X. Verizon's Service Quality Performance

Verizon's service quality results were better than Pacific's, but also showed problems in some areas. Verizon should focus on making improvement in the following areas:

      · Residential and business installation intervals and business installation commitments met (see Section entitled "Installation - Verizon," below).

      · Business trouble reports for repairs (see Section entitled "Repair - Verizon," below).

      · Staffing levels (see Section entitled "Staffing - Verizon," below).

As it does for Pacific, ORA alleges that under NRF Verizon - albeit to a lesser extent than Pacific - has "reduced [its] quality of service, grossly inflated staffing claims, . . . moved portions of the labor force out of California . . . , and had sustained facilities shortages. . . ."192

TURN cites several specific problems with Verizon that allegedly support its claims about NRF. It states that "like its TRSAT, Verizon's BOAT was often below the GO 133-B standard, until shortly after the SBC/Pacific Bell merger decision, wherein the Commission stated that it would enforce the standards."193

While Verizon points out that TURN's witness Gayatri Schilberg conceded on cross examination that one of the areas in which Verizon's performance had improved under NRF was the BOAT results, in fact Ms. Schilberg's testimony was not so clear cut. Rather, she emphasized that improvements only occurred when the Commission took action to enforce its service quality requirements, a point consistent with TURN's claim that regulatory intervention is necessary to ensure quality:

TURN also cites the Commission's decisions twice penalizing Verizon for marketing abuses in connection with its Language Assistance Center.195 GTE, Verizon's predecessor, agreed to pay $13 million to settle a case alleging that sales staff at its foreign Language Assistance Center charged non-English speaking subscribers for optional services, such as Call Waiting or Call Forwarding, which the customer did not order during the 1989-92 period. We found in 1998 that,


    [T]he information provided to the Commission in 1992 regarding marketing abuse was incomplete because GTEC wrongfully informed the Commission that the abuses were short-term in duration and discovered through `routine quality control procedures.' Contrary to GTEC's representations, both reports contend that there is evidence which indicates the marketing abuses sporadically occurred beginning in 1989, rather than 1992, and were discovered through non-routine monitoring of customer calls, rather than routine monitoring.196

Although the underlying abuses occurred a decade ago, the true facts concerning its magnitude did not come out until the Commission's Consumer Services Division (CSD) and GTEC each conducted investigations in 1997. Before that time, Resolution T-15404 had found that the abuses were related only to the 1992 time period. After the 1997 investigation, the parties both found that the abuses were far more widespread than originally thought and covered at least a 3-year period. (Indeed, CSD contended the abuses continued after 1992.197)

Because GTEC did not bring these facts to light until 1997, it is not fair to say that its misconduct ended in 1992. Rather, the company continued to conceal the true facts until the 1997 investigation. Thus, we agree with TURN that this case is relevant to our determination regarding Verizon's service quality during the NRF period.

Otherwise, the parties make no allegations with regard to Verizon specifically, but rather include Verizon in their general allegations about the effects of NRF on service quality. We therefore incorporate by reference our general discussion from the Section entitled "NRF Incentives and Service Quality - Pacific," above, and note that the same general conclusions apply to Verizon. We find some merit in the claim that NRF creates incentives to save money at the expense of service quality, but analyze Verizon's specific performance in later sections of this decision.

Verizon responds that "Verizon's service quality results are compelling evidence that NRF gives strong incentives to provide high quality service." Thus, it agrees that we must examine its specific service quality results in order to determine the veracity of TURN's claims. However, Verizon also claims that NRF "encourages carriers to focus on service quality," citing several measures that Verizon has employed that go beyond the bare bones reporting that this Commission and the FCC require.198

We do not find that NRF itself causes carriers to establish self-imposed service quality measures. Indeed, the fact that Pacific could not identify any specific service quality measures it imposes that exceed the standards the FCC and we require (See the Section entitled "Self Monitoring - Pacific," above) helps disprove Verizon's hypothesis. If it were true that NRF itself causes carriers to make "customer service results . . . an integral part of . . . management practices,"199 then we would presumably see Pacific imposing the same type of voluntary service quality standards as Verizon claims. The evidence did not establish this fact.

Nor did the decision establishing the NRF framework institute particular service quality reporting requirements, as TURN points out.200 Instead, the NRF decision cited the Commission's need to be vigilant about service quality lest incentives to cut costs caused the carriers to cut too deeply.

TURN focuses its argument in this area on Pacific. While it initially notes that "Recently, Verizon's California predecessor (GTEC) and Pacific Bell have respectively been merged into the nation's largest and second largest carriers,"201 it identifies no specific problems stemming from the Verizon merger. TURN notes the FCC's MCOT requirements stemming from the Verizon-Bell Atlantic merger expired in November 2002, but in its motion seeking an order continuing Pacific's parallel reporting requirements, TURN stated that Verizon agreed voluntarily to continue these reporting requirements until after a final decision issues in this proceeding. To make Verizon's obligation parallel the order we make here regarding Pacific's MCOT reporting, we will require Verizon to continue to report MCOT data to this Commission until further notice. We agree with TURN that we should consider the usefulness of MCOT data in Phase 3B of this proceeding and determine whether we should require the carriers to continue to report such data even after their merger obligations expire.

Moreover, the assigned Administrative Law Judge made clear during the hearing that regulatory changes in this area are outside the scope of Phase 2A, and instead should be addressed in Phase 3B.202 We therefore defer this issue as to Verizon to later in this proceeding.

We note that there is one significant difference between Pacific and Verizon in the area of advanced services such as DSL, on which TURN focused much of its concern. While Pacific continues to offer its advanced services in a separate affiliate, Verizon seeks to transfer those services back to the regulated utility.203 If granted, the transfer may limit the concerns TURN raises, but it is premature to address this issue in this phase.

In the meantime, however, the record establishes that TURN has some reason to be concerned. Verizon's service quality witness reported that she does not have access to data regarding the company's DSL service quality because DSL is provisioned out of a separate subsidiary:


    Q. Do you have the same internal service quality standards for advanced technologies such as DSL or broadband as you do for POTS204 service?


