Loading...
HomeMy WebLinkAboutCALLPLAN.docxMemorandum To:Commissioner Nelson Commissioner Smith Commissioner Hansen Myrna Walters Tonya Clark Gary Richardson Stephanie Miller Dave Schunke Eileen Benner Beverly Barker Don Howell Weldon Stutzman From:Lynn Anderson Subject:Staff Review of GTE's LCP in Kootenai and Bonner Counties Date:July 18, 1995 Attached is a copy of Staff's review of GTE's Local Calling Plan (LCP) as implemented on a trial basis in Kootenai and Bonner Counties.  GTE submitted its Report on April 3, 1995, as required by Order No. 24877 in Case Nos. GTE-T-90-1 and GTE-T-92-2,-3,-4.  (On July 10, 1995, GTE submitted its formal application to expand LCP to its remaining exchanges in Case No. GTE-T-95-2.  The Staff review does not include analysis of the proposed expansion.) Following is the conclusion from the Staff review: GTE's Local Calling Plan helps satisfy the local calling needs of customers in exchanges previously without adequate local calling areas.  This satisfaction comes primarily from the expansion of local calling areas, rather than the expansion of rate options from two to four.  In fact, the number of rate options contributes to the confusion of many customers, which leads to dissatisfaction and poor choices.  Obviously not all customers believe LCP is an improvement, some even think its worse than what they had before.  But with Staff's seven suggested modifications to rates, billing practices and route selections, overall customer satisfaction should be increased without diminishing GTE's rate-of-return.   cc:  Working File, GTE-T-90-1, et al GTE's Local Calling Plan in Kootenai and Bonner Counties Idaho Public Utilities Commission Staff Review July 18, 1995 Introduction On April 3, 1995, GTE Northwest, Inc. submitted its Report on the Implementation of Its Local Calling Area Plan (LCP) as required by Order No. 24877 dated May 20, 1993, in Case Nos. GTE-T-90-1, GTE-T-92-2, GTE-T-92-3 and GTE-T-92-4. GTE's Report concludes that LCP satisfies most customers in eleven exchanges located primarily in Kootenai and Bonner Counties and has not adversely affected GTE's revenues.   Background:  Summary of Events/Petitions Leading to LCP In May of 1989 the Commission issued Order No. 22464, Case No. U-1002-67A, which redesigned rates following a general rate increase.  Various types of Preferred Area Calling (PAC) were among the many services for which rates were restructured and increased.  (PACs were optional toll discount plans for calls of  distances limited to 32 or 40 miles and were sometimes used as an alternative to toll-free extended area service or EAS.)  The restructured discounts, called TollPAC, provided new flexibility by allowing each customer to choose up to three routes of up to 50 miles for inclusion in the 30% discount rate.  Unfortunately the new TollPAC rates represented more than a doubling of average PAC rates on existing routes, with some increasing as much as 200 percent.  Many of GTE's customers subsequently complained to the Commission through telephone calls, letters and petitions.  Especially vocal were those in small exchanges lacking EAS, such as Spirit Lake, Bayview, Hope and Clark Fork. Also in Order No. 22464 was the requirement for GTE to implement EAS among  the Rathdrum, Post Falls and Hayden Lake exchanges.  (At that time the Rathdrum-Coeur d'Alene EAS was a fairly recent addition.)  In its implementation of these three EAS routes the Company inadvertently implemented EAS between Spirit Lake and Rathdrum, Post Falls, Coeur d'Alene and Hayden Lake, which continued for several months before the error was discovered and toll charges reinstated.  This mistake and subsequent correction contributed to customer dissatisfaction in the Spirit Lake exchange.     By November of 1989 the Commission opened Case No. GTE-T-89-7 to investigate EAS on several routes and to reexamine TollPAC.  GTE's response in 1990 to these complaints was its MARC I and II toll discount plans (subsequently renamed GTE Discount Plans I and II).   Unlike TollPAC, the rates for Plans I and II included monthly charges rather than minimum amounts and the usage discounts were applicable to all toll routes within certain distance limits rather than just a customer's choice of three.  Customers had to choose between Plan I's 50% discount for calls up to 23 miles and Plan II's 35% discount for calls up to 40 miles.  Beginning August 1, 1990, the Plans were available to customers in most GTE exchanges.  After a trial period it was determined that revenues exceeded GTE's expectations and rates were reduced to current levels effective September 1, 1992.     Meanwhile complaints and EAS petitions continued to be filed with the Commission.  