded 1.2 billion shares per day, and
its market share had fallen to 29 percent. **
When the management perceives the market share to be inadequate, they lower their com-
mitment to the strategic orientation and additionally re-direct their attention away from floor
firms to customers. Customer orientation increases and with it increases perceived pressures for
e-trade from customers; the perception of pressures for the floor diminishes. More e-trade fol-
lows. The causal link from market share to the NY SE decision making can be seen in Figure 13.
Reorientations thus are “[... - | radical and discontinuous changes driven by the opposing pressures
of performance and inertia.
information technology
(and resulting tmarket time
to execution)
wt. on speed.
vs. price
+
+
i +|Dissatisfaction
rel, time to 5
execution wi Lime to
spread ig
B B
a pressure for 2
it commitment” e
securities held by to strategic R fraction of] + e-trade
institutions + orientation Jy) C-trade
pressure for
% - floor
No. of specialist B
Institutional participation #
ustomers +
market quality + customer
from specialist > pressure for floor
B
resistance
e for floor
Floor Valuation of Power of Floor
Floor : Firms.
Figure 13: Market share
This figure also shows the determinants of market share itself, so that the relationship
“market share > NY SE decision on e-trade > market share” becomes an endogenous feedback
http://apps.nyse.com/commdata/PubInfoMemos.nsf/A ll PublishedInfoM emosNyseC om/85256FCB005E19E88525
ei BSB ot ‘6/$FILE/Microsoft% 20W ord%20-% 20D ocument%20in%2007- pdt accessed 20 Aug. 2008.
°° Cf. Lucchetti: NY SE Plans to Revise Specialist-Trader Rules, P. C4.
®! Cf. Chapman, Peter: Rule Changes: NY SE Euronext Bets on Specialists to Revive! Broken Floor, Traders Maga-
zine, accessed 20. Aug. 2008: hip: ://www.tradersmagazine. cope TOL ITH Lh
© Tushman and Romanelli: Organizational Evolution, 1985, p. 197.
24
mechanism. Market share rises by a fast relative time to execution, i.e. when the fraction of e-
trade is high, which creates a balancing feedback loop. It also rises with high market quality
which creates a reinforcing mechanism, indicating that a move towards electronic trading first
reduces market quality, then market share and calls for more electronic trading. Therefore, the
implementation of e-trade has ambiguous effects on market share, but due to the rising impor-
tance of institutional customers the effect of speed prevails.
A further determinant of market share is the liquidity loop in Figure 13.°° Available liquid-
ity reduces the spread between the bid and ask price. Low spreads (i.e. high price quality) in tum
attracts more market share. Or as Duncan Niederauer, the now CEO of the New Y ork Stock Ex-
change expresses it: “liquidity begets liquidity”. ** Since high order flow has the effect of reduc-
ing the spread, this creates a reinforcing feedback loop which works positive for the exchange
that has the greatest market share.
The consideration of the management's decision-making creates a rather radical shift from
manual to electronic trading, as can be seen in the red line number 3 in Figure 14. Taking inertia,
market, share, the commitment to the current strategic orientation, and customer orientation into
account, creates a long period during which electronic trading is not implemented. The observed
delay could thus be explained.
NYSE Fraction of E-Trade
Dmal
o
re
19 976 1984 1992 2000 2008 2016 2024 2032 2040
Time (Year)
"NYSE Fraction of €-Trade” : s adaptation
"NYSE Fraction of €-Trade" : s ex. + end. pressures
"NYSE Fraction of €-Trade” : s w.0. scandal w.o regulation
"NYSE Fraction of €-Trade" : s base run
Figure 14: Adaptation, Culture and Resistance, and Mgmt
Figure 15 reveals the different forces leading to this shift. When the pressure for e-trade (line 3)
starts exceeding the pressure for the floor (line 4), a very small portion of e-trade gets imple-
mented. The following quick implementation comes about due to the abrupt decline of market
share. This drop leads to a drastic rise of the management’s customer orientation which then
shifts attention from pressures for the floor to pressures for automation.
%3 A ppendix VIII: Market Share shows the stock and flow view in more detail.
“NY SE Exchanges Blog: Duncan Niederauer, NY SE Euronext President and Co-COO and Head of U.S. Cash Mar-
kets, now CEO, http://exchanges.nyse.com/archives/2007/08/compensation.php (25. March 2008).
