Stuart A. Umpleby
Research Program in Social and Organizational Learning
The George Washington University
Washington, DC 20052 USA
umpleby@gwu.edu
Abstract
The year 2000 computer problem became the largest technical project in human history and the greatest management challenge since World War II. It led to unprecedented peacetime cooperation. What will be the effects on academic disciplines? This paper suggests ways in which a variety of disciplines may be affected. The year 2000 problem can be used to test social science theories. Theorists have made predictions of effects, which can be tested against events. Activists helped organizations and communities prepare and in so doing learned a great deal about how people respond to hypothetical, technological dangers. The failure to foresee, or correctly to perceive, the year 2000 computer problem may lead to changes in research programs in several disciplines.
1 Introduction
Several before and after studies of y2k are being conducted in the years 1999 and 2000 in an effort to turn the great uncertainty associated with the year 2000 computer problem (y2k) into a learning experience for managers and social scientists. Several of these studies are being conducted within the systems science community. A broad range of theories and action research strategies are being tested. See www.gwu.edu/~y2k. Click on Before and After Studies.
2 Psychology
Several psychological phenomena were revealed by y2k:
* When managers learned the potential effects of y2k, they often went through a period of denial before taking action. This delayed the start of serious work.
* When members of the public learned about y2k, some people felt considerable anxiety.
* Public officials were afraid of public panic. Preventing “public overreaction” was often given as much attention as fixing vulnerable equipment.
* Once the government defined preventing public panic as a goal, the credibility of optimistic government reports declined.
* Even after January 3, the first day of business, many people were slow to change their estimates of the possibility of disruptions. Theories in psychology help to explain these y2k phenomena.
The theory of cognitive dissonance [Festinger et al., 1956] helps to explain why people were slow to act on the y2k threat and why opinions did not change very much as projects were successfully completed. Y2k has been a threat unlike any we have experienced before. The closest parallel might be the use by the Romans of lead pipes for water. Never before did the technology, which we rely upon for a high standard of living, threaten to bring about global social, economic, and political disruption. Many people chose not to think about the implications.
Solomon Ash [1963] conducted experiments on social stress. For example, in a classroom of students he drew three lines of different lengths on the board. He told everyone in the class to say that the line of middle length was the longest. He then brought in a student who had not been given these instructions. He went around the class asking which line was the longest. The original students in the class all said that the line of medium length was the longest, and most of the time the naïve student went along with the group. Only a minority of the students not in the original group said that his or her classmates were mistaken. This theory helps to explain how governments were able to shape public opinion regarding y2k. In order to maintain public confidence, governments put an optimistic face on y2k status reports. These reports tended to diminish the credibility of those advocating that citizens and communities prepare. These reports also created doubt in the minds of activists that preparation and community participation were in fact needed.
Irving Janis’s [1982] work on “groupthink” shows that once an authoritative position has been defined, the members of an organization are reluctant to speak diverging views. Albert O. Hirschman’s [1970] book Exit, Voice and Loyalty defines the alternatives that are available to employees whose views differ from the authoritative view.
Gregory Bateson’s [1972] theory of a “double bind” can be used to explain the dilemma of activists who felt that y2k might be more serious than official estimates. If an activist spoke out against the official line, he or she risked being labeled as crazy or a troublemaker. If the activist did less than he thought was needed, then he was increasing the danger that disruptions would threaten his family and community.
In addition to academic theories there have been efforts by y2k activists to characterize the way different people reacted to y2k. Sally Strackbein has suggested that there are “tree people” and “forest people” from the saying, “Some people cannot see the forest for the trees.” Tree people focus on the details of fixing computer code. Forest people worry about societal implications. My conversations with people about y2k have led me to believe that there are three distinct time intervals for which people choose to prepare. A small group of people attempts to assess the threat and then chooses to prepare for the worst, as a precaution. A second group acknowledges the threat and then decides to prepare for about three days of disruption, since that is easy to do. Finally, most people decide to wait and see what happens. For this last group, y2k may change only the verb tenses: From “that could not happen” to “this cannot be happening,” to “that could not have happened.”
