are sensitive to such refinements, and to the extent 

 that the state of the art permits accurate refinement 

 of the estimate. Appropriate proxy measures may 

 be used in lieu of individual assessments at each 

 element in the system when evaluating system 

 delays. 



(b) Interaction of supply and demand schedules. 

 The entire evaluation procedure (2.6.4 through 

 2.6.15) is based on an assumption that the supply 

 and demand schedules are Independent; but in 

 fact, they are not. This problem is most acute when 

 considering the variance in delays at high levels of 

 lock utilization. Essentially, shippers will face not an 

 expected delay value but rather a highly uncertain 

 delay value. Shippers' response to uncertainty (as 

 reflected in the demand schedule) may be quite dif- 

 ferent from their response to an expected shipping 

 cost (as reflected by the intersect of the supply and 

 demand schedules). 



(c) User fee collection. The incremental collection 

 of user charges, fees, or taxes is not a NED bene- 

 fit. It is a transfer of resources between the private 

 and public sectors of the economy, manifesting 

 itself as resources committed to the proposed navi- 

 gation system. The increased collection of these 

 charges, fees, or taxes is therefore considered a 

 decrease in the public sector's contribution to the 

 proposed system. 



(d) Sensitivity analysis. Project benefits are calcu- 

 lated on the basis of "the most probable" with-pro- 

 ject and without-project conditions. However, risk 

 and uncertainty should be addressed in the analy- 

 sis of NED benefits and costs. In particular, major 

 uncertainty exists in the proper measure of savings 

 to shippers, namely the difference in long-run mar- 

 ginal costs. To the extent that rates or other prices 

 vary from long-run marginal costs, savings to ship- 

 pers will contain a component of transfers varying 

 from real resource savings. This element of uncer- 

 tainty should always be identified or acknowledged 

 in estimates of benefits. In dealing with uncertainty, 

 three techniques may be used: establishing consist- 

 ent sources of data, expanding the data-gathering, 

 and estimating the range of benefits. Use the fol- 

 lowing two specific approaches to implement the 

 third technique, and display the results in terms of 

 their effects on project benefits in tabular form in 

 the project report. 



(1) Prespecified sensitivity analysis. Compute the 

 following and include it in the report: 



(i) Current tonnage, new waterway. For new wa- 

 terways, compute benefits for the recommended al- 

 ternative on the basis of current phased-in tonnage 

 (steps 3 and 9c), current rates, and current fleet 

 characteristics. 



(ii) Current rates, fleet For both new and existing 

 watenways, compute benefits for the recommended 

 alternative on the basis of tonnage over time, cur- 

 rent rates (step 3), and current fleet characteristics. 



(iii) Growth beyond 20-year period. Compute the 

 benefits for alternatives carried forward for final dis- 

 play assuming no growth in tonnage or changes in 

 fleet characteristics or costs beyond 20 years in the 

 future. 



(iv) Interest rate. For projects whose authorized 

 discount rate is different from the current discount 

 rate, compute annualized benefits using the current 

 rate. 



(v) User charges. Estimate the effect on benefits 

 of full cost recovery through user charges. 



(2) Other in addition, the report should contain 

 such other sensitivity analyses as are necessary to 

 meet the objective of a clear, concise report pre- 

 senting a range of benefit levels that represent 

 data and assumptions about which reasonable per- 

 sons might differ. 



(e) Data sources. The following discussion sum- 

 marizes key data sources, including problems in 

 their use. 



(1) Interviews. Interview data may be used in 

 steps 1 through 9. (Use only forms approved by the 

 Office of Management and Budget.) Collect data 

 not available from secondary sources by personal 

 interviews. Use statistically sound techniques for 

 selecting the interview sample and for devising the 

 questions. The questionnaire and a summary of re- 

 sponses should be compiled and displayed in the 

 final report in such a way as to prevent the disclo- 

 sure of individual sources. Describe the errors and 

 uncertainty inherent in the sampling methods and 

 responses. 



(2) Other The basic organizational source for 

 systematically collected waterway data is the Office 

 of the Chief of Engineers. 



2.6.16 Report and display procedures. 



Clear presentation of study results, as well as 

 documentation of key input data assumptions and 

 steps in the analysis, will facilitate review of the 

 report. Tables 2.6.16-1 through 4 are suggested 

 presentations for all reports that include navigation- 

 al objectives. In addition to detailed data on the 

 NED benefits of a project, summary tables may 

 present useful information on other aspects of the 

 project such as its impact on commodity flows, on 

 other modes of transportation, and on the location 

 of economic activity. See the following sample 

 tables. 



56 