25
Pressure and E-Trade
1 Dmnl
100 pressure unit
0,5 Dmnl
50 pressure unit
0 Dmnl
0 pressure unit
1960 1968 1976 1984 1992 2000 2008 2016 2024 2032 2040
Time (Year)
NYSE Fraction of E-Trade* : s wo. scandal w.0 regulation Dmal
NYSE Market Share :sw.o. scandal wo regulation Dmnl
“total pressure forecrade from customers’: sw.c. scandal w.o regulation —3——— 3 ressure unit
total pressure forthe floor: s wo. scandal w.o regulation pressure unit
Figure 15: Underlying forces
As the reader may have noticed, the current forces are able to explain the general behavior pattern
of a radical change from floor to e-trade, but they do not get the timing right. Two additional ex-
temal effects will be analyzed in their effect on the timing, the Grasso scandal and a regulatory
change. As mentioned earlier, the long-term CEO of the NYSE, Richard Grasso, got involved
into a acandal about his retirement package. “Toward the end of his career, Mr. Grasso seemed to
consider himself bigger than the institution he ran, leading to a series of missteps that attracted
the scrutiny of the Big Board's secretive practices, [... ].” When information about his pay pack-
age became public it led to a scandal. The support of the formerly aligned management quickly
diminished, and it ended in a vote in which many of Grasso’s former “buddies” asked him to re-
sign.®° The leave of the CEO and his replacement were events that diminished the organizational
inertia suddenly and from outside. About two years after the scandal the US Securities and Ex-
change Commission implemented a new regulation that had effects on the reinforcing “liquidity
begets liquidity” loop of market share. The National Best Bid and Offer Rule (NBBO rule) re-
quires exchanges to be connected and send orders to the exchange currently offering the best
price. In 2005 the Securities and Exchange commission suggested and later introduced the Regu-
lation National Market System (Regulation NMS).°” This regulatory change reduces market share
for two reasons. First, it requires markets to trade fast in order to participate in the national mar-
ket system that direcly sends orders to the trading venue with the best price. Second, even if an
exchange quotes the best price, after the change in regulation it is not the full order any more that
gets executed at the exchange offering the best price, but often just the first fragment of the order.
This weakens the reinforcing liquidity loop, and both effects diminish NY SE maket share. Since
a falling market share diminishes the confidence in the strategic orientation and enhances cus-
tomer orientation, the Grasso scandal and the introduction of Regulation NMS are responsible for
that the radical shift towards electronic trading occurs about three years earlier.
$8 Morgenson, Gretchen: The Fall of a Wall Street Ward Boss, in: The New Y ork Times (2003), Issued 19. Oct.
pel.
°° Cf. Gasparino: King of the Club, 2007, pp. 275-281.
°7 Securities and Exchange Commission (SEC): Regulations NMS; Final Rule, in: Federal Register, Vol. 70 (2005),
No. 124, p. 37496.
26
Pressure and E-Trade
1 Dmnl
100 pressure unit
0,5 Dmnl
50 pressure unit
0 Dmnl
0 pressure unit
1960 1968 1976 1984 1992 2000 2008 2016 2024 2032 2040
Time (Year)
"NYSE Fraction of E-Trade" : s base run Dmni
NYSE Market Share : s base run Dmni
“total pressure for e-trade from customers" : s base run, 33333 pressure unit
total pressure for the floor : s base run pressure unit
Figure 16: Underlying forces
As can be seen in Figure 16, the base run of the model now to a great extent matches the refer-
ence mode, and the radical automation of the NY SE trading starts in 2007. This run reproduces
the real events and behavior as closely as possible and marks the reference point for further
model analysis It also becomes obvious that, due to the regulatory change, the market share does
not recover to the former high standard any more.** Taken as a whole, the decline in inertia and
market share decline made the New Y ork Stock Exchange rethink its past strategic orientation,
focus on the customer, and it even triggered a change in corporate culture.
“Within the company, a for-profit, publicly traded corporation since last year, the two
[NY SE executives Niederauer and Leibowitz] are trying to transform what was tradi-
tionally an inward-looking, exchange floor-dominated culture into an outward-
looking, customer-focused culture. Instead of running the organization for the benefit
of specialists, for example, Niederauer and Leibowitz say ther ey are trying to satisfy the
people that deliver the order flow- [i.e.] the broker-dealers. ual
The feedback loops also tell a story. In the beginning the floor valuation by floor firms domi-
nates. Additionally market share keeps the management inattentive to customers. The customer
dissatisfaction loop is not able to affect greater changes in the way of trading. Only when the
feedback from the relative speed of the NY SE to market share gets strong, this loops triggers a
rethinking in strategic orientation, and the falling inertia also triggers more and more change. The
process then stops when the system reaches its perceived goal.
6 Presentation and Analyses of Model Behavior
Now, after the structure of the model as well as its behavior have been revealed, it will be impor-
tant to gain some more confidence in this model. This is done with the help several tests which
are applied to the model’s structure and behavior. Here, one of the benefits of the system dynam-
ics method also becomes apparent: unlike econometrical models, we are not only able to test
°8 Current simulation results match real data quite closel ely. Simulation results of 45 and 33 pero for for the vears 2006
and 2008 match the observed data of 44 percent and 29 percent for the years 2007 and 2008 quite closely.
