XI. Assignment of Proceeding

Susan P. Kennedy is the Assigned Commissioner and Sarah R. Thomas is the assigned ALJ in this phase of this proceeding.

Findings of Fact

1. The Order Instituting Rulemaking for this proceeding called for an examination of the service quality of Pacific and Verizon in this phase of the proceeding.

2. The Order Instituting Rulemaking for this proceeding asked us to assess how the quality of service fared under the New Regulatory Framework (NRF).

3. Substantial methodological difficulties arise in the assessment of service quality because there is no record of service quality in the period preceding the adoption of NRF and because we lack a control group of utilities that did not change regulation during this period.

4. More typical methodological obstacles arise because companies often define or measure a particular service quality attribute differently.

5. With the passage of time, the activities of utilities change, and those changes affect what is measured in particular service quality attribute. These changes can arise from corporate reorganizations, which move certain telecommunications activities to subsidiaries not subject to regulation, or from changes in a company's sales, which affect the distribution of activities included in a particular measure of service quality.

6. General Order (GO) 133-B defines specific measures associated with the quality of telecommunications services and set standards for all but one.

7. The Federal Communication Commission's (FCC) ARMIS measures permit the comparison of utility performance against a reference group and the comparison of Pacific and Verizon with each other.

8. Following certain mergers, the FCC began measuring additional service attributes in a series known as "MCOT" measures.

9. Direct measures of utility performance do not necessarily indicate how consumers value specific service attributes.

10. Survey data that directly asks customers their view of service quality is a valuable means of acquiring insight concerning customer views.

11. Pacific has presented results from its ARMIS 43-06 survey, its NRF monitoring report, surveys conducted by IDC and a survey by JD Power.

12. Verizon has presented results from its ARMIS 43-06 survey, and surveys it conducts as part of its customer service operations.

13. Complaints filed at this Commission also offer a source of information on customer experiences.

14. Through GO 133-B, the Commission requires utilities to provide data on seven different measures and has established performance standards for six.

15. Under GO 133-B, the utilities report to the Commission the requests for primary telephone service delayed over 30 days due to lack of utility plant. There is no performance standard for this measure.

16. Under GO 133-B, utilities report to the Commission the percentage of installation line energizing commitments met. Currently, the performance standard for this measure is that a utility should meet 95% of its installation line energizing commitments.

17. Under GO 133-B, utilities report to the Commission the number of initial trouble reports relating to dissatisfaction with telephone company-provided equipment and/or service. The performance standards adopted vary with the size of the end office. Currently, for end offices with more than 3000 lines, the standard is 6 or fewer trouble reports for 100 lines.

18. Under GO 133-B, utilities report the percentage of toll and operator assistance calls answered within 10 seconds. Currently, the performance standard is that 85% of all calls should be answered within 10 seconds.

19. Under GO 133-B, utilities report the percentage of directory assistance calls answered within 12 seconds. Currently, the performance standard for this measure is that the utility must answer 85% of these calls within 12 seconds.

20. Under GO 133-B, utilities report to the Commission the percentage of calls reporting trouble that are answered within 20 seconds. This is known as "Trouble Report Answering Time (TRSAT)." Currently, the performance standard for this measure is that the utility must answer 80% of these calls within 20 seconds.

21. Under GO 133-B, utilities report to the Commission the percentage of calls to the business office that are answered within 20 seconds. This is known as "Business Office Answering Time (BOAT)." Currently, the performance standard for this measure is that the utility must answer 80% of these calls within 20 seconds.

22. Pacific defines as "primary service orders" those containing to the first line into a house. Verizon defines as "primary service orders" the first order for any line.

23. GO 133-B fails to define "primary service" in an unambiguous way.

24. R.02-12-004 is the appropriate forum for resolving how to define "primary service" in an unambiguous way.

25. GO 133-B fails to address the amount of time that a customer spends dealing with "Automated Response Units (ARU)" and neither Verizon nor Pacific consistently track this time.

26. Pacific stated that the time its residence customers spend in its ARU system ranges from a low of 50 seconds to a high of 300 seconds.

27. GO 133-B's failure to address the use of ARUs reflects changes in technology since the Commission adopted the standard, and this technology gap should be closed.

28. GO 133-B fails to track the number of calls blocked by busy customer service lines or the number of calls abandoned by customers.

29. Pacific measures the number of primary service held orders in a way that is inconsistent with GO 133-B's intention to have any order older than 30 days reported to the Commission. Pacific counts orders held over 30 days only once a month.

30. Pacific's "primary service held orders" shows significant improvement during the NRF period. However, Pacific's definition of primary service and erroneous measurement of orders held over 30 days prevent us from making a finding.

31. Verizon's "primary service held orders" shows great volatility over the period 1990 to 2001. Statistical analysis suggests improvement over the NRF period, but the trend of improvement is not statistically significant.

32. When measured on an annual basis, Pacific has exceeded the standard for "installation line-energizing commitments met" throughout the entire NRF study period. We find no significant time trend during the period under NRF regulation.

33. When Pacific's performance is measured on a monthly basis, we find that Pacific met the GO 133-B standard for "installation line-energizing commitments met" for every one of the 144 months for which we have data.

34. When measured on an annual basis, Verizon has exceeded the GO 133-B standard for "installation line-energizing commitments met" throughout the entire NRF study period. We find no significant time trend during the period under NRF regulation.

35. When Verizon's performance is measured on a monthly basis, we find that Verizon met the GO 133-B standard for "installation line-energizing commitments met" for every one of the 144 months for which we have data.

36. When measured on an annual basis, Pacific has consistently exceeded the GO 133-B standard for "number of trouble reports per 100 lines" throughout the entire NRF study period. We find no significant time trend during the period under NRF regulation.

37. When measured on a monthly basis, Pacific has met or exceeded the GO 133-B standard for "number of trouble reports per 100 lines" for each of the 144 months for which we have data.

38. When measured on an annual basis, Verizon has consistently exceeded the GO 133-B standard for "number of trouble reports per 100 lines" throughout the entire NRF study period. A statistical test shows that Verizon has significantly improved its performance during the period under NRF regulation.

39. When measured on a monthly basis, Verizon has exceeded the GO 133-B standard for "number of trouble reports per 100 lines" for each of the 144 months for which we have data.

40. When measured on an annual basis, Pacific has consistently exceeded the GO 133-B standard for "operator assistance answer time, yearly average" throughout the entire NRF study period. A statistical test shows no statistically significant time trend during the period under NRF regulation.