In total, following implementation of TollPAC and up until LCP was implemented, the Commission received about a dozen formal petitions for relief from toll charges on twelve routes in Kootenai and Bonner Counties. Summary of LCP In response to the many EAS petitions filed in northern Idaho, GTE implemented on a trial basis 22 new EAS routes in December of 1993 and following months.  As part of the package, customers in the eleven affected exchanges in Kootenai and Bonner Counties (nearly two-thirds of all GTE customers) were given four rate options instead of  just the traditional two options of flat-rate or local measured service.   The four rate options and the old and new EAS routes were named by GTE as its Local Calling Plan (LCP).  The specifics of LCP as contained in Order No. 24877 were slightly modified from those proposed by the GTE. Under LCP each exchange's local calling area was expanded to include all other exchanges within 23 miles (as measured using standard toll distance calculations)  and customers had to choose one of the following rate options: 1-- Basic Calling -- All calls within the 23-mile radius including calls made in the customer's immediate exchange are billed based on measured usage rates in addition to a monthly charge.  The measured rates were patterned after GTE's previous local measured service rates and range from less than one-third to about one-half the toll rates for the same distances. 2-- Community Calling -- Customers are billed a higher monthly charge than for Basic but calls within the immediate exchange are free.  The monthly charge varies considerably and increases as the number of lines in the immediate exchange increases.  Calls within 23 miles but outside the immediate exchange are billed local measured usage rates. 3-- Community Plus -- For a flat monthly rate, this option provides free calling within a customer's home exchange and to specified nearby exchanges that were previously toll but thought to have high need for EAS, as well as to pre-existing EAS exchanges.  The monthly charges have less variation between exchanges than in the Community Calling option because the number of lines in the various flat-rates area have much less variation.  Calls within 23 miles but outside the identified Community Plus area are billed local measured usage rates. 4-- Premium Calling -- For a monthly charge that is twice that of the Community Plus option customers may make an unlimited number of calls to all exchanges within the 23-mile radius local calling area. Attachment 1 shows most local and toll rates before and after LCP implementation in each exchange.  Attachment 2 lists local calling areas of each exchange before and after LCP and the percent of customers subscribing to each rate option. Summary of GTE's Analysis and Expansion Proposal GTE's Report says that LCP in the eleven exchanges is a success, that most customers are satisfied by it, and that Company revenues are in tact.  Thus, GTE says LCP should be expanded to all intraLATA, intrastate GTE toll routes of 23 or fewer miles.  On July 10, 1995, GTE filed an application outlining the details of the proposed expansion, which has initiated Case No. GTE-T-95-2.  Analysis of the proposed expansion is not part of this review. Staff Analysis Customer Choices and Migration to Measured Service In varying formats, GTE's Exhibit A and Staff's Attachments 2 and 5 show by exchange the percent of customers choosing each LCP rate option. According to information supplied in Exhibit A of GTE's Report, by December of 1994, 83% of the customers subscribed to the Community Plus option, which was the default option of most customers who failed to make a choice.  This percentage has been on a slight decline since January of the same year.  Fewer than 1% of the customers had the Premium option by December, which has also been declining in popularity.  However, these percentages vary widely and sometimes unexpectedly by exchange.  For example, it is not surprising that in the Bayview exchange, where calls to Coeur d'Alene are flat-rated only with the Premium option, nearly 15% of customers chose that option.  But it is surprising that in the similarly situated Spirit Lake exchange only 1.3% chose that option. Just prior to LCP implementation 1.5% of GTE's lines in the 11 exchanges had local measured service. (For comparison, U S WEST's LMS penetration was about 3.5% after having offered optional LMS for many more years than had GTE.)   By December 1994, GTE's local measured service penetration had increased more than five-fold, to 7.9%, not counting the measured service associated with the 9.3% of customers who chose the Community option.  In some exchanges the Community option is more like LMS than not because most local calls made are measured rather than flat-rated.  