°° Chapman, Peter: Men At Work, in: Traders Magazine, Vol. 20 (2007), No. 274, Issued Nov. 2007.
27
model behavior, but model structure as well. This structural plausibility analysis was mainly done
in parallel with the model description as causal relationships were explained with either literature
or data from the case study. Additionally the stepwise development of the model structure re-
vealed that adding new structure from the adaptation point of view via the culture and resistance
view to the management view created new behavior. The simulation runs in Figure 14 showed
this explicitly. Concerning the boundaries of the system, it can thus be said that it was necessary
to take these three views into consideration.
A validation test for behavioral consistency illustrates that the model also shows sensible
behavior under extreme conditions. If there are no technological developments that allowed for e-
trade, no e-trade develops in the model. If there are no institutional customers at the NY SE, no e-
trade gets implemented there either. This helps to gain confidence in the model because in reality
electronic trading started to get implemented at the New Y ork Stock Exchange only after both
forces— institutional customers and technological possibilities for e-trade— were present.
The initial questions concerned adaptive vs. inertial and vs. management forces. The case
study of the New Y ork Stock Exchange reveals how inertial forces, endogenous pressures from
culture and resistance, and the management impinge on the adaptation process. Figure 17 illus-
trates the importance of endogenous forces. Concerning this trend of the market, two different
assumptions are modeled: a linear growth of electronic trading in the remaining market as well as
a more realistic exponential growth with a limit. These two different market behaviors help ana-
lyze how closely the adoption pattern of e-trade in the remaining marked shaped electronic trad-
ing within the New Y ork Stock Exchange. The resulting model behavior for the automation of
trading does not significantly differ from the base run behavior of the model. The development of
electronic trading in the market is necessary for the New Y ork Stock Exchange to automate, but
how the automation is executed and how the implementation process unfolds within the NY SE is
subject to endogenous management and stakeholder dynamics.
Linear Development of E-Trade in Market
Oma
°
a
1960 1968 1976 1984 1992 2000 2008 2016 2024 2032 2040
Time (Year)
“fraction of e-trade in remaining market: s linear techn
“fraction of e-trade in remaining market" : s base run
"NYSE Fraction of E-Trade" : s linear techn
"NYSE Fraction of E-Trade" : s base run
Figure 17: Linear development of e-trade in market
In the following, the endogenous aspects will be analyzed more closely, particularly the effects of
each of these determinants on the implementation of e-trade, pressures for and against e-trade,
and the market share.
28
6.1 Sensitivity Analysis
Sensitivity analyses are conducted to analyze model robustness and the relationship between
structure and behavior.'”° In the following, first, the sensitivity for the pressure of institutional
customers, for the pressure of non-institutionals from culture and resistance, and for management
attitudes will be presented.
6.1.1 Analysis of Culture and Resistance Aspects
The upper left graph in Figure 18 shows that the extent to which e-trade gets implemented not
sensitive to variations in the reference level of resistance, i.e. to variations in how much resis-
tance is created by a certain inadequacy of employability and profitability.’" Different strengths
of resistance create more or less pressure for the specialist system (right). But it does not affect
the automation (left) since this happens only after the New Y ork Stock Exchange started imple-
menting e-trade and already shifted its attention towards institutional customers.
sens resistance ens resistance
50% 75% 95% 9 100% 50% 75% 995% (100%
"NYSE Fraction of E-Trade” total pressure for floor from floor
1 100
0,875 875
075 75
0,625 625
05 50
0,375 375
0,25 25
0,125 125
“9601970 1980 1990 2000 2010 2020 2030 2040 ‘i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
Figure 18: Sensitivity for resistance
The pressure from the floor is also determined by the power of floor firms. Floor firms gain
power from the participation of the specialist in trading and his positive effect on market quality.
They loose power due to the rising number of institutional traders who like to bypass the floor
intermediaries. When the sensitivity of the model is checked for changes of the reference power
of floor firms the fraction of e-trade shows a very low sensitivity, too. It seems that a group of
forces determines the implementation of e-trade, so that changes in the cohesiveness of culture
and in cultural pressures may have more effect since they operate earlier.
Therefore, as a next step, the sensitivity of the model is checked for three variables at the
same time—for the degree of cohesiveness of floor firms, the reference cultural pressure per
floor firm and reference resistance— and here the output graphs show more variation. Figure 19
suggests that when all three of them work together the effects on the implementation of e-trade
are much greater. For the case this reveals that electronic trading could have been implemented
much earlier had the three variables been lower. If the cultural cohesiveness and the resistance
10 Cf. Sterman: Business Dynamics, 2000p. 830.
10! Tn a sensitivity analysis, the model is run 200 to 5000 times while a certain parameter or a group of parameters is
changed. An overview of parameter values in the base run and ranges in the sensitivity analyses can be found in
Appendix IX: Institutional Customers, Program Trading, and Institutional Owners..