41. When measured on a monthly basis, Pacific has exceeded the GO 133-B standard for "operator assistance answer time, yearly average" for 107 of the 108 months of the NRF period for which we have data.

42. When measured on an annual basis, Verizon has consistently exceeded the GO 133-B standard "operator assistance answer time, yearly average" throughout the entire NRF study period. A statistical test shows no statistically significant time trend during the period under NRF regulation.

43. When measured on a monthly basis, Verizon has met or exceeded the GO 133-B standard "operator assistance answer time" for 140 of the 144 months in the NRF study period

44. When measured on a yearly basis, Pacific has consistently exceeded the GO 133-B standard of 12 seconds for "directory assistance answer time, yearly average" from 1993 to 2001. A statistical test shows no statistically significant time trend during the period under NRF regulation.

45. When measured on a monthly basis, Pacific has met or exceeded the GO 133-B standard of 12 seconds for "directory assistance answer time" each of the 108 months for which we have data.

46. When measured on an annual basis, Verizon has consistently exceeded the GO 133-B standard of 12 seconds for "directory assistance answer time, yearly average" from 1993 to 2001. A statistical test shows no statistically significant time trend during the period under NRF regulation.

47. When measured on a monthly basis, Verizon has met or exceeded the GO 133-B standard for "directory assistance answer time" for 142 of the 144 months under NRF.

48. When measured on an annual basis, Pacific failed to meet the GO 133-B standard of answering 80% of all "trouble service calls answer time, yearly average" (TRSAT) within 20 seconds from 1991 to 1998, but has met the standard in 1999-2001.

49. When measured on a monthly basis, Pacific met or exceeded the standard of answering 80% of all "trouble service calls answer time" (TRSAT) within 20 seconds for 92 of 132months for which we have data.

50. When measured on an annual basis, Verizon's trouble service report answer time (TRSAT) failed to meet the minimum standard of 80% of trouble calls answered within 20 seconds in 1993 and in 1995. Verizon met the standard in 1994 and 1996-2001. Verizon's performance shows statistically significant improvement over the NRF period.

51. When measured on a monthly basis, Verizon's trouble service report answer time (TRSAT) met the minimum standard of 80% of trouble calls answered within 20 seconds for 104 of the 132 months for which we have data. Verizon has met the standard every month since December 1996.

52. GO 133-B requires the exclusion of billing inquiries from its measure of business office answer time (BOAT).

53. Pacific included billing inquiries in its measure of business office answer time (BOAT) from 1992 until February 1999, but then excluded them. This complicates interpretation of this measure.

54. When measured on an annual basis, Pacific failed to meet the GO 133-B standard of answering 80% of calls to its business office within 20 seconds (excluding billing calls) in 1993, 1995, and 1996. Pacific met the standard in 1992, 1994, 1997-2001. Pacific shows a statistically significant improvement over the NRF period. Unfortunately, changes in policy over the inclusion of billing inquiries make it impossible to interpret this as an indication of improvement.

55. When measured on a monthly basis, Pacific met or exceeded the business office answer time standard 67 of 109 months under NRF for which we have data.

56. GO 133-B fails to require measurement of the time taken to answer an inquiry concerning billing.

57. Evidence in this record show that Pacific's response to billing inquiries is particularly slow, with only 50% of billing calls now answered within 20 seconds.

58. Despite GO 133-B's prohibition, Verizon included billing inquiries in its measurement of business office answer time (BOAT).

59. When measured on an annual basis, Verizon met the BOAT performance standard in 1994, 1995, and 1998-2001. Verizon failed to meet this performance standard in 1993, 1996 and 1997. Verizon shows a statistically significant improvement over the NRF period.

60. When measured on a monthly basis, Verizon met the BOAT performance standard in 80 of the 109 months under NRF for which there is data.

61. During the NRF period, Pacific's performance showed statistically significant improvement on business office answer time (although this measure has serious data problems).

62. During the NRF period, Pacific has shown no statistically significant change in the percentage of line-energizing installation commitments met, the number of customer trouble reports per 100 lines, the yearly average of toll operator assistance answer time, the yearly average of directory assistance answer time, and the trouble service answer time.

63. During the NRF period, Pacific did not show a statistically significant decrease on any GO 133-B measure of service quality.

64. During the NRF period, Verizon's performance showed statistically significant improvement on the number of customer trouble reports, trouble service answer time, and business office answer time.

65. During the NRF period, Verizon has shown no statistically significant change on the held orders, the percentage of line-energizing installation commitments met, the yearly average of toll operator assistance time, and the yearly average of directory assistance time.

66. During the NRF period, Verizon did not show a statistically significant decrease on any GO 133-B measure of service quality.

67. We find no evidence from Pacific and Verizon's performance that supports the hypothesis that NRF regulation decreases the quality of customer service.

68. The Automated Reporting Management Information System (ARMIS) data stem from FCC Common Carrier Docket No. 87-313, which implemented service quality reporting requirements for local exchange carriers such as Pacific and Verizon.

69. The FCC requires the carriers to make reports on several aspects of service quality, and the results for relevant years appear in the record of this proceeding.

70. The ARMIS 43-05 report contains service quality performance measures which track, among other things, whether Pacific or Verizon meet their installation commitments for residential and business customers, trouble reports and repair intervals (e.g., both initial and repeat trouble reports, and the time required to dispatch and complete repairs in response to trouble reports), and switch downtime incidents.

71. The ARMIS 43-06 report tracks customer perceptions of Pacific and Verizon's service quality.

72. ORA challenged the accuracy of Pacific's reports to the FCC.

73. Through the course of the proceeding, it became apparent that mismatches in data between FCC and data provided to ORA arose from differences in the data provided to ORA and the data contained in final reports provided to the FCC.

74. ORA failed to show that Pacific's historic data on service installation is inaccurate.

75. There is no need to conduct a data audit.

76. The resolution of complicated data issues requires professional cooperation, not adversarial interaction.

77. Adversarial interaction concerning issues arising from data collection invariably waste Commission time and hinders the development of a clear evidentiary record.

78. ORA did not provide sufficient evidence to permit us to conclude that Pacific is closing installation orders prematurely.

79. ORA argued that the presence of "duplicate" records among the data Pacific provided to ORA indicates that there are errors in Pacific's data.

80. ORA also argued that Pacific should include duplicate records in Pacific's calculation of its installation intervals.