Both Basic and Community LCP subscriberships have increased since the beginning of 1994.     Potential Policy Problems Associated with Migration to Measured Service As shown in the preceding section, LCP has resulted in large increases in the number of customers with local measured service (LMS).  One problem, at least in theory, is that as more customers are attracted to measured service's lower monthly rates, telephone companies want to recover the resulting revenue loss by increasing rates for non-measured services, which then encourages even more customers to choose measured service, with this cycle continuing until virtually all customers are eventually forced onto local measured service.  The Commission apparently recognized this problem in 1985 in its generic consideration of the potential social and economic costs and benefits of local measured service.  In Order No. 19717, Case No. U-1000-73, the Commission stated that it would not allow telephone companies to encourage rapid migration to LMS by pricing it too attractively. (p.10)  It is worth noting that GTE and Contel were the only two parties in that case who openly advocated mandatory LMS.   Local measured service is sometimes claimed by the industry to be the most efficient way to price local exchange service -- the argument being that if customers pay for the calls they make then telephone facilities will be used more efficiently.  The main problem with this argument is that telephone calls are a steeply declining cost commodity, i.e. once facilities are in place to accommodate a given number of calls, the cost of allowing additional calls is much less than the first.  The authors of a study published by the Rand Corporation in 1987 concluded that "...contrary to conventional wisdom, measured service pricing of telephone calls is likely to be somewhat less efficient than traditional flat rate pricing." (footnote: 1)  This finding was reaffirmed in 1993 by a study published in the Rand Journal of Economics, wherein the authors concluded that even on routes made local by virtue of two-way EAS, the aggregate public and private benefits associated with flat rate pricing can be expected to exceed those associated with one-way  EAS or of marginal cost pricing.(footnote: 2)  (Note:  LCP has similarities to  one-way EAS in that calls that are flat-rated in one direction may be measured in the other.) Local measured service is also often marketed to regulators as a way to encourage telephone subscribership to low-income households, but measured service customers are often, if not usually, double-income-no-kids householders who have access to office telephones for making most of their personal calls.  It is likely that the low-income householders who do subscribe to measured service in order to save a few dollars are foregoing much local communication and doing so completely needlessly when viewed from an economic efficiency perspective. The problem of LCP resulting in customer migration to LMS is not insurmountable.  It can be mitigated or eliminated through rate changes.   GTE's Customer Survey GTE's report concluded that,  "An overwhelming majority of our business and residential customers expressed satisfaction in the LCP as opposed to the service previously held.  The LCP provided a range of options for the customers to enhance their calling needs, at reasonable prices."  (Report, p. 4)    However, GTE's conclusion appears overstated since only 40% of the survey respondents said LCP was better than what they had before and 35% said the price was too high.  And while 22% said they liked having choices, 23% said the options were too confusing.  Overall, customers seem to like the expanded local calling area aspect of LCP much more than having several rate options. GTE said its survey included 20,000 "randomly selected" customers, but the four-question survey was sent to 36% of residence customers and only 20% of business customers.  It would have been statistically more valid to survey a higher percentage of businesses than residences because the business customer population is much smaller than the residence customer population.  In other words, when analysis of subgroups within populations is desired a stratified sample is appropriate and smaller subgroups require larger sample size percentages than do larger subgroups for equally reliable statistical extrapolation to their population universes. As it turned out, the business class was underrepresented not only in the survey design but also in the response rate, which was 8% compared to 12% for residences. The total response rate of 11.5% is far below what is generally considered necessary for reliable statistical extrapolation to the population universe -- a 50% response rate may be considered adequate and 70% very good.  