In a sensitivity analyses, it is assumed that the Regulation NMS comes into effect, but the effect of the Grasso
scandal is left out..
29
had been stronger, it would have been implemented at the same time as in the base run, but it
would have been implemented to a somewhat lesser extent since the pressure for the floor system
would have been stronger.
The effect of feedback becomes obvious in the lower part of Figure 19. The different extent
of the pressures for the floor (lower right) affects the fraction of e-trade (upper left) which then
also affects the pressure for e-trade (lower left) and market share (upper right).
sens floor floor
50% 75% (95% 100% ay 50% 75% MN 95% (100%
"NYSE Fraction of E-Trade" NYSE Market Share
1 0.8
0,875
0,75
0,625
07
06
05
04 —
03
02
01
1960 1970 1980 1990 2000 2010 2020 2030 2040 ‘i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
sens floor ens floor
50% 75% (NNN 95% 9 100% 50% 75% (95% (100%
“total pressure for e-trade from customers" total pressure for floor from floor
80 200
70 175
60 150
50 125
40 d 100
30 75
20 50 |
10 ta 25 |
9601970 1980 1990 2000 2010 20202030 2040 ‘lo6o 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
Figure 19: Sensitivity for resistance, culture, and cohesiveness of floor firms
The sensitivity analyses show that floor firms with low power, culture and resistance enable ear-
lier automation. If floor firms show strong power, culture, and resistance, they do not have much
direct influence on the point of time when electronic trading gets implemented; it happens around
2007. But the analysis showed that in a situation where cultural cohesiveness and resistance are
high and/or where the affected stakeholders are bound to the firm, they can reduce the extent to
which electronic trading gets implemented.
6.1.2 Analysis of Management Aspects
In the following, it will be tested how sensitively the system reacts to changes in the time to im-
plement trading, i.e. the organization’s general openness to change, as well as the management’ s
general responsiveness to customers. In the case of general openness to change, the reference
time to change trading is varied. A long reference time means that the current fraction of elec-
tronic trading is mainly kept whereas a short reference time leads to a quick adaptation to pres-
sures from outside. Figure 20 shows a sensitive reaction of the extent or pace to which e-trade
gets adopted, not to the timing when the launch starts. As the bottom left picture shows, the pres-
sure for e-trade from institutional customers reacts very sensitively to lower and higher levels of
30
e-trade. The floor shows some reaction, too. Depending on the extent of e-trade market share can
vary for some time.
sens time trading fens time trading
50% 75% NN 95% 9 100% 50% 75% (NN 95% (100%
"NYSE Fraction of E-Trade" NYSE Market Share
1 0.8
0,875 0,7
075 f 06
0,625 05
05 a 04
0,375 03
0,25 0.2
0,125 O21
"960 1970 1980 1990 2000 2010 2020 2030 2040 ‘i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
sens time trading ens time trading
50% 75% NN 95% IN 100% 50% 75% (INI 95% (100%
“total pressure for e-trade from customers" total pressure for floor from floor
100 80
875 70
75 60
625 50
50 40
375 30
25 = 20
125 10
“9601970 1980 1990 2000 2010 2020 2030 2040 “i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
Figure 20: Sensitivity for the time to implement trading
A similar behavior can be observed in the analysis of the management's responsiveness in its
orientation to customers. The model's sensitive reaction to changes in the reactions of the refer-
ence time to change customer orientation can be seen in Figure 21. When market share is inade-
quate, a low reference time to change customer orientation only leads to a slight orientation to
customers. As a consequence, the organization still directs most of its attention towards pressures
from the floor, and it only very slowly adopts electronic trading. This means that also market
share stays low fora long time. The effect on the time to change customer orientation is similar to
that of Figure 20.
31
sens time cust orient time custorient
50% 75% (NN 95% (100% 50% 75% (NN 95% 100%
"NYSE Fraction of ETrade" NYSE Market Share
1 08
0,875 07
0,75 06
0,625 05
05 = 04
0,375 ha 03
0,25 02
0,125 | ol
“960 1970 1980 1990 2000 2010 2020 2030 2040 ‘i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
sens time custorient
50% 75% (NNN 95% (100%
customer orientation
1
0,875
0,75,
0,625
05
0,375
0,25
0,125
“960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year)
Figure 21: Sensitivity for the time to change customer orientation
The sensitivity analyses for the management's general openness to change and for its responsive-
ness of customer orientation supports the impact of the management and the organization on the
implementation of electronic trading. The graphs show that the shift towards e-trade could have
happened somewhat faster had the organization been more reactive. The implementation could
also have gone much slower if the organization had been more rigid in its implementation time
and its reaction of customer orientation to market share. Thus, the management turns out to be a
crucial factor when it comes to the implementation of electronic trading.