My impression is that many people will be struggling to understand y2k even long after it happens. Attempts to persuade some people to seek to understand y2k before it happens are largely fruitless. As a result of the y2k experience psychological factors associated with foresight and preparation may receive increased attention.
3 Management
It is commonly observed in writings on organizational behavior that good news travels up the chain of command faster than bad news. A variation on this phenomenon is that reports tend to become more optimistic as they travel upward. While working on y2k projects, many consultants observed that assessments of y2k readiness became more pessimistic as they went down in an organization. One consultant reported that reports of what was “done” could be misleading. Consider a middle manager who has ten systems that need repairs. After the manager has signed a contract with a vendor to repair six of them, the manager would often count the six as “done,” and report this to his superior. Although repair work had not yet started, the superior was being told that 60 percent of repairs were “done.”
However, another phenomenon related to organizational politics was also at work. A manager of a large y2k project in the UK said that when top managers became very concerned about y2k, he did not share his view that the effects would be small, because he needed top management commitment to the project in order to get changes made.
In the field of software engineering, according to Ed Yourdon, Barry Boehm [1981] wrote his dissertation on why software development projects are usually late. Many possible causes were investigated – organization of the projects, the personality or training of the manager, funding, planning, etc. Overwhelmingly the most important factor was when the project started. That is, projects which start late finish late.
Since many organizations started their y2k projects late, the high success rate, at least as of January 2000, was due to several factors:
* The embedded systems part of y2k was not as serious as originally thought.
* A great deal of equipment was shut down for the rollover. Some equipment was vulnerable only at the moment of the rollover. Hence, although the consequences of a malfunction could be very serious, in some cases the fix or work-around was extremely easy to implement.
* Information gained from projects that began early was shared very quickly around the world via email and websites. One lesson of y2k is that it is now easy and inexpensive to conduct a world-wide information sharing activity.
Fred Brooks [1995], known as the “father of the IBM System 360,” wrote a book with the title, The Mythical Man-Month. In this book he proposed “Brooks’ Law” – adding people to a software development project delays it. Since the deadline of January 2000 was fixed, the violation of Brooks Law was the nearly universal “solution” to the year 2000 computer crisis.
The history of y2k was that estimates of the seriousness of the problem rose, as assessments were done, and hence budgets to make repairs rose as well. For example, the U.S. government estimated in 1997 that its y2k project would cost 2.3 billion. In early 1999 the estimate was 8.3 billion, and in late 1999 the estimate was 10 billion.
The strategic implications of significant y2k effects often were not given thorough consideration. When doing strategic planning, corporations generally assume the infrastructure will work. The number of consumers, suppliers, and competitors is thought to remain approximately the same. Attention is usually focused on maintaining or increasing market share, not on providing a minimum level of service in the event of major disruptions. Perhaps a wider range of possible environments will be considered in future strategic planning exercises.
The vast majority of the management literature is aimed at optimization. By squeezing even a small expense out of a large process, operations researchers can save a company millions of dollars a year. Comparable attention has not been given to resilience – the ability of an organization to respond effectively to unusual disturbances. A just-in-time inventory system is one example of a highly effective optimization method which can lower the resilience of an organization.
Y2k may lead to greater interest in forecasting and preparation. The fact that technology forecasters did not warn of y2k dangers in a timely manner may lead to a loss of prestige for forecasting.
So far most organizations have succeeded in containing y2k-related failures within the organization. However, if problems with computer equipment continue to occur during the year 2000, y2k may yet be like a Darwinian extinction event for corporations, somewhat like the asteroid impact which is thought to have wiped out the dinosaurs. There may be studies of corporations that failed and those that survived to determine what factors influence the resilience of corporations.