However, a low response rate does not necessarily indicate an inadequate sample size, rather the real problem is the likelihood that there are significant differences between those who respond and those who do not.  One way to assess whether a low response rate has biased the sample is to do a follow-up telephone sample survey of nonrespondents and compare the results to that of the first survey.  GTE apparently did not have such a follow-up survey completed.       GTE analyzed the survey responses by residence and business and by LCP choice, but not by exchange.  This was problematic because LCP was designed primarily to respond to the needs of customers in exchanges with very small local calling areas.  Thus, it would have been appropriate to stratify the survey by exchange subgroups with smaller subgroups having larger sample size percentages, as indicated above.  The staff has since obtained the survey results by exchange (but there is no indication that survey size percentages differed between large and small exchanges).  Attachment 3 shows, not surprisingly, that customers in exchanges with smaller local calling areas prior to LCP were nearly twice as likely to respond to the survey as those in larger local areas pre-LCP.  Some other notable differences between these two groups are that 48% of respondents in the smaller areas said they liked the new local calling areas compared to only 32% of those in the larger areas.  In the smaller areas 27% said they disliked their new calling area compared to 17% in the larger areas.(footnote: 3)   Of respondents in the smaller areas, more in the Bayview and Spirit Lake exchanges disliked LCP calling areas (36% and 39%, respectively) than in any other exchange (which ranged from 10% to 23%). In generalrespondents from smaller areas were much more likely to say that LCP was an improvement (59%) compared to those from larger areas (37%).  Whether respondents were from small or large areas, nearly 20% said LCP was worse than what they had before.  Not surprisingly,  respondents from the larger areas were much more likely to be indifferent as to whether LCP was better or worse (44% said it was the same or did not answer) compared to those in the smaller areas (24%.) Comparing survey responses by LCP option chosen to actual penetration data shows that those who said they had the Basic option overrepresented their subgroup population by over 4.5 times, while those who said they had the Community Plus option underrepresented their subgroup by less than one-half. (See Attachment 4)  One out of every 12 respondents did not identify their LCP option, presumably because they did not know what it was.  This last group was generally more negative in their LCP questionnaire responses than were others.   Following are some overall observations of positive correlation: --Smaller flat rate calling areas and liking LCP choices; --Smaller flat rate calling areas and liking LCP prices; --Smaller flat rate calling areas and disliking LCP calling areas; --Larger flat rate calling areas and disliking LCP prices; --Smaller flat rate calling areas and saying that options are too confusing; --Smaller flat rate calling areas and choosing more dislikes than likes; --Larger flat rate calling areas and rating LCP as better than before; --Smaller flat rate calling areas and rating LCP as worse than before; --Smaller flat rate calling areas and rating LCP as same or not answering. In general, it appears that respondents who chose (or who were assigned by default) Community Plus and Premium options are more satisfied with LCP than those who have Basic and Community options.  Those with Basic and Community expressed more satisfaction with prices and having choices, but then expressed more confusion and dissatisfaction with their flat rate calling areas. It is tempting to further analyze responses by exchange and LCP option, but in most cases the sample sizes are very small at this level of disaggregation and the overall problem of a likely biased sample becomes more acute.   Similarly, the small and likely biased business sample discourages much analysis of business versus residence customer satisfaction.  Suffice it to say that the business responses generally are only slightly less positive (or slightly more negative) than are the residence responses. In summary, it appears by the very low survey response rate (11.8%) that GTE's customers are largely indifferent to LCP.  Nearly a quarter of those who did respond said the LCP options were too confusing and it could be hypothesized that an even greater percentage of the apparently indifferent nonrespondents were too confused by LCP to respond.  