The examples so far concemed the effect of changes in a single or a few variables on the
implementation of electronic trading; it will be interesting to see what possibilities combined ef-
fects open up. Figure 22 shows a sensitivity analysis for a simultaneous random change of all the
variables discussed above in the sensitivity analysis.
32
sens all all
50% 75% (NN 95% (100% 50% 75% (NN 95% (100%
"NYSE Fraction of ETrade" NYSE Market Share
1 0.8
07
06
05
04 —
0,875
0,75
0,625
05
0,375
0,25,
0,125
03
0,2
Or —
ia |
“9601970 1980 1990 2000 2010 2020 2030 2040 ‘i960 1970 1980 1990 2000 2010 2020 2030 2040
Time (Year) Time (Year)
Figure 22: Sensitivity for all
The upper left graph of Figure 22 makes it obvious that the history and the future of the New
Y ork Stock Exchange could well have been different. The Exchange could only in very extreme
cases have inhibited the shift to electronic trading forever, but the extent to which it got imple-
mented could have been different. On the other hand, automated trading could also have been
implemented in the 1990s already. The differences arise from variations in the pressure for and
against automation, how vigorously the stakeholder groups pressure for their aims and how pow-
erful their pressure is. Different setups in the management can also lead to different trajectories.
As the market share graph of Figure 22 indicates, these different theoretical possibilities lead in
different positions of the New Y ork Stock Exchange in the market so that different futures would
have been possible.
7 Conclusions
7.1 Insights from the Modeling Process
The analysis reveals that the course of events in the implementation of electronic trading at the
New Y ork Stock Exchange could well have been different. It shows that specialists, floor brokers
and non-institutional customers were successful at delaying the adaptation process by continuing
to pressure for the traditional system. It also shows that, had the floor firms been more cohesive
and/or had their cultural pressure and power been higher, the pressure for the old system would
have been fiercer.
The model analysis also demonstrates that a management which even in times of perform-
ance crisis does not direct its attention to customers could have been a strong force for a further
delay in e-trade. Likewise, it was the initial situation where customer orientation was very low
due to the historical importance of the floor that was the reason that the shift did not yet happen
in the 1990s.
The modeling process is also able to illustrate why the changes appeared to be so radical. It
was a long accumulation of pressures, e.g. from dissatisfaction, that for a long time was held up
by management. Metaphorically speaking, the shift in management functioned like a valve, so
that once the opportunity was apparent, these pressures got released.
Including the management decision-making into the analysis proved to be necessary. At the
same time, narrowing the focus to the management only would not provide the desired results
since both the environmental developments as well as the management's decisions create en-
33
dogenous forces like customer and stakeholder pressures that feed back to the management's sub-
sequent decision-making.
7.2. Relationship to Organizational C hange Literature
The different scenarios have shown that, on the one hand, the change towards electronic trading
is an adaptation process of the New Y ork Stock Exchange to its competitive environment. Since
the technology provided for the possibility of e-trade and since there was a growth in institutional
customers, automated trading finally also got implemented at the New Y ork Stock Exchange.
Without these external forces, no change in the NYSE’s way of doing trading would have
emerged.
Yet on the other hand, in the case of the New Y ork Stock Exchange the rational adaptation
point of view does not give the full picture and cannot explain how—i.e. at what point of time
and at what pace— and why the shift to electronic trading at the NYSE really happened. This
limitation is apparent in much of the organizational change literature. Therefore, multiple views
as suggested by Morgan are necessary in order to provide a rich understanding of the change.”
Bamett and Carroll divide organizational change theory into adaptive and selective approaches.
Whereas the adaptive point of view regards change as an adaptation to technological or environ-
mental change, the selective point of view assumes that organizations neither change quickly nor
easily.’ Apart from this selective point of view there also exists what is called the organiza-
tional or population ecology as a theory of change. Van de Ven and Poole note that in this evolu-
tionary theory competitive survival is the central metaphor. For them retention (including inertia
and persistence) is a major force that antagonizes the variation (adaptation) loop.'°* The selective
or evolutionary perspective bases on the assumption that organizations are inert and do not
change easily. This picture fits the idea of different pressures pressing for the retention of the
floor. It also matches the fact that for a very long time the NY SE management did not attend to
its customers and only reacted to pressures from floor firms. Although the NY SE finally adapted,
its corporate culture was well able to delay this process, thus providing evidence of retention,
inertia and persistence.