    A. As you well know, DSL is provisioned out of a separate subsidiary. So I don't have access to their actual customer satisfaction results or their internal metrics. So I can't really answer your question.205

The witness went on to state that even if the Commission approves Verizon's application to return its advanced services subsidiary to the regulated utility, it will track DSL and POTS service quality separately.206 This separate reporting may continue to make Verizon's GO 133-B and ARMIS results look better than they would be if DSL results were also included. Whether housed separately or only reported separately, advanced services results are key to our understanding of Verizon's service quality. In Phase 3B, parties should be prepared to address how we can ensure that we have a complete picture of Verizon's service quality - including that of its DSL operation.

TURN's allegations in this area relate only to an argument Pacific raises. Pacific claims that basic service is subsidized and that it lacks an incentive to cut costs in that area because no amount of cost cutting would make the service profitable. Verizon does not address this argument, and therefore we simply incorporate by reference here the section entitled "Incentives to Cut Costs - Basic Service - Pacific," above.

As it does with regard to Pacific, TURN disputes any notion that competition necessarily improves service quality: "Their [Pacific and Verizon's] theoretical argument, such as it is, rests on the thin air of hypothetical `competition.'"207 Verizon does not address this point. Because this is the same argument TURN made with regard to Pacific, we incorporate our discussion of the issue by reference here.

Before addressing Verizon's specific performance data, we address ORA's criticisms regarding the accuracy of Verizon's performance data. We find for the most part that ORA's evidence on this point lacked merit. Despite this finding, we believe Verizon shows problems in the areas of business and residence installation intervals and business installation commitments met, business trouble reports for repairs, and staffing levels. We discuss those results after analyzing the accuracy of Verizon's data.

We do not find merit in ORA's claims that Verizon's performance data are erroneous. Rather, we find that the concerns ORA's Linette Young raised are the result of a mismatch between her data and the data on which Verizon bases its results, of miscommunication between the parties, and of other reasons for which neither party is blameworthy.

For example, ORA contended in testimony that Verizon's installation data are inaccurate because they contain an excessive number of duplicates. However, it later became clear that the records that appeared to be duplicates in fact were separate records with different service order numbers, and not duplicates at all. When Verizon tested ORA's data using the proper definition of duplicate records, it determined that there were no duplicates in any of its data sets. Since ORA did not mention its contention about duplicate records in briefing, we conclude it no longer disputes Verizon's claim.

ORA also offered testimony that Verizon's data are inaccurate because installation intervals from different data sets do not match. However, Verizon demonstrated that these separate sets contain a different mix of products or employ different criteria to measure installation intervals. As a result, the installation intervals from these data sets were not supposed to match. ORA made no further mention of this issue in its brief, and we do not find a problem in Verizon's data on this issue.

ORA had a more valid criticism when it came to calculating Verizon's installation intervals. Verizon is open and performs installation on weekends, but does not count weekend days as "working days" in determining whether it has met installation standards imposed on it as part of its MCOT (merger) reporting. ORA included days that Verizon is open that fall on weekends as "working days" - a stance that makes sense because Verizon is "working" on those days - and came up with different installation intervals than Verizon. However, since ORA calculated the interval based on a different period than did Verizon, it is no surprise that the two sets of data did not match.

The real concern with this interval data is not that Verizon's data is unreliable, but that Verizon may be misinterpreting the meaning of the FCC-imposed requirement of counting "working days." Verizon's Linda Thoms testified that she had discussed the matter with the FCC and was told that Verizon was interpreting the term correctly to exclude weekends and holidays. ORA introduced email correspondence between its witness and an FCC staff person who did not "recall making any changes to [Verizon's] business rules for the consumer service quality reporting requirements . . . that would change days to `business days.'" Ultimately, we find the evidence is inconclusive on whether the FCC means the term "working days" to include weekend days on which the carrier actually does business. While the more sound interpretation in our view is that it does, we are dealing with an FCC requirement that we did not institute.

The only issue for our determination related to the "working days" definition is whether Verizon mis-reports its results. Because of the definitional ambiguity, we do not find that ORA has established anything misleading in Verizon's data on this point.

ORA also offered testimony that Verizon is mis-reporting data because its installation data do not match the information in certain informal complaints on file with the Commission. However, ORA admitted that the information in the complaints might not be accurate because the Consumer Affairs Branch representatives who take the complaints rely on customers' statements identifying the dates problems arose. It would not be surprising for customers' recollections of precise dates to vary from Verizon's own computer records. We do not find this variance to raise any red flags about the accuracy of Verizon's reporting.

ORA also challenged Verizon's performance data because Verizon reported commitments met in the mid- to high 90th percentile range for 1998-1999, whereas ORA came up with an inexplicably low percentage - 0.88%, or lower than 1 percent - for these years. While ORA's witness did not concede that such a low result was obviously wrong, it is clear that it was the result of a miscalculation. As it turned out, ORA used a different definition of "commitments met" than did Verizon, resulting in the ORA calculation error.

While ORA explained that it asked for and received the incorrect definition from Verizon, causing the error, in fact this is not what happened. ORA asked Verizon for the definition using the present tense, and Verizon gave it an answer about its current behavior. While Verizon should probably have followed up with ORA to clarify its question, it answered the question posed. ORA then assumed that Verizon's definition was the same one it had used in the past, and came up with a mismatch in the data. While this miscommunication was unfortunate, we find that both sides acted unintentionally but that there is no underlying misrepresentation in Verizon's data.

ORA's witness also contended a Verizon affiliate falsified service quality data in New York, but dropped this claim when it was revealed that the New York Public Service Commission found the claim to be unfounded.

We also reject ORA's final contention about the accuracy of Verizon's data. ORA concluded that because Verizon's data included more than one installation order for the same phone number within a single 60-day period, Verizon must be prematurely closing installation orders without resolving them to make its figures look better. ORA assumed that if it could not complete the installation quickly, it would simply close out the installation order without doing the work, and then open a new order for the same work to conceal the delay.