81. There is no basis for finding that duplicate records among Pacific's data are erroneous.

82. ORA alleged that certain data were suspicious because ORA believed that no order for services could flow though Pacific's systems without a commitment date.

83. Pacific pointed out that some orders do not require a commitment date when a new resident takes over the phone service of an existing customer.

84. There is no basis for ORA's claim that the "anomalous" data are suspicious.

85. ORA raised issues concerning Verizon's data and Verizon successfully responded to each of them.

86. Pacific's witness, Dr. Hauser, presented a comparison of Pacific's performance on ARMIS measures against the average performance of a "reference group" of the top ten local exchange carriers in the country.

87. It is reasonable to compare the performance of Pacific and Verizon on ARMIS measures against the average performance of a reference group of similar large telecommunications utilities.

88. The ARMIS measure, "initial trouble reports," when normalized on the number of access lines in a utility, permits a comparison among carriers and over time.

89. For residential access lines, a visual inspection shows that Pacific's initial trouble reports per 100 access lines is better than that of the reference group and suggests that its performance is improving over time.

90. Statistical analysis shows that Pacific `s average residential initial trouble reports per access line is significantly better than the performance of the reference group, when statistical significance is assessed at the 1% level.

91. Statistical analysis shows that the improvement in Pacific's average initial residential trouble reports per access line is not statistically significant.

92. For business lines, visual inspection indicates that that Pacific's average number of initial trouble reports per access line is better than that of the reference group and has improved during the NRF period.

93. For business lines, Pacific's average number of initial trouble reports per access line has significantly improved over the NRF period when significance is assessed at the 1% level.

94. For business lines, Pacific's average number of initial trouble reports per access line is significantly better than the performance of the reference group of utilities when significance is assessed at the 1% level.

95. For residential lines, Verizon's average number of initial trouble reports has decreased (and improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 5% level.

96. For business lines, Verizon's average number of initial trouble reports has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level.

97. For residential lines, Verizon's average number of initial trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

98. For business lines, Verizon's average number of initial trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

99. For residential lines, Verizon's average number of initial trouble reports is less than (and therefore better than) that of Pacific, and that difference is statistically significant when significance is assessed at the 1% level.

100. For business lines, Verizon's average number of initial trouble reports is more (and therefore worse) than that of Pacific, but that difference is not statistically significant.

101. Both Pacific and Verizon show excellent performance on both the business and residential measures of initial trouble reports per 100 access lines.

102. For residential lines, Pacific's average number of repeat trouble reports has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

103. For business lines, Pacific's average number of repeat trouble reports has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level.

104. For residential lines, Pacific's average number of repeat trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

105. For business lines, Pacific's average number of repeat trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

106. For residential lines, Verizon's average number of repeat trouble reports has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level.

107. For business lines, Verizon's average number of repeat trouble reports has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 5% level.

108. For residential lines, Verizon's average number of repeat trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

109. For business lines, Verizon's average number of repeat trouble reports is less than (and therefore better than) the performance of the reference group, and that differences is statistically significant when significance is assessed at the 1% level.

110. For residential lines, Verizon's average number of repeat trouble reports is less than (and therefore better than) Pacific's, and that difference is significant when significance is assessed at the 1% level of significance.

111. For business lines, Verizon's average number of repeat trouble reports is not significantly different from that of Pacific, when significance is assessed at either the 1% or 5% level.

112. Both Pacific and Verizon show excellent performance on both the business and residential measures of repeat trouble reports per 100 access lines.

113. For residential lines, Pacific's average number of initial out-of-service trouble reports has improved since 1997. The coefficient of change for the entire NRF period is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

114. For business lines, Pacific's average number of initial out-of-service trouble reports has remained largely unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% or 5% level.

115. For residential lines, Pacific's average number of initial out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

116. For business lines, Pacific's average number of initial out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that differences is statistically significant when significance is assessed at the 1% level.

117. For residential lines, Verizon's average number of initial out-of-service trouble reports has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

118. For business lines, Verizon's average number of initial out-of-service trouble reports has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level.

119. For residential lines, Verizon's average number of initial out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

120. For business lines, Verizon's average number of initial out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

121. For residential lines, Verizon's average number of initial out-of-service trouble reports is less than (and therefore better than) Pacific's, and that difference is significant when significance is assessed at the 1% level of significance.

122. For business lines, Verizon's average number of initial out-of-service trouble reports is less than (and therefore better than) Pacific's, but that difference is not statistically significant when significance is assessed at the 1% or 5% level of significance.

123. Both Pacific and Verizon show excellent performance on both the business and residential measures of initial out-of-service trouble reports per 100 access lines.

124. For residential lines, Pacific's average number of repeat out-of-service trouble reports has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

125. For business lines, Pacific's average number of repeat out-of-service trouble reports has decreased during the NRF period. The coefficient of change is statistically different from zero when significance is assessed at 5% level.

126. For residential lines, Pacific's average number of repeat out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

127. For business lines, Pacific's average number of repeat out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that differences is statistically significant when significance is assessed at the 1% level.

128. For residential lines, Verizon's average number of repeat out-of-service trouble reports has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

129. For business lines, Verizon's average number of repeat out-of-service trouble reports has decreased (and therefore improved) during the NRF period, but that decrease is not statistically significant when significance is assessed at the 1% or 5% level.

130. For residential lines, Verizon's average number of repeat out-of-service trouble reports is less than (and therefore better than) the average performance of the reference group, and that differences is statistically significant when significance is assessed at the 1% level.

131. For business lines, Verizon's average number of repeat out-of-service trouble reports is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

132. For residential lines, Verizon's average number of repeat out-of-service trouble reports is less than (and therefore better than) Pacific's, and that difference is significant when significance is assessed at the 1% level.

133. For business lines, Verizon's average number of repeat out-of-service trouble reports is not significantly better than that of Pacific when significance is assessed at the 1% or 5% level.

134. Both Pacific and Verizon show excellent performance on both the business and residential measures of repeat out-of-service trouble reports per 100 access lines.

135. Pacific and Verizon had insufficient observations for the number of subsequent initial trouble reports and subsequent repeat trouble reports to permit a statistical finding.

136. For residential lines, Pacific's average number of initial "all other" trouble reports per 100 lines has increased (and therefore worsened) during the NRF period. The coefficient of change is statistically different from zero when significance is assessed at the 5% level.

137. For business lines, Pacific's average number of initial "all other" trouble reports per 100 lines has decreased during the NRF period. The coefficient of change is statistically different from zero when significance is assessed at the 1% level.