Yet, as said earlier, this population ecology view does not provide us with the full picture
either. Tushman and Romanelli study the evolution of organizations, and their proposal of the
punctuated equilibrium fits closest to how the New Y ork Stock Exchange evolved. A very long
phase of convergence was punctuated by institutional forces, market share forces, and the man-
agement’ s direction to customers and resulted in radical organizational change. The initial almost
exclusive attention to floor firms and the later almost exclusive attention to customers support
Tushman and Romanelli’s idea of performance-mediated perception bias.!°°
Child indicates that there is room for choice to either make change happen or to let go of
the opportunity.!°° The New Y ork Stock Exchange seized the opportunity for electronic trading
quite late, but when market share dropped, it did decide for electronic trading and against the
102 Cf, Morgan, Gareth: Images of organization, Thousand Oaks, California [u.a.] 1986, pp. 12-13 and 321-322.
13 Cf. Barnett, William P. and Glenn R, Carroll: Modeling Inteal Organizational Change, in: Annual Review of
Sociology, Vol. 21 (1995), No. 1, p. 218.
10 Cf. Van de Ven and Poole: Explaining development and change in organizations, 1995, p. 518.
105 Cf, Tushman and Romanelli: Organizational Evolution, 1985
106 Cf. Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997; and
Child: Organizational structure, environment and performance, 1972.
34
purely manual trading system. It did also decide against full automation and kept parts of its old
structure in a customized way in alignment with the new trading system. This was a deliberate
decision of the NY SE management that revealed high elements of choice.
Elements of all three theories are present in the New Y ork Stock Exchange’s shift to elec-
tronic trading. It is the element of choice as discussed above, but overall it is also an adaptation
process to what the competitive environment that paved the way to which the NY SE finally but
deliberately followed suit. Organizational decline due to the high inertia would have well been
another possible path that also many researchers and industry experts assumed the NY SE would
take. Thus rational adaptation, population ecology with its assumption of organizational inertia
and environmental selection, punctuated equilibrium theory and strategic choice are not that far
apart as some researchers suggest. It is important to note that the management view alone also
does not explain the changes. The adaptation view, culture and resistance view plus the manage-
ment view have to be applied simultaneously to make sense of the seemingly radical shift to elec-
tronic trading by the New York Stock Exchange. As Morgan points out, each new perspective
adds a little different understanding, thus leading to a more complete picture.’” It is their combi-
nation gives the full picture of the dynamics of how the changes unfolded.
The modular build-up of the model that first included elements of the rational adaptation
perspective only and was then extended with elements of culture and resistance and finally with
elements of management inertia and choice reveals the importance of all of these elements for the
explanation of how change came about at the NY SE. But then the sensitivity analyses show that
an instantaneous adoption of electronic trading or the maintenance of the floor would theoreti-
cally have been possible if the participants had made different decisions. This is consistent with
the findings of Sastry who also in a system dynamics model analyzes the differences of the adap-
tive decision-making literature, the population ecology perspective, and the punctuated equilib-
rium theory. She also comes to the conclusion that all three theories can be modeled by different
assumptions about parameters. The analysis of the case study of the New Y ork Stock Exchange
additionally points out that not only parameter changes create these different behaviors, but that
parameters have an influence on loop dominance and can put loops as well as whole model sec-
tors out of action. This suggests that different behaviors in the case study involve different
boundaries of the model. This further confirms that different organization theories to a great ex-
tent differ by their boundaries, i.e. the assumption what needs to be included and excluded in
their analysis.
7.3 Limitations and Future Research
This analysis is still limited by several factors concerning the model. First, technological devel-
opments are presented as an s-shaped development. This bases on developments of other tech-
nologies, e.g. like the mobile phone market, but it could not be grounded in real data. Second, the
factors that determine market share are limited to three: relative speed, relative spread, and mar-
ket quality from the specialist. In the literature, market quality is an important concept, but defini-
tions of what factors market quality includes differ. Others include prices and speed, but these
were modeled as separate influencing factors of market share. Third, the implementation of elec-
tronic trading always also involves a capacity decision, and it would thus be advisable to actually
model the capacity development of floor and e-trades. This part was left out since the fraction of
107 Cf. Morgan: Images of organization, 1997, pp. 12-13 and 321-322.
35
electronic trading also illustrates the changes, but including capacity could be included as a fur-
ther model and behavior test because it could be compared with NY SE data. Forth, concerning
the implementation of electronic trading at the NY SE the effects that change may have on experi-
ence was not modeled. Since manual trading that involves human experience was replaced by a
computer that took over their task, it was assumed to be of minor importance only and completely
left outside the model boundary.
36
1 References
Abolafia, Mitchel Y .: Making Markets: Opportunism and Restraint on Wall Street, Cambridge,
MA, London 1996.
Adler, Paul S. and Bryan Borys: Materialism and Idealism in Organizational Research, in: Or-
ganization Studies, Vol. 14 (1993), No. 5, pp. 657.
Amburgey, Terry L., Dawn Kelly and William P. Barnett: Resetting the Clock: The Dynamics of
Organizational Change and Failure, in: Administrative Science Quarterly, Vol. 38 (1993),
No. 1, pp. 51-73.