Verizon's testimony contradicted this contention. Verizon pointed out - and we agree - that it is normal for its data to contain two installation orders for the same number within the same 60 days:


    For example, a customer may order basic service at one location and move to a new location within 60 days. If the customer retains the original number at the new location, Verizon's records will appropriately reflect two installation orders for the same service created within 60 days. Similarly, a customer may inadvertently order basic service at the wrong address or a representative may inadvertently record the wrong address. If Verizon installs service at the wrong address and then is required to install service at the correct address, Verizon's records will appropriately reflect two installation orders created within a 60-day period.208

Moreover, we agree that Verizon adequately explained the specific examples ORA provided in which Verizon's data contained more than one record for the same phone number within a 60-day period.

ORA made a similar contention with regard to trouble tickets, arguing that Verizon prematurely closes tickets because "1.88% of the residential repeat out-of-service repairs in 2000 occurred within 24 hours of a previous repeat out-of-service repair."209 Verizon contends the number of such repairs is actually much smaller - 392 out of approximately 600,000 repairs.210 We cannot reconcile the difference in numbers, but because repeat out-of-service repairs represent a serious service quality problem, we do not take this evidence lightly. However, this is more an actual problem in service quality than substantiation of a claim that Verizon mis-reports its results, and we deal with this issue in the Section entitled "Repair - Verizon," below.

Overall, we find that Verizon had some performance problems in its installation intervals over the relevant period. We also find similar problems with respect to its installation "commitments met" performance for business customers.

TURN points out that, on certain installation measures, Verizon has shown a "pattern of both worsening and improving performance." Verizon explains that "a certain amount of variability on a performance measure is to be expected."211 We do not agree with the general principle that significant declines in service quality are acceptable just as long as the tide ultimately turns positive. Rather, we expect from all carriers steady, positive performance rather than performance that is irregular or inconsistent. Therefore, it is not a defense to TURN's claim that performance naturally ebbs and flows.

We have compared Verizon's and Pacific's installation performance both on installation intervals and installation commitments. The results appear below.212 With regard to installation intervals, the graph shows that Verizon performed less well than did Pacific for both residence and business installations from 1995-99. In 2000-01, Verizon California's213 performance improved: average installation intervals for residence customers decreased from nearly 5 days in 1998 to under 1 days in 2000 and 2001, while the same interval for business customers went from nearly 7 days in 1998 to just over two days in 2000 and 2001. Nonetheless, Verizon's installation intervals (business) were at 4 days or more from 1995 through 1999, as shown in the following graph:

Likewise, Verizon's installation intervals (residence) increased from about 1.5 days to over 4.5 days between 1994 and 1998, as shown in the following graph:

As the foregoing graphs reveal, Pacific's installation intervals were generally better than Verizon's during the NRF period, with business installation intervals remaining stable in the 3-4 day range during the entire period 1994-2001. Residence intervals were not as steady, with small spikes in 1995 and 1997, but the overall numbers were generally lower than Verizon's except in 1994-95 and 2000-01.

Data on installation "commitments met" track the percentage of cases in which a carrier installs according to the schedule it promised the customer. The ARMIS results for Verizon and Pacific appear below. They show the following:

Other than in 1999, when Verizon's percentage of residential commitments met dipped to below 97%, Verizon performed well during the 1993-2001 period on its residential commitments. Its performance did not vary much from Pacific's, and neither party shows major problems in the "commitments met" area during the 1993-2001 period.

Verizon's results were less stable in the area of business commitments met, as the foregoing graph reveals. Verizon's results showed a general declining trend between 1991 and 1998 and were most problematic in 1995 and 1998, dipping to 96% and 95.5% of commitments met for business customers in those years. For all years except 1999, the data show that Verizon's performance was worse than Pacific's.

Overall, the data show that Verizon's installation intervals were problematic, and that its "commitments met" performance also showed problems with respect to business (but not residential) customers.

TURN acknowledges that Verizon's long switch downtime incidents have decreased dramatically since 1991, but notes that this change occurred after regulatory action in D.94-06-011. There, GTEC (Verizon's predecessor) agreed to improve its major service interruptions by 30% from 1992 levels within 3 years, with a 45% improvement attributable to its GTD-5 switches. TURN therefore asserts that it is only through regulatory action that we can assure improvements in service quality.

We do not agree that service quality never improves without regulatory action. Verizon, for example, has a number of internal mechanisms to ensure positive service quality results that go beyond simple regulatory requirements. Indeed, we are impressed by the way in which Verizon supplements regulation with its own self-imposed expectations, and are disappointed that Pacific did not claim to use its own internal standards of excellence.214

For example, recognizing that it is difficult to compare ARMIS and GO 133-B results across carriers, Verizon uses other data to "benchmark" itself with other service providers.215 Verizon also has internal standards for installing or restoring service in a timely way and for completing work correctly the first time that go beyond GO 133-B standards.216

By the same token, we believe it is important to retain - and even expand where necessary - the reporting requirements we currently impose as part of GO 133-B or elsewhere. Indeed, we suggest that several holes in the GO 133-B requirements be plugged and that ambiguous language be clarified. Ultimately, we agree with the general premise we cited when we first instituted NRF - that pressure to lower cost and improve efficiency also poses the risk of reducing service quality. Therefore, we will continue closely monitoring service quality for the foreseeable future.

Verizon's repair performance showed some problems during the NRF period, especially in the area of trouble reports. In the area of repair intervals, Verizon's performance was consistently good through much of the NRF period, but has worsened in 2000-01, especially for residential customers. Nevertheless, Verizon's repair interval data throughout the NRF period was consistently better than Pacific's.

