4. Estimating Quantitative Values for Benefits

In the previous section, we explained that the shift in a shipper's demand for freight transportation would be the conceptual basis for estimating the benefits from a freight improvement. With the theoretical foundation in place, we turn to the problem of estimating actual values. It is one thing to build a framework of abstract reasoning on a demand curve as a theoretical construct; it is quite another thing to develop an empirical estimate of a demand curve or of a shift in a demand curve.

The approach that we have chosen to this empirical problem is based on a mathematical relationship that economists call "elasticity" of demand. If you know the price at which a good or service is actually being sold, and the quantity that is being sold at that price, and you know the elasticity of demand, then you have a great deal of information about the demand curve. For our purposes, elasticity of demand has two virtues: it provides useful information on demand, and therefore on benefits; and it may be feasible to estimate it.

Elasticity may be thought of as a measure of responsiveness. Elasticity of demand has to do with the response of demand for a good to a change in the price of that good (or service)—more precisely, it is a relationship between a change in price and the associated change in quantity sold.[7] Elasticity may be applied to other economic phenomena. Economists often investigate elasticity of costs, for example, with respect to other variables that may be of interest.

For our purpose of benefits estimation, we are concerned with the elasticity of demand for freight transportation with respect to the cost of freight transportation, cost including time costs. (The full mathematical argument in support of this point is set out in the White Paper, and there is no need to repeat it here.) The value of this elasticity can be known from the values of two other, related elasticities. One is about how shippers' demand for freight transportation responds to changes in time costs—elasticity of demand with respect to time costs. The other is about how shippers' logistics costs respond to changes in time costs—elasticity of logistics costs with respect to time costs.[8]

Intuitively, it seems reasonable that the response of shippers' demand and shippers' logistics costs to changes in time costs should be key factors in estimating benefits, and the mathematical argument in the White Paper demonstrates this point. Further, there is some prior experience in trying to estimate elasticity of cost with respect to time cost. As part of work done under NCHRP 2-17(4) [2], HLB Decision Economics developed a method for estimating the elasticity of logistics costs with respect to time costs. It appears feasible that the same method could be used for estimating elasticity of demand for freight transportation with respect to time costs.

The method used in 2-17(4) [2] was essentially one of very detailed interviews with representatives of a number of firms, the representatives being executives with close knowledge of their firms' logistics arrangements and likely investment strategies in the face of falling costs of freight carriage. The degree of time cost reduction—reduction in transit time and gain in reliability—required to cause a firm to make the investments required to reduce the number of its warehouses and increase their size is the most critical piece of information required for estimating both elasticity of logistics costs and elasticity of freight-transport demand with respect to time costs. Obtaining this information in sufficient detail and for a large enough sample of firms is probably the most challenging task confronting the team as we move forward into the data-gathering effort.

A number of other data items are also essential for this effort. The following is a partial enumeration of the most important information needs.

• Direct time-cost savings, from transit-time savings and increased reliability, due to highway improvements, and indirect savings from logistics reorganization effects;
• The threshold of time-cost savings at which reorganization would occur;
• The elasticity of logistics costs with respect to time-cost savings (derived from the above two items);
• Current level of firms' demand for freight transportation and elasticity of freight-transportation demand with respect to time-cost savings (also closely related to the first two items);
• Volume of freight traffic and product movement across transportation links by commodity and market; and
• Average travel time and variability following highway improvements, or other policy actions, for various transportation network links and nodes—other policy actions could include such changes as relaxed weight or height restrictions, port clearance characteristics, or the like.

Much of these data would be initially collected at the level of individual firms and then aggregated by type of industry. As noted above, the information would be obtained in detailed interviews with executives of firms that buy freight transportation, executives closely acquainted with logistics operations and strategy and likely investment strategy for achieving increased savings through improved logistics. This line of research would be based on sophisticated interview techniques and employ what are known as stated-preference methods. Stated preference involves designing interviews so as to elicit quantitative information from respondents on choices they would make when confronted with various changes in their business environments.

What we sketch out here is the overall strategy in terms of the kind of information needed and the general approach to obtaining it. Working out the details of the information-gathering process is the next major step for the study team, and it will be a challenging task. It will be necessary to find a substantial number of business executives who are willing to participate in the study and willing to share some detailed information on their firms' business operations.

previous | next