138. For residential lines, Pacific's average number of initial "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

139. For business lines, Pacific's average number of initial "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

140. For residential lines, visual inspection shows that Verizon's initial "all other" trouble reports is better than that of the reference group. For business lines, visual inspection shows that Verizon's performance is worse than that of the reference group.

141. For residential lines, Verizon's average number of initial "all other" trouble reports per 100 lines has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

142. For business lines, Verizon's average number of initial "all other" trouble reports per 100 lines has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level.

143. For residential lines, Verizon's average number of initial "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that differences is statistically significant when significance is assessed at the 1% level.

144. For business lines, Verizon's average number of initial "all other" trouble reports per 100 lines is not significantly different than the performance of the reference group when significance is assessed at the 1% level or 5% level.

145. For residential lines, Verizon's average number of initial "all other" trouble reports per 100 lines is not significantly different than Pacific's when significance is assessed at the 1% or 5% level of significance.

146. For business lines, Verizon's average number of initial "all other" trouble reports per 100 lines is worse than that of Pacific when significance is assessed at the 1%.

147. Both Pacific and Verizon show excellent performance on the business measure of initial "all other" trouble reports per 100 lines.

148. For residential lines, Pacific's average number of repeat "all other" trouble reports per 100 lines has increased during the NRF period. The coefficient of change is statistically different from zero when significance is assessed at the 5% level.

149. For business lines, Pacific's average number of repeat "all other" trouble reports per 100 lines has decreased (and therefore improved) during the NRF period. The coefficient of change is statistically different from zero when significance is assessed at the 1% level.

150. For residential lines, Pacific's average number of repeat "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

151. For business lines, Pacific's average number of repeat "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

152. For residential and business lines, visual inspection shows that Verizon's repeat "all other" trouble reports is better than that of the reference group.

153. Verizon's average number of repeat "all other" trouble reports per 100 lines has remained virtually unchanged during the NRF period. The coefficient of change is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

154. For business lines, Verizon's average number of repeat "all other" trouble reports per 100 lines has decreased (and therefore improved) during the NRF period, and that decrease is statistically significant when significance is assessed at the 1% level of significance.

155. For residential lines, Verizon's average number of repeat "all other" trouble reports per 100 lines is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

156. For business lines, Verizon's average number of repeat "all other" trouble reports per 100 lines is not significantly different than the performance of the reference group when significance is assessed at the 1% level or 5% level.

157. For residential lines, Verizon's average number of repeat "all other" trouble reports per 100 lines is better than Pacific's when significance is assessed at the 1% level of significance.

158. For business lines, Verizon's average number of repeat "all other" trouble reports per 100 lines is not significantly different than that of Pacific when significance is assessed at the 1% or 5% level.

159. Both Pacific and Verizon show excellent performance on the business measure of repeat "all other" trouble reports per 100 lines. For residential lines, Verizon also showed good performance on the repeat "all other" trouble reports per 100 lines.

160. Visual inspection indications that Pacific's residential initial out-of-service interval had considerable fluctuations over the NRF period, but we find no observable trend in business initial out of service intervals.

161. For residential lines, Pacific's initial out-of-service repair interval (in hours) has increased (and thereby worsened) during the NRF period. The coefficient of change, however, is not statistically different from zero when significance is assessed at the 1% or 5% level.

162. For business lines, Pacific's average number of initial out-of-service repair interval (in hours) has increased (and therefore worsened) during the NRF period. The coefficient of change, however, is not statistically different from zero when significance is assessed at the 1% or 5% level.

163. For residential lines, Pacific's initial out-of-service repair interval (in hours) is greater than (and therefore worse than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

164. For business lines, Pacific's average number of initial out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group, but that difference is not statistically significant when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

165. For residential and business lines, visual inspection shows that Verizon's initial out-of-service interval is better than that of the reference group.

166. For residential lines, Verizon's initial out-of-service repair interval (in hours) has deteriorated slightly during the NRF period. The coefficient of change, however, is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

167. For business lines, Verizon's initial out-of-service repair interval (in hours) has decreased (and therefore improved) slightly during the NRF period, but that decrease is not statistically significant when significance is assessed at the 1% or 5% level of significance.

168. For residential lines, Verizon's initial out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

169. For business lines, Verizon's initial out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group when significance is assessed at the 1% level.

170. For residential lines, Verizon's average number of initial out-of-service repair interval (in hours) is better than Pacific's when significance is assessed at the 1% level of significance.

171. For business lines, Verizon's initial out-of-service repair interval (in hours) is significantly better than that of Pacific when significance is assessed at the 1% level.

172. Since Pacific and Verizon were both subject to NRF regulation during this period, and since Verizon's performance is better than that of Pacific and that of the reference group, it is not reasonable to attribute Pacific's poor performance to NRF regulation.

173. Verizon shows excellent performance on the residential and business measure of initial out-of-service repair interval (in hours).

174. Since Pacific's initial out-of-service repair interval for residential service is worse than that of the reference group and its business service is indistinguishable from that of the reference group, we conclude that Pacific's performance in this area is below average.

175. Visual inspection indicates that Pacific's residential repeat out-of-service repair interval data appears curvilinear, where the data rises from 1994, peaks in 1998, and declines by 2001.

176. For residential lines, Pacific's repeat out-of-service repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

177. For business lines, Pacific's average number of repeat out-of-service repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

178. For residential lines, Pacific's repeat out-of-service repair interval (in hours) is greater than (and therefore worse than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

179. For business lines, Pacific's average number of repeat out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group, but that difference is not statistically significant when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

180. For residential and business lines, visual inspection shows that Verizon's repeat out-of-service interval is better than that of the reference group and relatively stable over the entire NRF period.

181. For residential lines, Verizon's repeat out-of-service repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at either the 1% level or 5% level.

182. For business lines, Verizon's repeat out-of-service repair interval (in hours) a coefficient of change that is not statistically significant when significance is assessed at the 1% or 5% level.

183. For residential lines, Verizon's repeat out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

184. For business lines, Verizon's repeat out-of-service repair interval (in hours) is less than (and therefore better than) the performance of the reference group when significance is assessed at the 1% level.

185. For residential lines, Verizon's average number of repeat out-of-service repair interval (in hours) is better than Pacific's when significance is assessed at the 1% level of significance.

186. For business lines, Verizon's repeat out-of-service repair interval (in hours) is significantly better than that of Pacific when significance is assessed at the 1% level.