In our analysis of Verizon's ARMIS results for trouble reports, Verizon's problems appeared primarily in the context of its business customers. For business customers, Verizon's results were worse than Pacific's for both initial trouble reports for conditions other than out of service conditions and repeat trouble reports for other than out of service conditions, as follows:

ORA emphasized one measure in particular, noting that 1.88 percent of the time, Verizon reports residential repeat out-of-service repairs within 24 hours of a previous repeat out-of-service repair.217 (Verizon gave the figure as a much smaller figure: 392 out of approximately 600,000 repairs in 2000.) While the number is small, it gives us concern because of the magnitude of the problems these customers experience. These are customers whose phones are completely out of service, and who have already had one repair visit. They then have two more repair visits before Verizon can resolve the problem.

According to TURN's witness, "repair times have been increasing for Verizon's (GTE California) residential customers. By 2001, the average repair interval for residential customers was the worst yet recorded since 1994 in ARMIS data for Verizon (GTE-CA) (Table 43-05) in the areas of initial out-of-service reports . . ., initial reports other than out-of-service . . ., repeat out-of-service reports . . ., and repeat reports other than out-of-service. . . ."218

ORA claimed that Verizon's business annual average repair intervals were not stable in 2001 and 2002. Verizon agreed that these intervals varied between 10 and 15 hours during this time frame.219 While ARMIS does not set standards for these intervals, Verizon concedes that its 2001 ARMIS average annual repair intervals for certain residential and business customers increased from the levels it achieved in prior years.220 While Verizon explained the steps it took in an attempt to remedy the situation, it is still too soon to know if the improving trend will continue. The worst performance occurred in the first quarter of 2001. Verizon attributed this poor performance to unusually heavy and prolonged rains in that quarter, but its witness also acknowledged that rain is an expected event for which the company should be prepared.221

According to the testimony, which covered a portion of 2002, Verizon's residential repair intervals improved somewhat in the first quarter of 2002. It is not clear whether this improvement represents a short- or longer-term phenomenon. Moreover, our charge in this proceeding is not only to examine Verizon's service quality at this precise moment. Rather, we must examine performance over the entire NRF period to look for trends or significant deterioration in quality that might give us reason to be concerned for the future of service quality. Thus, while it is true that most of Verizon's results show improvement, there were also times during the NRF period when they were problematic, and we rely on both performance trends in reaching our conclusions here.

While Verizon introduced 2002 repair data, TURN correctly points out that Verizon takes 3 months of 2002 data and annualizes it in order to reach the conclusion that results have improved significantly since its acknowledged problematic performance in early 2001.222 If, as TURN advocates, one compares Verizon's earlier performance to its 2001 data, its residential initial trouble reports have declined by only 3% from 1995 to 2001 (as compared to the 18% decline Verizon claims by comparing 1995 to its annualized 2002 data). TURN concludes that "[s]tarting in 1999 Verizon's residential repair intervals steadily worsened and last year were the worst ever. Verizon claims that there have been gains in the early months of 2002, but it remains to be seen if the final results for the year will show any actual improvement."223

While the parties tended to focus on the recent deterioration in Verizon's repair intervals, it is noteworthy that Verizon performed better than Pacific on all measures: 1) initial out of service repair intervals - residence; 2) initial out of service repair intervals - business; 3) initial all other repair intervals - residence; 4) initial all other repair intervals - business; 5) repeat out of service repair intervals - residence; 6) repeat out of service repair intervals - business; 7) repeat all other repair intervals - residence; and 8) repeat all other repair intervals - business, as shown in the Section entitled "Repair - Pacific," above.

Overall, our greatest concern is with Verizon's trouble report performance for its business customers. We are also somewhat concerned with the deterioration in Verizon's repair intervals in 2000 and 2001, especially for residential customers. Any further deterioration will be cause for more serious concern.

GO 133-B requires that the carriers report on answer times in both BOAT and TRSAT reports. We have prepared graphs showing Verizon's performance in both areas, which we reproduce below. As previously noted, the data used in the graph has been adjusted for Pacific to include billing calls for the period beginning May 1999 and thereafter to make the data comparable for all years and for both companies.224

We find that Verizon's BOAT results were superior to Pacific's in 1995, 1996 and 1998-2001. However, on average, Verizon's BOAT results failed to meet the minimum standard of 80% of calls answered within 20 seconds during the period from 1993 through 1997.225 Verizon's BOAT performance was clearly substandard during the early part of the NRF period, but has shown steady improvement since 1997.

(The minimum standards were 70% beginning December 3, 1992; 75% beginning October 4, 1993 and 80% beginning July 5, 1994.)

We note that the Commission in 1994 approved a settlement in which GTEC agreed to implement a service guarantee program.226 Under the program, if Verizon failed to meet the G.O. 133-B answer time standards for any three months of a six-month moving period, Verizon would be obligated to provide a refund to ratepayers. Verizon's BOAT performance thereafter showed improvement.

Verizon's TRSAT results were generally better than Pacific's from 1994-2001 (except in 1995 when the two carriers' results were about equal). Verizon's TRSAT results failed to meet the minimum standard of 80% of calls answered within 20 seconds from 1991 through 1993 (with extremely poor performance in 1992) and in 1995.

According to the testimony, since 1994, there have been four months in which Verizon failed to meet the standard under TRSAT (Trouble Report Service Answer Time).227 ORA's witness testified that Verizon had met the TRSAT standard 90% of the time since 1992.228 We do not agree with TURN that these results for Verizon can be attributed to the Commission's action against Pacific after the Commission's Pacific Telesis-SBC merger order (D.97-03-067) threatened Pacific with sanctions for noncompliance with the TRSAT standard. The connection between the two actions is not only too tenuous to draw conclusions with any certainty, but is also undermined by the fact Verizon generally exceeded the TRSAT minimum standard after D.94-06-011 and before the Commission issued the merger order.

Verizon uses Automated Response Units (ARUs) or Interactive Voice Response Units (IVRUs) "to facilitate provision of customer service." While Verizon correctly points out that the Commission also uses ARUs in some circumstances, it is not the use of ARUs, per se, about which we are concerned. Rather, it is that carriers' reported call answering times - which we prescribe at very low levels numbering in the seconds - do not include the time a customer spends navigating ARU options.