Argyris, Chris and Donald A. Schén: Organizational Learning: A Theory of Action Perspective,
Vol. 1, Reading, Mass. [u.a.] 1978.
Astley, W. Graham and Andrew H. Van de Ven: Central Perspectives and Debates in Organiza-
tion Theory, in: Administrative Science Quarterly, Vol. 28 (1983), No. 2, pp. 245-273.
Barnett, William P. and Glenn R. Carroll: Modeling Internal Organizational Change, in: Annual
Review of Sociology, Vol. 21 (1995), No. 1, pp. 217-236.
Blume, Marshall E., Jeremy J. Siegel and Dan Rottenberg: Revolution on Wall Street: The Rise
and Decline of the New Y ork Stock Exchange, New Y ork, London 1993.
Buchanan, James M.: An Economic Theory of Clubs, in: Economica, Vol. 32 (1965), No. 125,
pp. 1-14.
Chapman, Peter: Men At Work, in: Traders Magazine, 20 Jg. (2007), No. 274, Issued Nov. 2007,
pp. 44-75.
Chapman, Peter: Rule Changes: NY SE Euronext Bets on Specialists to Revive Broken Floor,
Traders Magazine, accessed 20. Aug. 2008:
http://www.tradersmagazine.com/news/101217-1.html.
Child, John: Organizational structure, environment and performance: The role of strategic choice,
in: Sociology, Vol. 6 (1972), pp. 1-22.
Child, John: Strategic Choice in the Analysis of Action, Structure, Organizations and Environ-
ment: Retrospect and Prospect, in: Organization Studies, Vol. 18 (1997), No. 1, pp. 43.
Clemons, Eric K. and Bruce W. Weber: Information Technology and Screen-Based Securities
Trading: Pricing the Stock and Pricing the Trade, in: Management Science, Vol. 43
(1997), No. 12, pp. 1693-1708.
Darwin, Charles: The Origin of Species, London 1971.
Ellul, Andrew, Craig W. Holden, Pankaj Jain and Robert Jennings: Order dynamics: Recent evi-
dence from the NY SE, in: Journal of Empirical Finance, Vol. 14 (2007), No. 5, pp. 636-
661.
Ewing, Jack: When CEOs Talk to Each Other, World Economic Forum in Davos, Switzerland,
2005, in: Business Week Online from 31. Jan. 2005,
http://www.businessweek.com/bwdaily/dnflash/jan2005/nf20050131 5572.htm (2. April
2008)
Feldman, Stuart: Electronic Marketplaces, in: Internet Computing, IEEE, Vol. 4 (2000), No. 4,
pp. 93-95.
Gasparino, Charles: King of the Club: Richard Grasso and the Survival of the New Y ork Stock
Exchange, New Y ork 2007.
Hannan, Michael T. and John Freeman: The Population Ecology of Organizations, in: The
American Journal of Sociology, Vol. 82 (1977), No. 5, pp. 929-964.
Hannan, Michael T. and John Freeman: Structural Inertia and Organizational Change, in: Ameri-
can Sociological Review, Vol. 49 (1984), No. 2, pp. 149-164.
37
Kieser, Alfred, Cornelia Hegele and Matthias Klimmer: Kommunikation im organisatorischen
Wandel, Stuttgart 1998.
Larsen, Erik R. and Alessandro Lomi: Resetting the clock: a feedback approach to the dynamics
of organisational inertia, survival and change, in: Journal of the Operational Research So-
ciety, Vol. 50 (1999), No. 4, pp. 406-421.
Larsen, Erik R. and Alessandro Lomi: Representing change: a system model of organizational
inertia and capabilities as dynamic accumulation process, in: Simulation Modelling Prac-
tice and Theory, Vol. 10 (2002), No. 5-7, pp. 271-296.
Lucchetti, Aaron: NY SE Plans to Revise Specialist-Trader Rules, in: Wall Street Journal - East-
em Edition (2008), pp. C 4.
Malone, Thomas W., Joanne Y ates and Robert I. Benjamin: Electronic Markets and Electronic
Hierarchies, in Allen, Thomas J. and Michael S. Scott Morton (Ed.): Information Tech-
nology and the Corporation of the 1990s: Research Studies, New Y ork, Oxford 1994, pp.
61-83, originally published in: Communications of the ACM, Vol. 30 (1987), No. 6, pp.
484-497.
McGeehan, Patrick: Next to Downsize On Wall Street? The Exchange Floor, in: New Y ork
Times (2007), Issued 23. Sep. 2007, pp. 37.
Mellahi, Kamel and Adrian Wilkinson: Organizational failure: a critique of recent research and a
proposed integrative framework, in: International Journal of Management Research, Vol.
5/6 (2004), No. 1, pp. 21-41.
