The results shown in Table 4 are substantially similar to those in Table 3, except that they suggest a somewhat lower contribution of the elimination of cabooses to rail productivity over the period (5% of 1997 costs rather than 8% in Table 3) and a higher contribution of the general time trend (42-66% rather than the 32-33% in Table 3).
rail freight industry during the post-Staggers period, and, for the first time, they allow us to isolate the effects of one crucial innovation, which is the operation of trains with no cabooses.
It is worth noting as well that some of the benefits of elimination of cabooses are not included in our figures.
Furthermore, the elimination of cabooses from freight trains had yielded all its benefits in productivity growth by the mid-1990s, and if further innovations did not occur, productivity growth would have slowed; this point was made earlier by Martland (1999, pp.
This will misestimate the savings because some of the savings will be operation and maintenance of the cabooses. Second, Maitland's analysis does not account for substitution effects in production.
For a description of the technological role of cabooses and how technology made cabooses obsolete, see White (1993, esp.
(11) Because the fraction of train miles with cabooses can take a value of zero, the natural logarithm of this variable is undefined in some cases.
This in turn requires that cabooses and their accompanying crews be "deadheaded" in one direction.
Second, it isolates the contribution to productivity and lower costs of a crucially important technological innovation implemented in the years shortly after deregulation, (2) namely, the ability to run trains with crews of two members, rather than four or five, and the elimination of the caboose, traditionally needed both for safety at the end of a freight train and as quarters for the two or three crew members now technologically obsolete.
The caboose mile variable (which is new to this paper and has not been used in previous research) ranges between zero and slightly greater than one.
To the extent that this trend toward shipper-provided cars is proxied by this interaction term, we have good reason to believe that our analysis controls for it, and that the effects we have measured for time and caboose miles represent true changes in productivity.
The interaction terms with the caboose variable are for the most part not significant.
The effects of relaxing the constraints on our quantitative results (productivity changes and the caboose's contributions to them) are discussed in the next section after the results of the constrained equations.
Finally, the combined effects of the two sets of technological progress indicate that, had both the time trend and caboose use been at 1983 levels for 1997 firms, costs would have been 20-77% higher, depending on the railroad, with an average increase of 43-44%, depending on the weighting scheme used.