Our call answering standards mean little if a customer spends more time navigating the ARU than waiting for a live operator. For that customer, the time spent navigating and waiting is far longer than the standard we prescribe "for the trouble report service attendant to answer trouble report calls" or "for the business office representative to answer business office calls."229 Because our current rules do not contemplate the use of ARUs or IVRUs, we are addressing this issue generically in our Service Quality OIR.

GO 133-B defines "Business Office" as "A Centralized Service Group which receives Small Business and/or Residence Customer requests for new installation or change in existing service. This does not include billing center inquiries."230 However, Verizon states that it includes billing inquiries in its BOAT measure.231 We do not wish to discourage such voluntary over inclusion, but we will require Verizon to notify us if it seeks to discontinue reporting billing inquiries. As we note elsewhere in this decision, one important use of the GO 133-B data is that we can use it to analyze a carrier's performance over time. Such comparability requires that a carrier seek prior Commission authorization before making changes to the way it reports its data.

Overall, we find that, while Verizon had serious problems with respect to both BOAT and TRSAT in the early NRF period, Verizon appears to have developed a consistent track record of solid performance since then.

Verizon offers its customers a "service performance guarantee" (SPG) when customers believe - "rightly or wrongly"232 - that Verizon has delivered problematic service.233 We wholeheartedly support the SPG program as a good way to offer recompense to customers immediately after they suffer service problems. However, in order for such a program to work fairly, Verizon should ensure it properly discloses the SPG to all customers. Verizon's witness gave inconsistent responses to questioning on whether Verizon discloses the availability of SPGs to customers. First she indicated that Verizon's service personnel do not disclose the SPG on every call:


Q.So when a customer calls and places an order, is that customer always told, "By the way, if you're not satisfied, we have this performance credit?


A. No. That's not what we say. We say we guarantee our service, and to meet our commitments to you; and if we fail to do that, call us back, and we will stand by our commitment.


Q. Okay, but . . . reps don't necessarily say, "by the way, there's this credit"?

Then, the witness said that in fact Verizon's representatives were required to disclose the availability of the credit up front:


A. We say, here's what we direct them to say. "We guarantee your service will be working on" - and this happens to be an installation practice - "and if we miss this commitment, please call us, and Verizon will issue you appropriate credit."

. . .


    Q. So does the result of your . . . checking of what . . . customer-care representatives do . . . verify to your satisfaction that, in fact, on every . . . initial installation call, the customer is told up front that they can get a credit if they want it?


    A. Certainly not on every one, but again, based on our audits and reviews, when we find that the rep has failed to do that, then we go back and provide the appropriate education and coaching. 234

Because a customer must request the credit in order to get it - "it's our procedure that the customer requests the credit"235 - it is very important that every customer know of the credit up front in order for it to be applied fairly. However, based on the foregoing ambiguous testimony, it is not clear that Verizon tells every customer of the credit. If Verizon does not, it is highly probable that customers who truly deserve SPGs are not receiving them. These customers may actually have experienced a missed commitment based entirely on objective measurements. However, if the customer does not affirmatively request a credit, we are concerned that the customer does not receive the immediate compensation the SPG program is designed to deliver.236

We require more information on this topic, and will examine during Phase 3B the extent to which Verizon discloses its SPG program to all customers. At the very least, Verizon should address the following issues in its Phase 3B testimony:

We agree with ORA that if customers are not being told of the availability of SPG credits, and therefore are not being granted a credit that a similarly situated customer might receive, there are service quality implications.

Verizon should not interpret our comments as a criticism of the usefulness of a service guarantee generally. To the contrary, a properly disclosed and applied program is a good tool for compensating customers as soon as they suffer harm, and for motivating carriers to improve service quality. Verizon shall not discontinue its program as a result of our comments in this decision. However, we believe the program as presently applied may require modification, and ask for the follow-up information above to determine whether we should take further action.

TURN cites two formal proceedings that it states show problems with Verizon's service quality:

      · A.92-05-002/D.94-06-011 regarding GTEC (Verizon's predecessor) answer times and switch outages.237 The Commission found that GTEC's answer times failed to meet minimum GO 133-B standards. For example, GTEC failed to meet the G.O. 133-B answering time standard for its Customer Care Centers in 17 out of the 24 months in 1991 and 1992. For the Customer Billing Centers, the average speed of answering time was approximately two minutes: 126.1 seconds and 113.1 seconds, respectively.238 GTEC also had a high customer billing error rate, a disproportionately high number of informal complaints, inconsistencies in its service quality monitoring data and problems with its calling cards.

      · C.98-04-004/D.98-12-084, approving GTEC's payment of $13 million to settle marketing abuse claims stemming from the period 1989-92.239 However, we later found that we did not have all the facts surrounding the abuse in requiring GTEC to distribute $ 3.2 million among local groups within the Hispanic community for the purpose of telecommunications education and to report the names of recipients and amounts of contributions above its normal contributions. At that time, we imposed no punitive fines against GTEC. Only after parallel investigations by GTEC and the Commission in 1997 did the true facts about the extent of the abuses come to light, and at that time GTEC agrees to make a $13 million "civil payment" and a personal apology to the Commission.240

Nevertheless, Verizon's formal complaint history during the NRF period compares favorably to Pacific's record. Both formal complaints against Verizon relate to conduct early in the 1990s and before. While GTEC did not bring to light the true facts surrounding the marketing abuses in the second case until 1997, the Commission did not find that the abuses themselves continued after 1992.

Therefore, Verizon's formal complaint history, standing alone, does not indicate a significant service quality problem.

The OII initiating this proceeding also attached Verizon's informal complaint record, as follows:

Number of Informal Complaints Filed at the Commission

Verizon - January 1, 1995, through July 12, 2001

 

Category of Complaint

1995

1996

1997

1998

1999

2000

2001

1

Delayed Orders & Missed Appoint.

20

7

44

94

44

80

44

2

Quality of Service (e.g., static, crossed lines, intermittent service, etc.)

