******************************************************** NOTICE ******************************************************** This document was converted from WordPerfect to ASCII Text format. Content from the original version of the document such as headers, footers, footnotes, endnotes, graphics, and page numbers will not show up in this text version. All text attributes such as bold, italic, underlining, etc. from the original document will not show up in this text version. Features of the original document layout such as columns, tables, line and letter spacing, pagination, and margins will not be preserved in the text version. If you need the complete document, download the WordPerfect version or Adobe Acrobat version, if available. ***************************************************************** APPENDIX F DESCRIPTION OF PROPOSED METHODOLOGY FOR ESTIMATING EXPENSES 1. Data Sources used in Regression Analysis. The use of multiple variables in the estimation process required that various data sources be used to determine the common support service expense model inputs. Because the reporting requirements and number of company study areas were different among the reports used, it was necessary to reconcile the data for 1996 expenses, access lines, and dial equipment minutes, as described below. 1996 Expenses Data Source: ARMIS, 43-01 Report, "Subject to Separations" Column. Study Areas (SAs) reconciled: Total SAs from ARMIS 43-01 Report: 127 Less: SAs combined to agree with access line data in ARMIS 43-08: (8) Study Area(s) Combined with PNID MSID CTRH and CTWC CTUP COCA GTCA GNCA GTOR COVA GTVA COTX GTTX PRCC PRPR SAs removed (because not in NECA Tier I reporting): (28) (ALPA, ALWR, COIA, COSI, COMO, COCM, COEM, GTGO, COPA, COQS, COIN, COIL, COWA, GLIL, COKY, CONC, COSC, CONM, COAL, GLMI, GLIN, CENC, CEVA, CEIL, UTMO, UTTX, UTOR, UTWA) SAs used in analysis: 91 Access Lines Data Source: ARMIS, 43-08 Report, Table III, Column (dj) "Total Switched Access Lines", and Column (dm) "Total Access Lines (Switched and Special)" Study Areas (SAs) reconciled: Total SAs from ARMIS 43-08 Report:: 115 Less: SAs combined to agree with ARMIS expense data and NECA usage data: (5) Study Area(s) Combined with NYNY (Conn) NYNY (New York) CBTC (IN & KY) CBTC (Ohio) LTNE (IA & KS) LTNE (Nebraska) SAs removed (because not in NECA Tier I reporting): (19) [ALPA, ALWR, GTNW, GTSO, COSO(AL, IN, MI), COTM, COAT, UTIM, CEVA, CEIL, UTMO(IA, KS, MS), CETO, UTTX, UTNW(OR, WA)] SAs used in analysis: 91 Dial Equipment Minutes Data Source: NECA filed statistics on network usage by carrier Study Areas reconciled: Total SAs per NECA data filing: 104 Less: SAs combined to agree with access line data: (9) Study Area(s) Combined with: GNCA GTOR COCA GTCA PNID MSID CBTC (KY) CBTC (Ohio) CTRH & CTWC CTUP PRCC PRPR COTX GTTX COVA GTVA SAs removed (because not in ARMIS reporting): (4) (GA Alltel Telecom, COTM, Micronesian Tel, Citizens Utilities DBA Citizens of Tennessee) SAs used in analysis: 91 2. Adjustments for Non-Supported Services. This section reviews the methodology that was used to adjust the proposed estimates for corporate operations and customer operations expenses by removing those costs in ARMIS that should not be included in the estimated cost of providing supported services. (a) Marketing. The data used to adjust ARMIS marketing expenses was based on a study done by Economics and Technology, Inc. This study analyzed data from companies providing basic telephone service to apportion marketing expenses between supported and non-supported services. This analysis attributed an average of 95.6 percent of marketing costs to non-supported customers or activities, such as vertical and new services. Based on this analysis, we propose to include 4.4 percent of the regression coefficient for marketing expenses as the base input value for ARMIS 43-08, Account 6610. (b) Other Non-Supported Services. Adjustments were also made to remove from the proposed expense estimates costs related to coin operations and collection, published directory, access billing, interexchange carrier office operation, and service order processing, because these activities are not necessary to provide supported services. Because these services were not specifically identified within the ARMIS 43-08 data used for the regression estimates, it was necessary to estimate the percentage of expenses attributable to these services through another publicly available data source. This was accomplished by using information from a time trend analysis of average ARMIS 43-04 expense data for five years (1993-1997), which separately identifies these services. First, we calculated the ratio of expenses attributable to non-supported activities to the total expenses in ARMIS 43-04 that represent the activities reported in the ARMIS 43-08 subaccounts. Applying these ratios, percentage reductions were then made to the portion of the average regression coefficients for that subaccount. For example, the 1996 ARMIS 43-08 Customer Services Subaccount represents 73 percent of the expenses reported in the Service Expense Account 6620. The total average expenses for non-supported customer services including Coin Collection and Administration, Carrier Access Billing, and Other Customer Services based on the time trend analysis were estimated as 12.46 percent of the total expenses reported in ARMIS 43-04 for those services that would be included in the ARMIS 43-08 Customer Services Subaccount 6623. The Service Expense Account 6620 coefficient was then reduced based on the adjusted sub-account. The following table shows the calculations used to exclude expenses for non-supported services.