4 Political Science
Political leaders did not do all they could to warn the public to prepare for y2k disruptions. The most widely cited explanation for this policy is that public officials feared “public overreaction.” Changes in consumer behavior would affect normal business processes, particularly for banks, food stores, and pharmacies. Other explanations for not warning the public are:
* Officials feared a loss of confidence in the economy, which would mean that less money would be available to make repairs.
* Political leaders feared a loss of control, if people lost confidence in government.
* Officials genuinely believed that there would be little disruption and that most equipment would be repaired or replaced in time.
One of the intriguing y2k phenomena was the wide variety of ways that official spokespersons would “spin” the status reports. The most common pattern was for a report containing alarming information to have an optimistic, calming executive summary.
Another government action to shape opinion was shortening the time period for which people were advised to prepare in the event of a disaster. Until 1999 U.S. government agencies advised citizens to prepare routinely for three days to two weeks of disruption in case of natural disasters such as floods, hurricanes, earthquakes or tornadoes. These websites were changed in early 1999 to recommend preparing only for three days in the event of y2k disruptions.
One way of dealing with uncertainty about what would happen was to define a variety of scenarios. Peterson, Wheatley, and Kellner-Rogers [1998] used two dimensions to define four possible scenarios. The two dimensions were few failures vs. many failures and social coherence vs. social collapse. Michael Marien [1999] defined two additional diagrams. In each diagram one dimension was few failures vs. many failures. In Marien’s first diagram the second dimension was no other major disturbances vs. some other major disturbances, such as a worldwide financial crisis. In his second diagram the second dimension was minimal global preparation vs. coordinated global preparation. Such diagrams call attention to the range of possibilities that planners need to consider.
In the United States and other countries y2k did not become a political issue. It was treated as a common threat. Opposing parties worked together. However, y2k may yet threaten domestic public support for the political parties currently in power, if some important government services fail in 2000. Ed Yourdon [1999] has suggested, “y2k community groups represent an embryonic populist movement that could become the foundation of a major shift in political power if y2k goes badly.”
5 Economics
Karl Mueller [1998] notes that y2k was the first serious crisis of contemporary information society, a crisis that was global and not transferable in time. In his book The Collapse of Complex Societies Joseph Tainter [1988] suggests that as societies become more complex, additional complexity produces diminishing returns. Eventually the added complexity cannot be sustained and the society collapses. As a society collapses, it falls back to a previous stage of development. If the previous stage of development cannot be sustained, the society falls back to a still earlier stage of development.
In January 2000 the immediate threat of y2k to the economy seems to have passed. Apparently modern societies have not reached a level of complexity that cannot be successfully managed. Systems are more resilient than was thought, and people are better at managing technology than was thought. However, economic disruption caused by efforts to cope with y2k could still occur. Specifically, in an effort to make funds available for y2k repairs and recovery and to counter the expected withdrawal of funds from banks, the Federal Reserve Board in the U.S. increased liquidity greatly. These funds will probably be withdrawn from the economy through higher interest rates. Higher rates usually produce a decline in stock prices.
There has been speculation over whether y2k will produce inflation or deflation. In January 2000 oil prices were rising due in part to fears of y2k caused shortages. The increase in the money supply is also expected to contribute to inflation.
The usual positive feedback of economic growth could be reversed, if business failures increase and ripple through the economy. Rather than growth in investment, employment, and consumption, instead we would see increasing numbers of business bankruptcies, greater unemployment, and reduced demand for goods.
George Soros [1994] maintains that gaps between perception and reality appear from time to time in both financial markets and political systems. When a gap becomes large, one can predict that the gap will not continue indefinitely and that perception and reality will eventually come closer together. Before January 2000 views of the immediate future were far apart. For example, the stock market had a more optimistic outlook than did government and business reports on y2k [U.S. Senate, 1999; Marcoccio, 1999].