183

250

243

217

193

188

77

3

Disputed Bill

502

655

767

807

489

692

365

4

Disconnections

29

56

61

106

61

59

35

5

Deposits

39

44

47

21

23

22

7

6

Disputed Customer of Record

27

21

53

59

67

37

12

7

No Notice

14

31

22

19

26

0

0

8

Late Payment Charge

3

3

5

7

4

0

0

9

Rate Design

300

28

47

67

9

9

6

10

Rules

20

52

74

69

16

20

21

11

Directory

25

31

47

107

39

0

0

12

Company Practice

26

79

54

58

21

60

44

13

Miscellaneous

76

54

47

77

61

57

25

14

Baseline

0

0

0

0

0

24

0

15

Surcharges/Taxes

15

2

18

36

28

8

14

16

Number/Area Code

1

0

15

14

22

0

1

17

Rate Protest

1

0

2

3

2

0

0

18

Master/Sub Meters

0

0

0

0

0

0

0

19

Bill Format

5

1

3

2

1

0

0

20

Commission Policy/Practices

0

1

1

0

0

0

0

21

Operator Services

0

2

8

6

9

0

0

22

Annoyance Calls

6

5

10

6

14

0

0

23

Payment Arrangements

30

17

38

73

28

5

3

24

Commitment

0

1

9

16

12

2

1

25

Pay Per Call Service

16

19

15

13

5

0

0

26

Refusal to Serve

11

2

14

12

2

1

1

27

Estimated Billing

0

0

0

1

0

0

0

28

Deaf Program

0

2

0

0

2

1

0

29

Balance/Level Pay Plan

0

0

1

1

0

0

0

30

Illegal Activities

0

0

0

0

0

2

0

31

COPT

2

0

5

3

0

0

0

32

Custom Calling Features

21

93

45

42

44

21

0

33

Inside Wiring

13

1

12

13

16

6

3

34

Abusive Marketing

10

35

31

36

19

22

21

35

Backbilling

2

0

3

2

2

1

1

36

Centralized Credit Check System

50

28

43

24

20

1

0

37

Female/Minority Business Enterprise

0

0

0

0

0

0

0

38

Mergers

0

0

0

0

0

0

0

39

Low Income Programs

14

3

18

0

5

8

2

40

New Incentive Regulatory

265

1

1

4

3

0

0

41

Safety

0

0

0

1

1

0

1

42

Electromagnetic

0

0

0

0

0

0

0

43

Landline to Cellular

0

0

0

0

1

0

0

44

Improper Advertising

0

0

0

0

3

0

0

45

Cramming

0

0

0

16

10

6

7

46

Outages

0

0

0

0

0

3

9

47

Anonymous Call Rejection

0

0

0

0

0

0

1

48

Prepaid Phone Card

0

0

0

1

1

1

0

 

TOTALS

1,726

1,524

1,803

2,033

1,303

1,336

701

Verizon's totals compare to Pacific's as follows:

Verizon

1995

1996

1997

1998

1999

2000

2001

Totals

1,726

1,524

1,803

2,033

1,303

1,336

701

Pacific

1995

1996

1997

1998

1999

2000

2001

Totals

5,203

6,130

8,926

8,191

5,515

6,974

2,784

However, Pacific also has more than 4 times the number of access lines in California than does Verizon, according to each company's annual reports for 2001:

CALIFORNIA LEC YEAR-2001 NUMBER OF ACCESS LINES241

 

SWITCHED

NON-SWITCHED

TOTAL

COMPANY

ACCESS LINES

ACCESS LINES

ACCESS LINES

PACIFIC BELL

17,548,599

7,858,177

25,406,776

VERIZON CALIFORNIA, INC.

4,721,336

1,621,152

6,342,488

Moreover, if one organizes Verizon's data into the same categories as we did for Pacific - that is, those most directly related to service quality, Verizon's numbers are far lower proportionately than Pacific's:

 

Verizon

1995

1996

1997

1998

1999

2000

2001

Total

Abusive Marketing

10

35

31

36

19

22

21

174

Quality of Service

183

250

243

217

193

188

77

1351

Operator Services

0

2

8

6

9

0

0

25

Safety

0

0

0

1

1

0

1

3

Outages

0

0

0

0

0

3

9

12

Delayed Orders & Missed Appts

20

7

44

94

44

80

44

333

Missed Commitments

0

1

9

16

12

2

1

41

TOTAL

213

295

335

370

278

295

153

1939

Pacific's comparable numbers - with four times the access lines - are as follows:

 

Pacific

1995

1996

1997

1998

1999

2000

2001

Total

Total

1067

1784

3416

2686

1743

2165

587

13448

Even if one multiplies the Verizon figures by 4, Verizon's proportional numbers are far lower than Pacific's.

Overall, we do not find that Verizon shows significant service quality problems based on the informal complaint data before us.

ORA's customer service survey for Verizon showed that in the minds of the customers surveyed, Verizon's service quality has improved since 1991. ORA's own witness acknowledged this improvement and the ORA survey clearly bears out this conclusion. Indeed, the fact that the survey is capable of measuring an improvement in service quality is, in our view, a validation of the survey's usefulness for drawing a similar - but far less favorable - comparison between Pacific's past and present performance.

We agree with TURN, however, that even if customers perceive service quality to be good, there may nonetheless be objective service quality problems that require a remedy. It is for this reason that in reaching conclusions about Verizon's service quality in this decision, we look not only at customers' subjective perceptions but also at the objective reports Verizon makes to this Commission. Where Verizon demonstrates objective problems - for example in the areas of installation and repair data - we expect Verizon to make improvement.

Verizon claims it surveys its California customers by conducting over 1,000 interviews per month covering Directory Assistance, Consumer and Business Provisioning (which covers installation of new service), Consumer and Business Repair (which covers diagnosis, repair, and restoration of existing service), and Consumer and Business Request and Inquiry (which covers requests and inquires directed to the Business Office regarding customer bills, products and services, prices, and company policies).242 The results of these surveys show that Verizon offers good service quality. Neither ORA nor TURN challenged the results of these surveys.