But customers in exchanges with inadequate local calling areas prior to LCP's automatic 23-mile EAS expansion appear to be much more interested in LCP and more satisfied with the results than customers who enjoyed relatively large local calling areas before LCP.  Because of problems associated with the survey design and implementation, few conclusive statements can be made about whether the LCP rate options satisfy customers more than pre-LCP rates for a given local calling area.  However, the primary selling point of LCP, i.e. customer choice of options, does not appear to contribute heavily to customer satisfaction--only a fifth of the respondents said they liked having choices, far fewer than those who said they liked having a larger local calling area and even fewer than those who said the options were too confusing.       Customer Complaints to the Commission Since LCP implementation the Commission has received 18 complaints specifically regarding that service.  One-third of the complaints were regarding GTE billing errors for measured calls, the billing format and charges associated with measured call bill detail, and/or confusion about LCP by customer service representatives.  Another one-third were complaints about the inadequacy or inequity of LCP routes or rates, including the requirement that foreign exchange lines be restricted to the Premium rate.  Finally, one-third of the complaints were from customers wanting LCP where it is not available, specifically in Bonners Ferry, Priest Lake, Harrison, Setters, and Mullan. Communication Effects Had the data been available, it would have been informative to compare total communication (calls and/or minutes) occurring before and after LCP implementation in each exchange and on specific routes.  GTE has provided both flat rated and measured usage by exchange for December, 1994, but not before LCP.  Route specific LCP data was also not available, which would have allowed comparisons to be made to toll usage before LCP. Attachment 5 contains a summary of originating minutes of use per line by exchange, by residence and business, and by LCP option, calculated from data provided by GTE.  (Attachment 7 contains this data sorted differently.)   In general, average originating minutes per line increase as measured service areas decrease or as the flat rate areas increase.  The few exceptions to this rule may be due to either or all of data errors, poor customer choices, and usage anomalies of a few customers.  Somewhat surprisingly, total business lines were reported to have had originating minutes of use about 5% below that of  residence lines.  It is expected that businesses typically originate more calls than do residences, but if the average duration of business calls is much less than average residence call duration, then total minutes of residences and businesses will be more equal than total calls.  The total business/residence usage ratios vary widely among exchanges, e.g. Spirit Lake and Hope at 1.50 and 1.31, respectively, compared to Clark Fork and Hayden Lake at 0.83 and 0.82, respectively.  Within a particular LCP rate option the highest ratio of business usage compared to residence is 8.84 by customers in Coeur d'Alene with the Premium option.  Surprisingly, the ratios in Hope for both Basic and Community customers are 2.82 and higher, while it is 0.20 for Premium customers in Post Falls.  It appears quite possible that GTE's data is inaccurate for one or more exchanges and customer classifications.  And it should be cautioned that this data, even if accurate, is insufficient to assess whether business usage demand during peak hours is diminishing relative to residence demand.  To the extent there are traffic sensitive costs for telephone calls, they are incurred primarily during peak busy hours, which typically occur during the business day and are the result of business activity.   A comparison of residence originating minutes and the number of lines in each flat rate calling area (sorted by the latter) is shown in Attachment 6 both in table and graph formats.  As expected, average usage generally increases as lines in the flat rate calling area increases.  This holds true whether counting only flat rate calls or the total of flat rate and measured rate calls, with positive correlations of 0.90 in either case.   Attachment 7 contains a table with revenue per line and minutes per line for residences and businesses by rate option and exchange.  In this table the data was sorted first by rate option and then by residence average bill.  The format of this table allows easy comparisons between residence and business customers as well as between customers with the same rate option in different exchanges and between rate options.  To the right of the table are comments questioning the wisdom of customer choices and/or the accuracy of the data indicated by shaded areas in the table.  