Relevant Costs 233Research by psychologists testing the decision behavior of individuals
including business managers clearly shows that sunk costs can adversely affect
judgment.^3 For instance, executives will choose rightly to make a substantial
initial investment in a simulated project, such as new product development,
an R&D effort, or a capital investment. Yet, they continue to make cash invest-
ments even when new information in the simulation is highly unfavorable. By
contrast, executives who enter the simulation only at the second decision with
the same information (here, previous management has made the initial deci-
sion) are much more likely to pull the plug and write off the investment. The
moral is clear; it’s difficult to be objective when one is already psychologically
invested in the initial decision (the more so the larger the initial sunk cost).
Initial investors tend to maintain an overly optimistic outlook (despite the unfa-
vorable new information) and adhere to the status quo established by their ini-
tial decision. Sunk costs also have effects in other contexts. For instance, in
ongoing business disputes ranging from labor impasses to law suits, the rival
parties frequently dig in as costs accumulate and refuse to settle (even when it
is in their self-interest), thereby escalating the conflict.
Government spending programs, particularly in energy, defense, and
basic science face similar challenges. During the 1980s and 1990s, the U.S.
government halted public spending on scores of energy projects, including
almost all synthetic-fuel programs ($25 billion spent). In 1989, Congress
authorized the largest pure science project ever undertaken, the
Supercollider program. Unhappily, the project’s cost estimates obeyed their
own law of acceleration, rising over the years from $4.4 billion to $6 billion
to $8.2 billion to $11 billion to $13 billion. In 1993, with $2 billion already
spent and 15 miles of underground tunnels dug, Congress voted to abandon
the program. Nonetheless, some weapons programs seem to have nine lives—
refusing to die even when their original Defense Department sponsors have
recommended cancellation. The futuristic Airborne Laser, conceived in the
1980s to shoot down enemy ballistic missiles, is a case in point.^4 Since 1996,
the Pentagon has spent some $5.2 billion on the program with only a poorly
performing test aircraft to show for it. Believing the concept to be unwork-
able, the Clinton, Bush, and Obama administrations have all recommended
cancellation. Yet defense contractors, the locales where the development
work is being done, and proponents in Congress have all lobbied hard to
continue funding.
In recent years, the budget axe has been used effectively in scrapping a
number of uneconomical large-scale programs. Nonetheless, as critics point(^3) For research on decision making and sunk costs, see H. Arkes and C. Blumer, “The Psychology of
Sunk Cost,” Organizational Behavior and Human Decision Process, 1985, pp. 124–140; and W.
Samuelson and R. Zeckhauser, “Status Quo Bias in Decision Making,” Journal of Risk and Uncertainty
(1988): 7–59.
(^4) This discussion is based on N. Hodge, “Pentagon Loses War to Zap Airborne Laser from Budget,”
The Wall Street Journal(February 11, 2011), p. A1.
c06CostAnalysis.qxd 9/29/11 1:46 PM Page 233