Neither TURN nor ORA made specific allegations about Verizon related to the impact of technological change on its service quality. However, Verizon's claim that it has improved service quality through network upgrades tends to support TURN's general argument that technological advance may tend to create a system of "haves" and "have-nots."

Verizon concedes that technological advances improve quality: "Verizon has removed nearly all of the analog pair gain devices from its network to reduce interference with digital services. Verizon has also placed more than 380,000 miles of fiber optic cable to date in its network to improve data transport rates." SONET rings, Verizon claims, "[allow] customers to receive uninterrupted service even when there is a service outage in the main route." 243 To the extent these upgrades are not ubiquitous, but rather are focused on the most densely populated areas - where SONET rings are most prevalent, for example - some customers are not benefiting from the upgrades.

This claim is in contrast to Pacific's refutation of TURN's claim that upgrading technology creates better service quality. While Verizon does not agree with TURN that the upgrades risk creating two classes of customers - those with and those without the latest technology - its assertion that SONET rings and fiber optics improve service tends to support TURN's arguments.

The underlying question is whether NRF can or should be modified to prevent this disparity in service. We certainly do not wish to discourage technological upgrades. The parties should be prepared to address any regulatory changes they believe we can reasonably make in Phase 3B of this proceeding.

TURN made the same arguments with regard to Verizon as it did regarding Pacific. It states - and we agree - that a carrier should not benefit from relaxed service quality expectations because it experiences a period of great growth in demand, access lines, customers, or company size. We agree with TURN that "[i]t is the duty of the carriers to manage their growth so as to maintain high service quality."244

Verizon does not disagree with TURN in this regard. Rather, it simply reports increases in demand for its services and explains the investments it made to meet this growth. Verizon actually cites instances in which it improved its service quality during times of exceptional growth: "For example, Verizon's continuing investment in the network significantly reduced the number of delayed order requests (DORs) experienced by its customers. . . . Verizon had 5,658 DORs in 1994 and had only 814 in 2001."245

However, we do not necessarily agree with Verizon's conclusion that "NRF provides the proper incentives . . . to maintain a high level of service quality. . . ." While Verizon may have used its increased flexibility under NRF to improve its service quality, Pacific's less positive results under NRF undermine any general argument that NRF alone, and not other factors, causes companies to improve service quality.

TURN shows that Verizon's field staff has declined over the period 1989-1994. TURN states that Verizon's field staffing declined by 35% from 1989, the year the Commission implemented NRF, to 2000, with a large reduction (42%) occurring from 1989-1994. While Verizon's reply testimony suggests that a smaller decline occurred, TURN claims it is misleading because it uses 1984 as the base year for comparison rather than 1989, when NRF was instituted. TURN shows the following staff levels over the period 1989-2000:

TURN also points out that Verizon "was not able to provide data on the levels of experience of its field personnel," so that it was impossible to go beyond the raw numbers to determine the quality of the field work being done by Verizon's employees.

Verizon acknowledges reductions in force, but claims that because Verizon redefined certain field positions, the raw numbers TURN used above and elsewhere in its testimony are misleading. It states that the total reduction in cable splicers or their equivalents was 17%, not the much higher percentage TURN claimed.

Nonetheless, these are deep cuts and could portend service quality problems down the road. The cuts may also explain some of Verizon's problems with respect to installation and repairs/trouble reports.

Verizon acknowledges that service quality suffered during the first quarter of 2001 due to unusually heavy rains: "[T]he [repair] intervals were extraordinarily high during the January, February, and March period, due to some prolonged rains that we experienced at that particular point in time."246 We discuss the specifics in the Section entitled "Repair - Verizon," above. However, Verizon's witness admitted that poor weather is no excuse for poor service quality, and that a company with significant outside plant, such as Verizon, should be prepared for inclement weather:


    Q. . . . [A]s a company with huge - hundreds, thousands, maybe millions of miles of outside plant, it wouldn't be responsible simply to sit back and hope that there isn't excess rain in a given year, would it?

A. It's - it's not how we would prepare for that actuality.

Q. Because it wouldn't be responsible to prepare that way, would it?

A. No, it wouldn't.247

We agree with Verizon, and find that inclement weather is an unacceptable excuse for the poor service quality it delivered during that period. While we appreciate the steps Verizon has taken since that time to improve its response to weather-related outages,248 they do not obscure the fact that Verizon delivered poor service quality during the first quarter of 2001.

In claiming Verizon has engaged in marketing abuse, TURN again cites C.98-04-004/D.98-12-084, in which the Commission approved GTEC's payment of $13 million to settle marketing abuse claims stemming from the period 1989-92.249 TURN also claims that Verizon has "misused customer contacts as marketing devices."250 However, this allegation is based on the same case.

We do not agree with Verizon that this case is irrelevant to our assessment of its service quality because similar incidents have not occurred in the years since then. By the same token, in discussing this case earlier in this decision, we did not believe the case indicated a pattern of problems. See Section entitled "Formal Complaints - Verizon," above.

No party alleges that Verizon's mergers and structural changes have had an impact on service quality. Nor does Verizon - in contrast to Pacific - argue that changes in the company attributable to its growth in size are mitigating factors that explain its service quality results. Thus, we do not find that in Verizon's case mergers or structural changes have had an impact on its service quality.