Usage reported for most Spirit Lake customers seems dubious.  The Sandpoint usage numbers are interesting, if not dubious, because there is very little difference between customers with Community versus those with Community Plus and, surprisingly, business customers with Community are shown as having more minutes than those with Community Plus. It should be pointed out that the usage data that is available is limited to originating calls only.  On the whole this limitation is inconsequential because each originating minute has a terminating minute, but when analyzing the effects on specific customer groups it could be useful to assess the effects on the number and duration of calls received as well as those made.  It may be that some types of customers' increases in calls made associated with expanded local areas are much different than increases in the number of calls they receive.   Customer Bills by Exchange and Originating Minutes of Use Attachment 8 contains a scatter graph of each exchange's average residence LCP bill versus its average minutes of originating usage.  On average, customers in the exchanges below the diagonal line (representing total average LCP revenue per minute) pay more per originating minute of use than those in exchanges above that line.  It is noteworthy that the exchanges below the line are all smaller than the smallest exchange above the line. Revenue and Cost Effects The capital investment necessary to accommodate the additional LCP traffic was reduced by GTE's innovative integration of toll and local traffic on the same facilities, thereby gaining per trunk capacity increases inherent with larger trunk group sizes.  GTE said it invested about $470,000 (about $7 per LCP line) to accommodate the traffic increases associated with the 22 new EAS routes in the LCP trial.  GTE says it needs to invest about $150,000 (less than $5 per line) to increase facility capacities necessary to expand LCP to the rest of its customers, which would convert toll traffic to local on 20 additional routes.  In total, GTE says that its conversion of 42 toll routes to local will have required an average capital investment of less than $15,000 per route or about $6 per line.  The Staff does not necessarily accept GTE's assumptions in its calculations of monthly carrying costs, but concludes that the reported costs are small enough to be relatively insignificant for the purpose of evaluating LCP effects. Attachment 9, p. 1, contains a table showing revenue per line by exchange before and after LCP implementation based on information supplied by GTE.  Total revenue increased by 9.6% ($1.44 million annually), but revenue per line decreased by 2.3%, mainly due to nearly a four-fold increase in the number of CentraNet lines, for which GTE reported only a portion of associated access rates.  The omitted revenue is from the rates for central office trunks, the number of which are unknown to Staff and which vary among customer systems.  Staff has estimated the revenue effect of LCP alone by hypothetically holding CentraNET line growth to the same percent as overall line growth.  This hypothetical estimate produces a  revenue increase of about $23,500 per month.  Subtracting GTE's claim of $15,100 in monthly costs associated with its LCP capital investments, results in a net revenue increase of $8,400 per month or about $100,000 annually.  Attachment 9, p. 2, compares GTE's inaccurate revenue analysis presented on page 1 of Exhibit C of its Report to the Staff's hypothetical analysis described above. Perhaps a more important shortcoming of either the Company's or Staff's revenue per line analysis is the implied assumption that all revenue associated with line growth should be retained by the Company in order to maintain its rate-of-return.  There are two flaws with this assumption.  First, the rate-of-return GTE was earning prior to LCP may have been above or below its Commission-authorized rate-of-return.  Second, the per line method of comparing revenues in order to maintain a rate-of-return is valid only if the cost of providing new lines is exactly the same as the average cost of old lines--this is generally a false assumption, not only because of general economies of scale, but also because technology improvements and in-filling of geographic areas should be reducing the costs of additional loops and associated switching and interoffice transmission.  In fact, GTE's LCP line growth rate of over 12% is about double the estimated population growth rate of the two counties, which indicates that much of the line growth rate is due to existing customers requesting additional lines.  