187. Since Pacific and Verizon were both subject to NRF regulation during this period, and since Verizon's performance is better than that of Pacific and that of the reference group, it is not reasonable to attribute Pacific's poorer performance in residential repeat out of service interval to NRF regulation.

188. Verizon shows excellent performance on the residential and business measure of repeat out-of-service repair interval (in hours).

189. Since Pacific's repeat out-of-service repair interval for residential service is worse than that of the reference group and its business service is indistinguishable from that of the reference group, we conclude that Pacific's performance in this area is below average.

190. Visual inspection indicates that Pacific's residential initial "all-other" repair interval data appears curvilinear, where the data rises from 1994, peaks in 1997, and declines by 2001. Visual inspection indicates that Pacific's business initial "all-other" repair interval appears more stable, comparable to the reference group, and improving over the NRF period.

191. For residential lines, Pacific's initial "all-other" repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

192. For business lines, Pacific's average number of initial "all-other" repair interval (in hours) has a coefficient of change that indicates an improving trend when significance is assessed at the 5% level.

193. For residential lines, Pacific's initial "all-other" repair interval (in hours) is greater than (and therefore worse than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 5% level.

194. For business lines, Pacific's average number of initial "all-other" repair interval (in hours) is not statistically different from that of the reference group when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

195. For residential and business lines, visual inspection shows that Verizon's initial "all-other" interval is better than that of the reference group and relatively stable over the entire NRF period.

196. Verizon's initial "all-other" repair interval (in hours) has a coefficient of change that indicates a deterioration of service when significance is assessed at the 1% level.

197. For business lines, Verizon's initial "all-other" repair interval (in hours) a coefficient of change that is not statistically significant when significance is assessed at the 1% or 5% level.

198. For residential lines, Verizon's initial "all-other" repair interval (in hours) is less than (and therefore better than) the performance of the reference group, and the difference is statistically significant when significance is assessed at the 1% level.

199. For business lines, Verizon's initial "all-other" repair interval (in hours) is less than (and therefore better than) the performance of the reference group when significance is assessed at the 1% level.

200. For residential lines, Verizon's average number of initial "all-other" repair interval (in hours) is better than Pacific's when significance is assessed at the 1% level of significance.

201. For business lines, Verizon's initial "all-other" repair interval (in hours) is significantly better than that of Pacific when significance is assessed at the 1% level.

202. Since Pacific and Verizon were both subject to NRF regulation during this period, and since Verizon's performance is better than that of Pacific and that of the reference group, it is not reasonable to attribute Pacific's poorer performance in residential initial "all-other" repair interval to NRF regulation.

203. Verizon shows excellent performance on the residential and business measure of initial "all-other" repair interval (in hours).

204. Since Pacific's initial "all-other" repair interval for residential service is worse than that of the reference group and its business service is indistinguishable from that of the reference group, we conclude that Pacific's performance in this area is below average.

205. Visual inspection indicates that Pacific's residential repeat "all-other" repair interval fluctuates over the NRF period.

206. Visual inspection indicates that Pacific's residential repeat "all-other" repair interval data appears curvilinear, where the data rises from 1994, peaks in 1998, and declines by 2001.

207. For residential lines, Pacific's repeat "all-other" repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

208. For business lines, Pacific's average number of repeat "all-other" repair interval (in hours) has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

209. For residential lines, Pacific's repeat "all-other" repair interval (in hours) is greater than (and therefore worse than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

210. For business lines, Pacific's average number of repeat "all-other" repair interval (in hours) is not statistically different from that of the reference group when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

211. Pacific shows a disturbing increase in the number of repeat problems within 24-hours of the initial repair. Such a pattern is consistent with not properly correcting the problem the first time.

212. Visual inspection shows that Verizon's repeat "all-other" repair interval is better than that of the reference group and relatively stable over the entire NRF period for both residential and business customers.

213. For residential lines, Verizon's repeat "all-other" repair interval (in hours) has a coefficient of change that indicates a deterioration of service when significance is assessed at the 1% level.

214. For business lines, Verizon's repeat "all-other" repair interval (in hours) has a coefficient of change that is not statistically significant when significance is assessed at the 1% or 5% level.

215. For residential lines, Verizon's repeat "all-other" repair interval (in hours) is less than (and therefore better than) the performance of the reference group, and that difference is statistically significant when significance is assessed at the 1% level.

216. For business lines, Verizon's repeat "all-other" repair interval (in hours) is less than (and therefore better than) the performance of the reference group when significance is assessed at the 1% level.

217. For residential lines, Verizon's average number of repeat "all-other" repair interval (in hours) is better than Pacific's when significance is assessed at the 1% level of significance.

218. For business lines, Verizon's repeat "all-other" repair interval (in hours) is significantly better than that of Pacific when significance is assessed at the 1% level.

219. Since Pacific and Verizon were both subject to NRF regulation during this period, and since Verizon's performance is better than that of Pacific and that of the reference group, it is not reasonable to attribute Pacific's poorer performance in residential repeat "all-other" out of service interval to NRF regulation.

220. Verizon shows excellent performance on the business measure of repeat "all-other" repair interval (in hours).

221. Since Pacific's repeat "all-other" repair interval for residential service is worse than that of the reference group and its business service is indistinguishable from that of the reference group, we conclude that Pacific's performance in this area has been below average. Recent residential service measures have matched the reference group, and if they prove stable, Pacific's performance will be average.

222. Visual inspection indicates that Pacific's residential and business average installation interval was consistent and improving over the NRF period.

223. For average installation interval, our reference group shows a declining trend over the NRF period.

224. Utilities include vertical services, which are easy to install, in their installation interval data. It is highly likely that these orders are responsible for the declining trend for all utilities.

225. For residential lines, Pacific's average installation interval has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

226. For business lines, Pacific's average number of average installation interval has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

227. For residential lines, Pacific's average installation interval is not statistically different than the performance of the reference group when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

228. For business lines, Pacific's average installation interval is not statistically different from that of the reference group when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

229. Visual inspection shows that Verizon's average installation interval was unstable over the entire NRF period for both residential and business customers.

230. Verizon's average installation interval has a coefficient of change that is not statistically significant at either the 1% or 5% level.

231. For business lines, Verizon's average installation interval has a coefficient of change that is not statistically significant when significance is assessed at the 1% or 5% level.