Mojtahedzadeh, Mohammad: Personal Communication, 2008, Apr. 2008.
Morgan, Gareth: Images of organization, Thousand Oaks, California [u.a.] 1986.
Morgenson, Gretchen: The Fall of a Wall Street Ward Boss, in: The New Y ork Times (2003),
Issued 19. Oct. 2003, pp. 1, 10.
NASDAQ website, NASDAQ Facts, Home > Nasdaq Corporate > Nasdaq Facts,
http://www.nasdag.com/reference/nasdaq_facts.stm (Dec. 7, 2007)
NYSE Euronext: Types of Members, [Home > Products & Services > NY SE Equities > Trading
Licenses > Overview > Types of Members], accessed 24. Oct. 2007:
http://www.nyse.com/productservices/nyseequities/1167954368183.html.
NY SE Euronext: Facts and Figures, [NY SE Time Series Data]:
http://www.nyxdata.com/nysedata/D efault.aspx?tabid=115.
NY SE Exchanges Blog: Comment by Tony Dey, owner of a day trading firm,
http://exchanges.nyse.com/archives/2007/11/men_at_work.php (25. March 2008).
NY SE Exchanges Blog: Comment by Tony Dey, owner of a day trading firm,
http://exchanges.nyse.com/archives/2007/06/duncan.php (25. March 2008).
NYSE Exchanges Blog: Duncan Niederauer, NY SE Group President and Co-COO, now CEO, at
the NY SE Analyst Day 2007, http://exchanges.nyse.com/archives/2007/06/duncan.php
(25. March 2008).
NY SE Exchanges Blog: Duncan Niederauer, NY SE Euronext President and Co-COO and Head
of U.S. Cash Markets, now CEO, NY SE Glossary: “Seat”,
www.nyse.com/glossary/Glossary.html (Dec. 7, 2007).
NYSE Group, Inc.: Annual Report of 2005, New Y ork City 2006.
NY SE Group, Inc.: Annual Report of 2006, New Y ork City 2007a.
NYSE Group, Inc.: NY SE Completes Hybrid Market Phase III Activation, NY SE Group, Inc.,
[News Release], accessed 11. Dec. 2007:
http://www.nyse.com/pdfs/HybridPhaselV 1.24.07.pdf.
NY SE Regulation, Inc.: Information Memo: Hybrid Market Implementation - Phase IV, No. 07-
12, 29 Jan. 2007
38
http://apps.nyse.com/commdata/PubInfoM emos.nsf/A llPublishedInfoMemosNyseCom/85
256FCB005E19E885257272006565C 6/$FILE/Microsoft%20W ord% 20-
% 20D ocument% 20in%2007-12.pdf, accessed 20 Aug. 2008.
. NYSE website: Facts and Figures > Institutional Investors > Holdings of corporate equities in
the U.S. by type of institution (Oct. 1, 2007).
Pareto, Vilfredo: Manual of Political Economy: Transl. from the French ed. of 1927, London
[u.a.] 1972.
Perrow, Charles: Complex Organizations: A Critical Essay, New Y ork [u.a.] 1986.
Picot, Arnold, Christine Bortenlanger and Heiner Rohri: The Automation of Capital Markets, in:
Joumal of Computer- Mediated Communication, Vol. 1 (1995), No. 3.
Picot, Arnold, Christine Bortenlanger and Heiner Rohrl: Organization of Electronic Markets:
Contributions from the New Institutional Economics, in: The Information Society, Vol.
13 (1997), No. 1, pp. 107-123.
Securities and Exchange Commission (SEC): Regulations NMS; Final Rule, in: Federal Register,
Vol. 70 (2005), No. 124, pp. 37495-37644.
Seligman, Joel: The Transformation of Wall Street: A History of the Securities and Exchange
Commission and Moder Corporate Finance, rev. Ed., Boston 1995.
Stengel, Martin: Psychologie der Arbeit, Weinheim 1997.
Sterman, John D.: Business Dynamics: Systems Thinking and Modeling for a Complex World,
Boston [u.a.] 2000.
Steverman, Ben: NY SE: Hooray for Market V olatility, in: Business Week Online (2007), Issued
5. Nov. 2007, pp. 25.
Tushman, Michael L. and Elaine Romanelli: Organizational Evolution: A Metamorphosis Model
of Convergence and Reorientation, in Cummings, Larry L. and Barry M. Staw (Ed.): Re-
search in Organizational Behavior, 7. Vol., Greenwich, CT 1985, pp. 171-222, originally
published.
Van de Ven, Andrew H. and Marshall Poole: Explaining development and change in organiza-
tions, in: The Academy of Management Review, Vol. 20 (1995), No. 3, pp. 510-540.
Weiss, Gary: The $140,000,000 Man, in: Business Week, (2003), No. 3849, Issued 15. Sep.
2003, pp. 84-92.
39
