Insider satire-with-data on defence programme cost growth.
Augustine ran Martin Marietta and later Lockheed Martin, and served as Under Secretary of the Army. He wrote fifty-two numbered laws about the defense industry and about management generally, each supported by data he assembled and presented with deliberate humour.
The best-known is the sixteenth law, which extrapolates the growth in unit cost of American tactical aircraft. Augustine plotted the cost of successive fighter generations on a logarithmic scale, found a straight line, and observed that if the trend continued the entire defense budget would buy exactly one aircraft in 2054, to be shared between the Air Force and the Navy on alternate days. The point is serious: unit costs of complex weapons have risen far faster than inflation or than budgets, and the resulting quantities fall toward numbers too small to be militarily useful.
Other laws address the same territory with the same method. He shows that the last ten per cent of performance generates one third of the cost and two thirds of the problems, that programme cost estimates are systematically optimistic and grow as programmes proceed, that software costs came to dominate hardware costs, that reducing quantities to save money raises unit cost and eventually saves nothing, and that organisations expand their staff independent of the work available.
The data collection is real and the underlying observations have held. Augustine's cost curve has continued roughly on trend through the F-22 and F-35 programmes, which is the strongest possible support. The book's format, with jokes and cartoons throughout, causes some readers to underrate the analysis. It is dated in its examples, since everything is drawn from the 1960s to the early 1980s, and it does not address the post-Cold War consolidation of the defense industry that Augustine himself led.
Anyone working in or investing in defense and complex systems programmes, and managers who want data-supported observations about cost growth and requirements creep.
Chosen by shared topics, the entries that hold them, and the people who recommended both.