232. For residential lines, Verizon's average installation interval is not statistically different from that of the reference group when significance is assessed at the 1% or 5% level.

233. For business lines, Verizon's average installation interval is not statistically different from that of the reference group when significance is assessed at the 1% or 5% level.

234. For residential lines, Verizon's average number of average installation interval is not significantly different from that of Pacific when significance is assessed at the 1% or 5% level of significance.

235. For business lines, Verizon's average number of average installation interval is not significantly different from that of Pacific when significance is assessed at the 1% or 5% level of significance.

236. Verizon shows performance on the residential and business measure of average installation interval consistent with that of the reference group.

237. Pacific's shows performance on the residential and business measure of average installation interval consistent with that of the reference group.

238. Visual inspection indicates that Pacific's switch downtime was consistent and excellent over the NRF period.

239. Pacific's average switch downtime has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

240. Pacific's average switch downtime is significantly better than the group of reference utilities when significance is assessed at the 1% level.

241. Visual inspection of the graph of switch downtime suggests that Verizon's service has deteriorated over the NRF period and that its performance is not as good as the reference group.

242. Verizon's average switch downtime has increased (and therefore worsened) over the NRF period, and it has a coefficient of change that is statistically significant at the 1% level.

243. Verizon's average switch downtime is worse than that of the reference group, but the difference is not statistically significant at 1% or 5% level.

244. Verizon's average switch downtime is worse than that of Pacific when significance is assessed at the 5% level.

245. Since Pacific and Verizon show different patterns of switch downtime and both were subject to the same NRF regulation, it is not reasonable to attribute Verizon's poorer performance to NRF regulation.

246. Pacific had only six observations for the number of switches down per switch. The statistical analysis shows that Pacific does not exhibit a statistically significant trend in the number of switches down per switch while Verizon exhibits a slight improvement in this area.

247. Pacific's performance does not show a statistically significant upward or downward trend in the number of occurrences over two minutes per switch and the percent unscheduled.

248. Pacific exhibited a downward trend for the number of occurrences under two minutes per switch.

249. Verizon has exhibited a downward trend for the number of occurrences under two minutes per switch and an upward trend for the percent unscheduled.

250. Visual inspection indicates that Pacific's residential installation "commitments met" was consistently good from 1991 to 2001, with the exception of 1997, but overall it appears to track the performance of the reference group.

251. For residential lines, Pacific's residential installation "commitments met" has a coefficient of change that is not statistically different from zero when significance is assessed at the 1% or 5% level.

252. For business lines, Pacific's installation "commitments met" has a coefficient of change that shows a worsening of service, statistically significant, when significance is assessed at the 1%.

253. For residential lines, Pacific's installation "commitments met" is not statistically different than the performance of the reference group when significance is assessed at the 1% level or 5% level. Thus, we cannot statistically distinguish Pacific's performance from that of the reference group.

254. For business lines, Pacific's installation "commitments met" is better than that of the reference group when significance is assessed at the 1% level.

255. Visual inspection shows that Verizon's installation "commitments met" tracks that of the reference group, although in 1999 Verizon shows a marked improvement in business service, and a marked decline in residential service.

256. For residential lines, Verizon's installation "commitments met" has a coefficient of change that is not statistically significant at either the 1% or 5% level.

257. For business lines, Verizon's installation "commitments met" has a coefficient of change that is not statistically significant when significance is assessed at the 1% or 5% level.

258. For residential lines, Verizon's installation "commitments met" is not statistically different from that of the reference group when significance is assessed at the 1% or 5% level.

259. For business lines, Verizon's installation "commitments met" is not statistically different from that of the reference group when significance is assessed at the 1% or 5% level.

260. For residential lines, Verizon's installation "commitments met" is not significantly different from that of Pacific when significance is assessed at the 1% or 5% level of significance.

261. For business lines, Verizon's installation "commitments met" is worse than that of Pacific when significance is assessed at the 1% level of significance.

262. Overall, Pacific's performance on installation "commitments" met is above average, while Verizon's is average.

263. As a condition of SBC's merger with Ameritech, the FCC required additional quarterly measures of state-by-state service quality. Categories of reporting for retail services include installation and maintenance, switch outages, transmission facility outages, service quality-related complaints, and answer time performance.

264. For the period July 1999 to June 2001, Pacific shows negative spikes in California in the following areas: 1) answer time performance (business customers), 2) trouble report rate per 100 lines (especially business customers), 3) percentage of installation orders completed within 5 working days (especially residential customers), and 4) percentage of installation orders delayed over 30 days (business customers).

265. Our statistical analysis of Pacific's MCOT data shows that Pacific's performance exhibits an improving trend in average answer time for residential and business customers, average trouble duration, and the report rate when significance is measured at the 1% level.

266. Pacific's operations appear largely unaffected by the Ameritech merger.

267. Our statistical analysis of Verizon's MCOT data shows that Verizon's performance does not exhibit an improving or deteriorating trend in majority of the measures.

268. Since the period for which we have MCOT data is so short and covers only part of the period subject to our investigation, it does not permit us to draw any conclusion concerning how NRF regulation affected Pacific's performance.

269. The FCC imposed a 36-month reporting requirement as a condition of the 2000 GTE merger with Bell Atlantic that created Verizon.

270. While GO 133-B measures the handling of business office calls, it does not track billing calls even though such calls account for half of the calls to the business office.

271. Verizon showed negative spikes in California on several service quality measures at the following times during the period July 2000-June 2001, as compared to the rest of that period: 1) percentage of dissatisfied customers (with business customers reporting 50% dissatisfaction in November 2000 and residential customers reporting 20% dissatisfaction in March 2001), 2) answer times (with business answer times in the 50-60 second range in September 2000 and in the 40-50 second range in January 2001 - as compared to a GO 133-B standard of 20 seconds); and residential times exceeding 20 seconds in November 2000 [30 seconds] and January 2001 [40 seconds], 3) repair intervals for both residential and business customers spiking in the period January-March 2001, 4) repeat trouble reports spiking for both types of customers in March 2001, and 5) trouble reports per hundred lines spiking in the January-March 2001 time period for residential customers.

272. In spite of these spikes, we have not observed a significant upward or downward trend in Verizon's performance for the following measures: complaints per one million lines (residential and business), the percentage of dissatisfied customers (residential and business), answer times (business), average repair interval (residential and business), the percentage of repeat trouble reports (residential and business), trouble report rates (residential and business), the percentage of orders completed within five working days (residential and business), the percentage of orders delayed over 30 days (business).