The efficient market hypothesis in finance maintains that stocks are always appropriately priced, given the information available. However, information on the implications of y2k was available for several years with apparently little affect on financial markets. One possible conclusion is that stock analysts look at quarterly earnings and interest rates, but do not question the viability of the infrastructure. Another possibility is that the markets did a good job of estimating risk.
Present economic theory assumes a return to equilibrium following a disturbance. When people with an economic background spoke about y2k, one heard statements such as the following: “If there is a problem, someone will devise a solution, since there is an economic incentive.” And, “the managers of corporations are bright people. They will not allow their organizations to fail.” Contemporary economic theory assumes that there is time to devise a solution to a problem. Most threats to the viability of a firm come in the form of new technology, government regulation, or increased competition. Survival entails adapting more quickly than one’s competitors. There is a gradual process of gaining or losing market share. However, y2k presented the specter of a sudden inability to function at all, due to either the failure of critical equipment, the failure of necessary infrastructure, or the unavailability of parts. The “relaxation time,” or the time required to respond to disturbances, may receive more attention in the economics literature in the future. One lesson of y2k is that the time required to respond to a global technological problem has been greatly reduced by email and the world wide web.
A key tenet in economic theory is that people are guided by rational self-interest. However, surveys have indicated that many small and medium-sized enterprises were not working to repair their equipment for y2k. Tradition, inertia, denial, or accustomed life patterns may be stronger than rational self-interest, particularly if significantly new patterns of thought are required. Perhaps these subjects will receive more attention by economists in the future.
6 Engineering
As a result of the expenditures on y2k and the glitches caused by y2k, there is likely to be considerable public discussion about the need to monitor or regulate information technology. Academic research may broaden from a preoccupation with producing new technology to concern also with insuring the reliable functioning of installed equipment. Information technology projects are likely to be better managed. Inventories of hardware and software will be kept and records of changes made in software will be maintained more carefully. Time, as it is represented in computer programs, will likely receive increased attention. Perhaps calendars showing needed changes will be maintained over a longer period of time. The cost of maintaining vs. buying IT equipment may be weighed more frequently and regularly.
7 Cybernetics
Several ideas in the field of cybernetics — the science of control and communication in animals, machines, and social systems – are particularly useful when studying y2k.
The law of requisite variety, as postulated by Ross Ashby [Conant, 1981], states that the amount of regulation that can be achieved is limited by the variety in the regulator. It is a simple but fundamental principal of wide applicability. In the case of y2k it means that the amount of equipment that can be repaired is limited by the number of people available to do the work, the tools available, and the time available. If the workers, tools, and time are insufficient, the task will not be completed.
The principle of self-organization, also proposed by Ashby, states that every isolated, determinant, dynamic system obeying unchanging laws will develop organisms adapted to their environments. Y2K as a stimulus produced a remarkable amount of self-organization. Families planned in ways they had not planned before. New community groups formed, and alliances and coalitions were established. Task forces were organized within companies and government agencies. Several new organizations at the federal level in the United States were created. There were regional meetings of country y2k project managers, and the United Nations also organized meetings. Although new organizations were created at all levels of society, cross-level communication, for example between community activists and federal officials, at times was difficult, no doubt due to different priorities. Activists sought to increase preparation, whereas officials sought to minimize panic.
The idea of self-reference – that statements reflect the point of view of the observer [Umpleby, 1997] – has been a key element in y2k discussions, primarily in the form of self-fulfilling or self-defeating prophesies and the idea that beliefs, real or imagined, are real in their consequences. In 1998 and 1999 government officials were very concerned that even if there were no significant disruptions, runs on banks or grocery stores could be very disruptive. Other variations of the self-reference theme are the following:
* If dire forecasts are made, people may feel more motivated to fix equipment and make preparations, with the result that actual outcomes are not as bad as anticipated.
* Statements by officials that all is well and there is no need for concern can create complacency and inaction, resulting in more disruption than might otherwise have been the case.