192 ORA Opening/Service Quality at 3. 193 TURN Reply/Service Quality at 7. 194 19 RT 2306:4-9 (Schilberg) (emphasis added). 195 Resolution T-15404 and D.98-12-084. 196 D.98-12-084, 1998 Cal. PUC LEXIS 910, at *13. 197 1998 Cal. PUC LEXIS 910, at *18. 198 Verizon Opening/Service Quality at 4. 199 Verizon Opening/Service Quality at 4. 200 TURN Reply/Service Quality at 6. 201 TURN Opening/Service Quality at 7. 202 18 RT 2263-67. Any reference to Phase 3B in this decision should be interpreted to include a separate phase if the Commission further segments this proceeding in the future. 203 A.01-11-014. The Commission has not yet acted on this application, in part due to uncertainty about whether the Commission should decide competitive issues Verizon's competitors raise with regard to DSL services in A.01-11-014 or in another more comprehensive proceeding regarding the incumbent local exchange carriers' obligations to share DSL lines with competitive carriers (R.93-04-003 et al.). 204 POTS is an acronym for "Plain Old Telephone Service." 205 20 RT 2481:16-22 (Anders). 206 20 RT 2483:3-14 (Anders). 207 TURN Reply/Service Quality at 8. 208 Exh. 2B:216 at 13:8-17 (Thoms Supplemental Reply Testimony). 209 Exh. 2B:133 at 21 (Hieta Direct Testimony). 210 Verizon Reply/Service Quality at 31, citing Exh. 2B:214 at 24-25/charts (Thoms Direct Testimony). 211 Verizon Reply/Service Quality at 9. 212 The installation interval and commitments met graphs are based on ARMIS data reported by carriers to the FCC. Prior to 1996, carriers reported ARMIS data on a quarterly basis, and thereafter, annually. For years reporting quarterly data, quarterly installation orders are summed to obtain annual installation orders. Annual installation intervals and commitments met are obtained by weighting and combining the quarterly data (i.e., multiplying quarterly installation intervals or commitments met by quarterly installation orders, summing the results and dividing the summed result by annual installation orders). Similarly, Verizon's annual installation orders and commitments met are obtained by summing GTE California (GTEC) and Contel installation orders. Verizon's installation intervals and commitments met are obtained by weighting and combining the GTEC and Contel installation intervals or commitments met. 213 The ARMIS data on which the graphs in text are based separates Verizon into three companies: GTE/California, Contel/California and West Coast/California. Because GTE California has approximately 80 percent of the access lines of the three entities, we compare the ARMIS figure for GTE California to Pacific's performance. 214 Compare Section above entitled "Self-Monitoring - Pacific," in which we quote from the testimony of Pacific's witness Mr. Resnick with 20 RT 2477-80 (Verizon Thoms/Anders/Fernandez panel). 215 20 RT 2477:18-28 (Anders). 216 20 RT 2479:2-19 (Anders). 217 Exh. 2B:133 at 21 (Hieta Opening Testimony). 218 Exh. 2B:507 at 19:8-13 (Schilberg Opening Testimony). 219 Verizon Reply/Service Quality at 12. 220 "ARMIS average annual repair intervals for business initial and repeat out-of-service repairs and initial and repeat all-other-trouble repairs varied temporarily in 2001 from the levels previously achieved by Verizon. Comparable residential intervals also varied in 2000-2001 from levels previously achieved." Verizon Opening/Service Quality at 17-18. 221 20 RT 2490:11-19 (Anders). 222 TURN Reply/Service Quality at 12. 223 Id. 224 Verizon includes billing related calls in its reported BOAT data, while Pacific began excluding billing related calls from its BOAT reports beginning May 1999. Pacific's adjusted BOAT data is from TURN's August 27, 2002 Supplemental Testimony. 225 Because monthly calls volumes are not part of the record, annual averages were estimated by summing the monthly results and dividing by 12. 226 D.94-06-011, 55 CPUC 2d 1, 53 (1994), 1994 Cal. PUC LEXIS 456, at *150 et seq. 227 19 RT 2318:28-2319:20. 228 Exh. 2B:138 at 6 (Piiru Direct Testimony). 229 GO 133-B, Sections 3.8 and 3.9, respectively. 230 Id., Section 1.3(b). 231 22 RT 2786:10-17 (statement by Verizon's counsel). 232 Verizon Reply/Service Quality at 23. 233 Verizon's SPG was originally a provision of Contel of California's (Contel's) tariffs prior to the GTEC/Contel Merger. ORA's predecessor argued during the merger proceeding that Contel's SPG was superior to Verizon's, and adoption of Contel's SPG by Verizon should be a condition of the merger. However, before the Commission ruled on the issue, Verizon voluntarily adopted the SPG contained in Rule Nos. 18 and 19 of its tariffs. Advice Letter No. 5521, filed August 30, 1993. 234 20 RT 2493:26-2493:19 & 2495:4-13 (Anders). 235 20 RT 2493:20-21. See also id. at lines 17-19 (Q. "Does a customer get . . . a credit without ever having called Verizon to complaint? A. No, they shouldn't be . . . ."). 236 Verizon's Tariff Rules No. 18(4) and 19(4) state, "A credit will be extended in accordance with the above conditions at the request of the customer." 237 1994 Cal. PUC LEXIS 456. 238 1994 Cal. PUC LEXIS 456, at *154. 239 We discuss this case in full in the Section entitled "NRF Incentives and Service Quality - Verizon - Introduction," above. 240 D.98-12-084, 1998 Cal. PUC LEXIS 910, at *13. 241 Source: Pacific and Verizon ARMIS 43-08 reports, Table III, for 2001, available at http://gullfoss2.fcc.gov/cgi-bin/websql/prod/ccb/armis1/forms/43-08/frame3.hts. 242 Verizon Opening/Service Quality at 51-52. 243 Verizon Opening/Service Quality at 54. 244 TURN Opening/Service Quality at 36. 245 Verizon Opening/Service Quality at 55. 246 20 RT 2488:5-7 (Anders). 247 20 RT 2490:11-19 (Anders). 248 Such improvements, according to Verizon, included reassigning technicians from non-demand activities such as preventative maintenance to service restoration activities; augmenting its construction workforce with contractors to free up additional employees to work on repair activity; using extensive overtime to increase man-hours committed to service restoration; making significant capital investment to help insulate Verizon's plant from rain related trouble; and developing procedures to more effectively deploy technicians during periods of extreme demand. Exh. 2B:219 at 8:16-17:5 (Anders Reply Testimony). 249 We discuss this case in full in the Section entitled "NRF Incentives and Service Quality - Verizon - Introduction," above. 250 TURN Opening/Service Quality at 42.

Previous PageTop Of PageNext PageGo To First Page