273. Verizon's performance shows slight improvement in the percentage of orders delayed over 30 days for the residential lines.

274. Verizon has shown improvement in the answer time performance for residential lines.

275. For Verizon, we conclude that despite a visual spike illustrating a decrease in the quality of service in the January to March 2001 time period, there is no statistically significant indicator of an ongoing decrease in quality.

276. ORA conducted a survey of service quality in 2001 using the same questions as it did in 1995, pursuant to Commission direction.

277. ORA's survey shows that Pacific's service quality declined over time, but a big drop in the response rate in the 2001 ORA survey from that of 1995 limits our ability to draw conclusions from the survey with statistical confidence.

278. Despite its statistical weaknesses, ORA's survey suggests that there has been a diminishment in customers' perception of Pacific's service quality.

279. The JD Power survey does not measure consumer satisfaction with Pacific's service performance, but provides an overall measure of consumer satisfaction with the company.

280. On the JD Power Survey, Pacific received a score of 110 in 2001 from J.D. Power, where 104 is the industrial average score.

281. On the JD Power Survey, Pacific ranked in the top six out of the sixteen local service providers surveyed.

282. An IDC survey of local exchange carriers found that Pacific's customers are more satisfied than the average local telephone customer for all attributes studied except one. Pacific's customers are the second most overall satisfied for customer service; Pacific's customers are the third most satisfied for voice quality; and Pacific is one of the top three providers in over 85% of the areas measured.

283. In responses on the IDC survey to customer service and voice or service quality, Pacific showed results comparable to other utilities. Because of the small sample size used in the IDC survey, it is not possible to draw conclusions with statistical confidence.

284. Market Insights conducts surveys for Pacific every month, 7-10 days following a service event to obtain information about the service interaction. The results of these surveys are reported to the FCC under ARMIS report 43-06 and to the CPUC under P.A. 02-04 in a slightly different format.

285. The Market Insights surveys indicate that Pacific's customers in the 1998-2001 period are three to six percentage points less dissatisfied than the average of the top ten LECs.

286. The Market Insights surveys indicated that Pacific's customers' dissatisfaction rose for only installation services for residential and large business customers and business office services for residential and large business customers between 1998 and 2001. The dissatisfaction declined for all other services and categories. This trend was largely counter that of the reference group of utilities.

287. ORA failed to discuss the survey results reported in P.A. 02-04 and ARMIS 43-06.

288. There is no basis for finding that the dispute over Pacific's P.A. 02-03 reporting requirements arises from anything other than confusion. The next phase of this proceeding offers the appropriate venue for resolving this confusion.

289. ORA's survey of Verizon's customers showed that service has improved since 1991.

290. Verizon's surveys of its customers show that Verizon offers good service quality.

291. Analyzing complaint data provides information on those customers having the worst experiences with telecommunications utilities.

292. Pacific has 25.4 million access lines.

293. Verizon has 6.3 million access lines.

294. Only an extremely small percentage of customers file a complaint at the Commission. In Pacific's worst year, 1997, only 0.04% of lines were affected by a service quality complaint. This is approximately 4 in 10,000 lines.

295. In 1995, Pacific's best year, only 0.02%, or 2 in 10,000 lines were affected by a service quality complaint.

296. In 1998, Verizon's worst year, only 0.03% of lines were affected by a service quality complaint. This is approximately 3 per 10,000 lines.

297. In 1995, Verizon's best year, only 0.02% of lines, or approximately 2 per 10,000, were affected by a complaint.

298. There have been six formal Commission proceedings finding problems with Pacific's service quality since 1995.

299. There is no evidence that NRF was responsible for the increase in formal complaints against Pacific, and our data indicate a general improvement in Pacific's service quality under most measures.

300. We cannot conclude, based on the record of this case, that NRF is responsible for this increase in formal complaint decisions adverse to Pacific, or that the problems that prompted these complaints would have been less significant under rate of return regulation.

301. Verizon does not show an increase in formal complaints, thereby casting doubt that NRF regulation is the cause of an increase in formal complaints.

302. There is no evidence in this proceeding that the deployment of new technologies in the telecommunications network will create classes of technology haves and have-nots.

303. There is not evidence in this proceeding that the deployment of net technologies in the telecommunications network is linked to socio-economic status.

304. In the past, eventually new technologies were ubiquitously employed throughout the public switched telephone network.

305. In the past, the deployment of new technologies has not been linked to socio-economic status.

306. Although during the NRF period Pacific dealt with economic growth and the regulatory requirements of unbundling, we do not consider these factors as legitimate excuses for poor service quality.

307. Pacific's staffing during the NRF period shows decreases followed by increases.

308. Pacific has increased its number of customer-facing staff in the last part of the 1990's.

309. The 1997-98 El Niño season with its substantially increased rainfall led to increases in Pacific's trouble reports.

310. Pacific's rate changes introduced during the NRF period had the approval of the Commission.

311. It is reasonable to continue Pacific's and Verizon's MCOT reporting requirements.

312. Verizon's customer service performance guarantee recompenses customers immediately after that suffer service problems.

313. Verizon provides this service pursuant to tariff, and there is no evidence that Verizon fails to follow its tariff.

314. Verizon's service performance guarantee is a welcome development in telecommunications service delivers.

315. Although Verizon has reduced its number of employees, we find no effect on service quality.

316. No party alleges that Verizon's mergers and structural changes have had an adverse impact on service quality.

317. The financial incentives affecting service quality under rate of return regulation and those under NRF regulation very similar but differ in degree. Both permit the shareholders to benefit from all savings that the company can realize, including those savings realized by reducing customer service. Under rate of return regulation, however, the failure to spend on service quality could lead to a lowering of rates, while under NRF regulation this result did not occur. Thus, financial incentives under NRF to reduce service quality costs exceed those under rate of return regulation.

318. Under cost-of-service regulation, this Commission rarely systematically measured or assessed service quality provided by telecommunications utilities.

319. The Commission first adopted a General Order concerning service quality and setting performance standards for telecommunications utilities in 1973.

320. In the pre-NRF period, the Commission reviewed service quality only intermittently for telecommunications utilities, with major proceedings taking place in 1976 for Pacific and 1980 for Verizon.

321. Since the adoption of NRF in 1989, the Commission has reviewed service quality in 1994 and again this year.

322. The examination of service quality in the telecommunications industry under NRF has been more comprehensive and consistent than that under cost of service regulation, and has created regulatory incentives to maintain or improve service quality.