Such lines generally require less outside plant investment than average and have less associated customer service costs. In conclusion, GTE's revenue analysis is insufficient to determine whether LCP has resulted in any change to its rate-of-return, although it appears certain that LCP has not caused a reduction and any increase is not likely to have been very large.  However, whether or not there has been a change in GTE's rate-of-return does not address whether the prior rate-of-return was within a reasonable range of that authorized by the Commission or whether the rate-of-return last authorized in 1988 is reasonable in today's environment. Analysis of LCP "fit" to Needs of Present Exchanges LCP's conversion to local calling on all routes less than or equal to 23 miles was a reasonably good response to the needs of customers in those 11 exchanges.  However, this distance limitation excluded at least two routes that exhibited some need for local calling--Spirit Lake-Sandpoint, which are bordering exchanges but have a 26-mile toll distance, and Bayview-Post Falls at 25 miles.  It also needlessly included two routes--Bayview-Hope and Bayview-Clark Fork, both at 23 miles but with Lake Pend Orielle as a natural barrier.  By moving the Bayview toll coordinates from the community of Bayview, which is located on the eastern extreme of the exchange, to a more central location, perhaps the Athol airport, the two unnecessary routes would be eliminated and the Bayview-Post Falls route would be added.  Depending on exactly where Bayview's coordinates were relocated, Bayview-Oldtown might also be added to LCP, which would not be entirely unwarranted.  No other LCP routes  would be affected by moving Bayview's coordinates, except perhaps by a local usage mileage band change. Staff Recommendations for Changes to LCP Routes, Rates, and Service 1)LCP's uniform 23-mile distance is unnecessarily and unreasonably restrictive.  Of the current exchanges with LCP, Spirit Lake-Sandpoint and Bayview-Post Falls are probably the worst LCP route omissions. The Bayview-Post Falls omission can be easily corrected by moving the toll rating coordinates as previously described or by waiving the arbitrary 23-mile limit.  The latter solution could also correct the Spirit Lake-Sandpoint route omission.  (The unworkability of the 23-mile standard will become more apparent as GTE expands LCP to the remainder of its service territory.)   2)Staff recommends that the charge for billing detail be eliminated.  The tariff charges for local calling detail, whether for LCP or non-LCP measured service, are $1.50 per month plus $.10 per page.  Measured or "metered" service by telephone companies, without a requirement to show the meter, requires unwarranted faith in telephone companies' billing systems.  Electric, gas and metered water utilities all have meters the customers can monitor.  While there may be a small cost to provide call detail to all customers, such detail is necessary for customers to assess the accuracy of their bills and to determine whether they have chosen the correct LCP rate option.  In addition, the cost of providing detail will be at least partially offset by the reduction in customer time and GTE resources now used to negotiate billing disputes. 3)The flat rates for LCP's Basic option should be increased to temper the migration to local measured service.  The rate increase should be moderate, perhaps $1.00 per month for residences and proportionately more for business customer classifications.  Such a rate increase should more than offset any cost increases associated with the universal provision of measured service bill detail recommended above. 4)The flat rates for Rathdrum's Community options should be reduced.  The rates were originally designed with a balance between flat rate calling area size and total LCP area size, but Rathdrum's rates now appear out of balance.  The residence rate of $14.00 is the same as in Coeur d'Alene, Hayden Lake and Post Falls even though its flat rate calling area (i.e. within Rathdrum) is much smaller.  Sandpoint's $12.75 rate seems a convenient mid-point between the current $14.00 rate and the $10.25 rate in the six smaller exchanges.  The business rates for the Rathdrum Community option should be reduced proportionately.  The revenue effect of this rate change for the 29 Rathdrum Community customers is negligible.   5)The flat rates for the Community Plus option in the Bayview and Spirit Lake exchanges should be reduced from the levels they share with Coeur d'Alene, Hayden Lake and Post Falls ($15.95, res.) to match those in Sandpoint, Hope, Clark Fork, Oldtown and Priest River ($15.00).  The flat rate calling area sizes of Community Plus in Bayview and Spirit Lake are more comparable to those in these smaller exchanges than to those in the larger exchanges.  