## l

The major differences between USPS and OTA were assumptions about the incentive rates for ZIP+4 usage (OTA and GAO treated these as a cost, since incentives appear to be required to get large mailers to use ZIP+4), multi-line OCR performance rates (OTA and GAO concluded that performance would likely be somewhat better than USPS estimated), ZIP+4 usage, and clerk/carrier savings rate.

For the latter three variables, OTA developed low, median, and high estimates. For low, median, and high estimates of multi-line OCR performance. OTA concluded that production model multi-line OCRs would read 60, 65. and 75 percent of 5-digit mail to the 9-digit level, respectively.

For ZIP+4 usage, USPS projects that usage would reach 90 percent of the machinable metered first class mail base within 5 years. This is considerably more optimistic than actual experience with either the U.S. 5-digit ZIP code or the Canadian

6-digit years,

postal code. The 5-digit ZIP took 12 years the 5-digit usage level was about 51 percent.

to reach 90 percent usage; after 5 Thus, the USPS projection shows

ZIP+4 reaching 90 percent about two and one-half times as fast as the 5-digit ZIP (in years rather than 12). At present, the projected actual volume of ZIP+4 mail for 1984 about 20 percent of the original USPS projection.

5 is

After reviewing all available evidence, OTA concluded that the USPS ZIP+4 projection should be considered optimistic (high), that an appropriate median estimate would be the 5-digit ZIP growth pattern, and that an appropriate pessimistic (low) estimate would be a growth pattern peaking out at about 40 percent ZIP+4 usage. At the present time, the first year ZIP+4 usage could turn out to be even more pessimistic. The estimated 2.73 billion pieces of ZIP+4 first class mail at year end 1984 represents about 5.4 percent of the target mail base as compared to about 7 percent under the low scenario, 13 percent under the median, and 28 percent under the optimistic scenario. (GAO made no estimate of ZIP+4 usage but did find that businesses are still concerned about the cost of converting and whether the USPS presort discount will be modified because of ZIP+4.)

For clerk/carrier labor savings, OTA concluded that the USPS baseline estimate was probably somewhat optimistic, since the quality and mix of the OCR processed mail may be less than anticipated, and labor costs (e.g., for maintenance) may be more than expected. OTA assumed high, median, and low labor savings rates of 100 percent, 90 percent, and 80 percent of the USPS estimate.

Results of decision analysis. With respect to internal rates of return (ROIs), every option except option F (cancel), under all conditions modeled, shows an ROI above the 15

percent threshold established by USPS.

# OTA assumed, therefore, that

scenario,

the

# Phase

# I

single-line

# OCRs

already

purchased

would

be

kept

in

under any service. All

ROIs, net present with the Phase I

values, and cash flows were calculated net of

single-line

OCRs.

Use

of

ROIs

for

decision

cash flows associated making has a serious

limitation. When more than one option clears the hurdle rate (that is, has more than minimum required ROI, in this case 15 percent), the ROI itself gives no indication of cash flow differences of the various options as a basis for comparing the options.

the the An

alternative to ROI frequently used in capital investment decision making value (NPV). NPV discounts the cash flows of each option at the hurdle or in this study 15 percent.

is net present threshold rate,

10