323. No party to this proceeding claims that competition will ensure service quality in the provision of telecommunications services.

324. The key to determining how NRF regulation affects service quality is to look at and measure the performance of Pacific and Verizon, not arguments from first principles.

325. NRF created a series of regulatory and organizational incentives that increased the attention given to measuring and reviewing the service quality records produced by Pacific and Verizon.

326. Phase 3B of this proceeding will offer an opportunity for parties to suggest how to build on the record of high service quality produced under NRF and to improve on those areas of weakness in service quality.

CONCLUSION OF LAW

1. NRF's impact on service quality was a key concern when we adopted the new framework in 1989.

2. Pursuant to the plan of this proceeding and our previous rulings, we have conducted extensive fact finding on the quality of service in this phase of the proceeding.

3. Although D.89-10-031, the decision establishing the NRF framework, did not institute specific service quality reporting requirements, the Commission subsequently adopted specific monitoring requirements and has periodically and systematically examined service quality.

4. Any changes that we make to NRF should be coordinated with revisions to GO 133-B that result from the rulemaking we recently opened to make such revisions, Rulemaking 02-12-004.

5. GO 133-B defines a held order as "[r]equests for primary (main) telephone service delayed over 30 days for lack of utility plant."

6. Pacific's method of counting held orders fails to comply with GO 133-B.

7. It is reasonable to require Pacific to continue to report the FCC's MCOT results until further notice of the Commission.

8. It is reasonable to require Verizon to continue to report the FCC's MCOT results until further notice of the Commission.

9. It is reasonable to use GO 133-B measures to assess Pacific and Verizon's service quality.

10. It is reasonable to use the FCC's ARMIS measures and a reference group to assess Pacific and Verizon's service quality.

11. It is reasonable to use statistical methods to assess Pacific and Verizon's service quality.

12. It is not reasonable to draw conclusions based on single deficiencies in performance since service quality depends on a company's overall performance.

13. It is appropriate to consider the results of formal complaints and other formal Commission proceedings initiated during the NRF period in assessing Pacific and Verizon's service quality.

14. The Commission intends to consider clarifying the meaning of the term "primary (main) telephone service" in GO 133-B in its Service Quality OIR, R.02-12-004. GO 133-B's reference to "primary (main) telephone service" is unclear and requires clarification. It is unclear whether the quoted phrase refers to a class of service that includes basic exchange service and that the sequence of lines to an address is not a factor in the definition of primary service; or whether "primary (main) telephone service" refers only to the first line into a home. However, this is not the appropriate proceeding to revise this term.

15. GO 133-B specifically excludes billing inquiries from its measure of BOAT.

16. Pacific did not violate Pub. Util. Code § 451 in connection with its answer times for billing calls.

17. The Commission intends to consider whether to include billing calls within the GO 133-B standards in its Service Quality Order Instituting Rulemaking.

18. It is appropriate pursuant to Commission Rule 73 we take official notice of the Commission's actions in the complaints or other formal proceedings as cited herein.

19. It is reasonable to use survey data in assessing customer perceptions concerning service quality.

20. It is reasonable to use statistically valid methods to determine the evidentiary weight to assign to a specific survey instrument.

21. It is reasonable to conclude that meeting GO 133-B standards adopted by the Commission is a sign that a carrier is providing good service quality.

22. It is reasonable to conclude that a utility that earns better scores on ARMIS service quality measures than the scores of a reference group that includes the major large utilities is providing good service quality.

23. It is reasonable that any assessment of the quality of service offered by a telecommunications utility focus on direct measures of service quality.

ORDER

IT IS ORDERED that:

1. Pacific Bell (Pacific) shall conform its method of counting orders held over 30 days to the requirements of General Order (GO) 133-B as stated in this decision. Within 30 days of the effective date of this decision, Pacific shall file a compliance document in this docket demonstrating its compliance with the requirements of this decision with regard to the calculation of such held orders.

2. Pacific shall continue to report data to this Commission for measures required under the Federal Communications Commission's (FCC's) Merger Compliance Oversight Team (MCOT) requirements that expired in November 2002 until further notice of the Commission.

3. Verizon California, Inc. (Verizon) shall continue to report data to this Commission for measures required under the FCC's MCOT requirements contained in its order FCC 00-221 until further notice of the Commission.

4. We deny the Office of Ratepayer Advocates' (ORA's) recommendation that we conduct an audit of Pacific's or Verizon's historic installation data to determine the extent of data error and its subsequent impact on reported service quality results during the New Regulatory Framework (NRF) period

5. Pacific shall file and serve data in the form of a compliance filing in this docket that answers the questions concerning closed installation orders containing multiple lines as enumerated herein within 30 days of the effective date of this decision.

6. We deny The Utility Reform Network's (TURN's) request for a finding that Pacific has violated Pub. Util. Code § 451 with regard to its billing calls.

7. The Commission will consider in Phase 3B of this proceeding what regulatory actions it should take to ensure the continuation of high quality service by Pacific and Verizon and the improvement of service, where necessary or possible. The parties shall address such issues in their Phase 3B testimony.

8. Verizon shall notify us in advance if it seeks to discontinue reporting billing inquiries in its Business Office Answer Time (BOAT) results to make any other change in the types of calls it includes in its BOAT reporting.

9. Neither Pacific nor Verizon shall change the way they count their GO 133-B results (except as ordered herein) without advance permission of this Commission.

10. The Commission will consider in Phase 3B of this proceeding how to resolve the issues concerning the reporting of survey data under the P.A. 02-03 and P.A. 02-04 filing categories as described herein. The focus of the Commission's consideration is to determine whether additional unreported data exists from the period under review and how Pacific and Verizon should file survey data on a going forward basis.

11. The parties shall address any needed regulatory changes related to the findings this decision makes in Phase 3B of this proceeding.

This order is effective today.

Dated October 30, 2003, at San Francisco, California.

MICHAEL R. PEEVEY

President

GEOFFREY F. BROWN

SUSAN P. KENNEDY

Commissioners

I reserve the right to file a concurrence.

/s/ GEOFFREY F. BROWN

Commissioner

I will file a concurrence.

/s/ SUSAN P. KENNEDY

Commissioner

I reserve the right to file a dissent.

/s/ LORETTA M. LYNCH

Commissioner

I reserve the right to file a dissent.

/s/ CARL W. WOOD

Commissioner

APPENDIX A

END OF APPENDIX A