This change will mitigate some of the much higher than average dollar-per-minute bills for Bayview and Spirit Lake customers as shown on Attachment 8.  The revenue effect of this rate change is relatively small (under $30,000 annually), however the rate change may not be warranted if the LCP route omissions previously discussed (S.L.-Sandpoint and Bayview-P.F.) are remedied. 6)The rates for the Premium option in all exchanges should be reduced.  Subscribership to this option is consistently in the 0% to 1% range in all exchanges for both businesses and residences, except for the 17% of residences in Bayview.  It appears to be price-prohibitive for most customers in the few exchanges where it offers value.  The revenue effect of a moderate rate reduction for the 473 residence and 59 business customers is relatively small and will likely be offset by increased subscribership.  An alternative to reducing Premium rates is to eliminate it as an LCP option.  However, this alternative would require that the Community Plus routes be expanded to include both Bayview and Spirit Lake with Coeur d'Alene and perhaps a few other routes of lesser importance. 7)The restriction of foreign exchange (FX) customers to Premium rates should be eliminated unless GTE can show why it is not in the public interest to allow FX customers the same choices as all other customers.   Conclusion GTE's Local Calling Plan helps satisfy the local calling needs of customers in exchanges previously without adequate local calling areas.  This satisfaction comes primarily from the expansion of local calling areas, rather than the expansion of rate options from two to four.  In fact, the number of rate options contributes to the confusion of many customers, which leads to dissatisfaction and poor choices.  Obviously not all customers believe LCP is an improvement, some even think its worse than what they had before.  But with Staff's seven suggested modifications to rates, billing practices and route selections, overall customer satisfaction should be increased without diminishing GTE's rate-of-return. Following section is not part of the Review of Current LCP Analysis of Expanded LCP "fit" to Needs of Proposed Exchanges Converting the rest of GTE's toll routes under 24 miles to local calling through LCP would provide much needed relief on many routes, but the fit of LCP's current distance limit to probable need will not be as good as it was in the first eleven exchanges.  Following is at least a partial list of GTE's exchanges currently without LCP and their toll routes for which there may be justification, in part due to common borders and/or county seat ties, for conversion to LCP or some other type of EAS but which are longer than LCP's 23-mile limit: ExchangeToll Routes > 23 Miles with Possible Local Calling Need Priest LakePriest River @24mi. and Sandpoint @25 mi.     Bonners FerrySandpoint @32 mi.     KelloggCoeur d'Alene @33 mi. and Harrison @32 mi.     HarrisonKellogg @32 mi.     St. MariesBovill @31 mi. and possibly Potlatch @32 mi.     BovillSt. Maries @ 31  mi. and Moscow @30 mi.     MoscowBovill @30 mi. GTE proposes to limit LCP to intrastate, intraLATA toll routes even though this restriction would eliminate many routes where both exchanges are owned by GTE and where local calling may be arguably justified and needed.  Following is at least a partial list of these routes: ExchangeInterLATA, Intrastate Toll Routes with Local Need     Plummer/WorleySetters, Rock Creek, and Tensed     Setters Coeur d'Alene and Plummer/Worley     Rock CreekPlummer/Worley     TensedSt. Maries, Plummer/Worley and Potlatch     EvergreenPotlatch, Moscow and Cora     CoraEvergreen GTE's proposal does not mention the Lakeview exchange owned by Midvale Telephone.  However, based on GTE's answer to a staff production request in Case No. GNR-T-93-13, the Company appears to be in favor of including Lakeview in its LCP.  Midvale Telephone may want Lakeview to be exclude from LCP even though its toll routes under 24 miles are all intrastate and within GTE's LATA, and there may be some need for local calling between Lakeview and Bayview and other exchanges. FOOTNOTES 1: Optimal Peak-Load Pricing for Local Telephone Calls, Rolla Edward Park & Bridger M. Mitchell, The Rand Corporation, Rand Publication Series, Santa Monica, California, 1987, p.35. 2: "Demand and Pricing of Telecommunications Services:  Evidence and Welfare Implications", John W. Mayo (Dept. of Econ., U. of Tenn. and Carlos Martins-Filho (Dept. of Econ., Or. St. U.), Rand Journal of Economics, Vol. 24, No. 3, Autumn 1993, pp. 439-454. 3: Staff's percentages may not always agree with those reported by GTE because its Exhibit C erroneously includes survey responses from customers outside the LCP exchanges.