The interplay between reality and perception was widely discussed within y2k circles and was an important contributor to uncertainty.
8 Conclusion
The year 2000 computer problem can be regarded as a large-scale social experiment. In 1999 I thought that interest in y2k among academics would rise dramatically in January 2000, because I assumed there would be at least some significant disruptions. However, in January 2000 interest in y2k declined dramatically among the press and the general public. Y2k teams in business and government began to be disbanded. Now there is a danger that the academic community may never become engaged in thinking about y2k as a subject for research. If this happens, we shall have lost a unique opportunity to do worldwide coordinated research on a simultaneous social event that has threatened all modern societies.
References
[Asch, 1963] Solomon Asch. Effects of Group Pressure
Upon the Modification and Distortion of Judgments. In Harold S. Guetzkow, editor, Groups, Leadership and Men. New York: Russell and Russell, 1963, pp. 177-190.
[Bateson, 1972] Gregory Bateson. Steps to an Ecology of
Mind. New York: Ballantine Books, 1972.
[Boehm, 1981] Barry W. Boehm. Software Engineering
Economics. Englewood Cliffs, NJ: Prentice-Hall, 1981.
[Brooks, 1995] Frederick P. Brooks. The Mythical Man-
Month: Essays on Software Engineering. Reading,
MA: Addison-Wesley, 1995.
[Conant, 1981] Roger Conant, editor. Mechanisms of
Intelligence: Ross Ashby’s Writings on Cybernetics.
Seaside, CA: Intersystems Publishers, 1981.
[Festinger et al., 1956] Leon Festinger, Henry W.
Riecken and Stanley Schachter. When Prophesy Fails.
Minneapolis: University of Minnesota Press, 1956.
[Hirschman, 1970] Albert O. Hirschman. Exit, Voice and
Loyalty: Responses to Decline in Firms, Organizations
and States. Cambridge, MA: Harvard University Press, 1970.
[Janus, 1982] Irving Janus. Group Think. Second
Edition. Boston: Houghton Mifflin, 1982.
[Kubler-Ross, 1971] Elizabeth Kubler-Ross. On Death
and Dying. New York: Macmillan, 1971.
[Marcoccio, 1999] Lou Marcoccio. Year 2000 Global
State of Readiness and Risks to the General
Business Community, March 26, 1999.
http://gartner5.gartnerweb.com/public/static/aboutgg/pressrel/testimony1098.html
[Marien, 1999] Michael Marien. More Y2K Scenarios to
Worry About. The Futurist, January, 1999, p. 4.
[Mueller et al., 1999] Karl Mueller, Peter Purgathofer and
Rudolf Vymazal. Chaos 2000: Das globale Zeitbeben.
Vienna, Austria: Doecker Verlag, 1999.
[Peterson et al., 1998] John L Peterson., Margaret
Wheatley, Myron Kellner-Rogers. The Y2K Problem:
Social Chaos or Social Transformation? The Futurist, October 1998, pp. 21-28.
[Soros, 1994] George Soros. The Alchemy of Finance:
Reading the Mind of the Market. New York: John
Wiley and Sons, 1994.
[Tainter, 1988] Joseph A. Tainter. The Collapse of
Complex Societies. Cambridge University Press, 1988.
[Umpleby, 1997] Stuart A. Umpleby. Cybernetics of
Conceptual Systems. Cybernetics and Systems, 28:
635-651, 1991.
[Umpleby, 1999] Stuart A. Umpleby. Why We Missed
the Year 2000 Computer Problem. Eleventh
International Conference on Systems Research, Informatics and Cybernetics, Baden-Baden, Germany, August 2-7, 1999.
[U.S. Senate, 1999] United States Senate, Special
Committee on the Year 2000 Technology Problem.
Investigating the Year 2000 Problem: The 100 Day Report. Sept. 22, 1999, http://www.senate.gov/~y2k.