FCC 03J-2 Federal Communications Commission Before the Federal Communications Commission Washington, D.C. 20554 CC Docket No. 96-45 ) ) ) In the Matter of Federal-State Joint Board on Universal Service ) RECOMMENDED DECISION Adopted: March 27, 2003 Released: April 2, 2003 By the Federal-State Joint Board on Universal Service: Chairman Bob Rowe, and Commissioners Lila A. Jaber and Billy Jack Gregg issuing separate statements. TABLE OF CONTENTS Paragraph Number I. INTRODUCTION..............................................................................................................1 II. BACKGROUND................................................................................................................2 III. DISCUSSION.....................................................................................................................6 A. Eligibility...............................................................................................................11 1. Background ..............................................................................................11 2. Discussion..................................................................................................15 a. Income-Based Criteria.................................................................15 b. Program-Based Criteria..............................................................20 c. d. National Eligibility Standard for Receipt of Federal Support..........................................................................................25 Duration of an Individual’s Eligibility for Lifeline/Link- Up ..................................................................................................28 B. Verification of Eligibility.....................................................................................31 1. Background ..............................................................................................31 2. Discussion..................................................................................................32 C. Outreach ...............................................................................................................47 1. Background ..............................................................................................47 2. Discussion..................................................................................................50 D. Other Issues ..........................................................................................................57 FCC 03J-2 Federal Communications Commission IV. CONCLUSION ................................................................................................................63 V. RECOMMENDING CLAUSE .......................................................................................64 APPENDIX A: PARTIES FILING INITIAL COMMENTS APPENDIX B: PARTIES FILING REPLY COMMENTS APPENDIX C: LIFELINE/LINK-UP STATE SURVEY APPENDIX D: ESTIMATED INCOME REQUIREMENTS FOR A HOUSEHOLD AT OR BELOW 135% OF THE FEDERAL POVERTY GUIDELINES APPENDIX E: LIFELINE/LINK-UP STATE PROCEDURES: EXAMPLES OF STATE PROGRAMS SUBMITTED BY COMMENTERS APPENDIX F: JOINT BOARD STAFF ANALYSIS OF IMPACT OF ADOPTING AN INCOME-BASED CRITERION OF 135% OF FEDERAL POVERTY GUIDELINES SEPARATE STATEMENT OF COMMISSIONER BOB ROWE, MONTANA PUBLIC SERVICE COMMISSION SEPARATE STATEMENT OF COMMISSIONER LILA A. JABER, FLORIDA PUBLIC SERVICE COMMISSION SEPARATE STATEMENT OF BILLY JACK GREGG, DIRECTOR OF THE CONSUMER ADVOCATE DIVISION, PUBLIC SERVICE COMMISSION OF WEST VIRGINIA 2 FCC 03J-2 Federal Communications Commission I. INTRODUCTION 1. In this Recommended Decision, the Federal-State Joint Board on Universal Service (“Joint Board”) provides the Commission its recommendations regarding Lifeline and Link-Up (collectively “Lifeline/Link-Up”), two federal support programs that are used to advance universal service and to ensure that quality telecommunications services are available to lowincome consumers at just, reasonable, and affordable rates. Since its inception, Lifeline/Link-Up has provided support for telephone service to millions of low-income consumers.1 Despite this success, the Joint Board believes that the program can be further improved. Therefore, the Joint Board recommends that the Commission expand the default federal eligibility criteria to include an income-based criterion and additional means-tested programs. In addition, the Joint Board recommends that the Commission require states, under certain circumstances, to adopt verification procedures. Finally, to more effectively target low-income consumers, the Joint Board recommends that the Commission provide outreach guidelines for the Lifeline/Link-Up program. The Joint Board believes that the recommendations set forth in this document will make the Lifeline/Link-Up program more inclusive and robust, consistent with section 254(b) of the Communications Act of 1934, as amended (“Act”), to meet the twin goals of affordability and increased subscribership.2 II. BACKGROUND 2. Section 254 of the Act codified the Commission’s and states’ historic commitment to advancing the availability of telecommunications services for all Americans. Specifically, section 254(b) establishes principles upon which the Joint Board and the Commission shall base policies for the preservation and advancement of universal service. These principles state that: (1) quality services should be available at just, reasonable, and affordable rates; (2) access to advanced telecommunications and information services should be provided in all regions of the Nation; (3) consumers in all regions of the Nation, including low-income consumers, should have access to telecommunications and information services, that are reasonably comparable to those services provided in urban areas and that are available at rates that are reasonably comparable to rates charged in urban areas; and (4) there should be specific, predictable and sufficient Federal and State mechanisms to preserve and advance universal service.3 Additionally, section 254(e) states that only eligible telecommunications carriers designated pursuant to section 214(e)4 shall be eligible to receive federal universal service support.5 To be designated an eligible telecommunications carrier pursuant to section 214(e), a carrier must 1 See FCC, Trends in Telephone Service Report, Table 7.2, 7.4 (May 2002) (2002 Trends Report) (estimating that 6.2 million people paid reduced rates under the Lifeline program in 2001 and 12.1 million people paid reduced charges under Link-Up since 1987). 2 47 U.S.C. § 254(b). 3 47 U.S.C. § 254(b). 4 47 U.S.C. § 214(e). 5 47 U.S.C. § 254(e). 3 FCC 03J-2 Federal Communications Commission throughout its service area “offer the services that are supported by Federal universal service support mechanisms under section 254(c).”6 Additionally, eligible telecommunications carriers must “advertise the availability of such services and the charges therefor using media of general distribution.”7 Carriers have flexibility to determine appropriate outreach methods and materials employed, subject to any state requirements. 3. The Lifeline/Link-Up program is one of several universal service support mechanisms.8 Universal service support advances the availability of telecommunications services to all consumers, including low-income consumers and those living in rural, insular, and high cost areas at rates that are reasonably comparable to those charged in urban areas.9 Universal service support also promotes the ability of schools, classrooms, libraries and health care providers to have access to advanced telecommunications services.10 The Schools and Libraries program helps to ensure that the nation’s classrooms and libraries receive access to the vast array of educational resources that are accessible through the telecommunications network. The Rural Health Care program helps to link health care providers located in rural areas to urban medical centers so that patients living in rural America will have access to the same advanced diagnostic and other medical services that are enjoyed in urban communities. 4. Lifeline provides low-income consumers with monthly discounts on the cost of receiving telephone service for a single telephone line in their principal residence.11 Link-Up provides low-income consumers with discounts on the initial costs of commencing telephone service.12 Recognizing the unique needs and characteristics of the tribal community, enhanced Lifeline and Link-Up provides qualifying low-income individuals living on tribal lands with additional discounts on the cost of receiving telephone service and the initial costs of commencing telephone service.13 5. On December 21, 2000, the Commission referred low-income support issues to the Joint Board.14 In the Referral Order, the Commission requested the “Joint Board to undertake a 6 47 U.S.C. § 214(e)(1)(A). 7 47 U.S.C. § 214(e)(1)(B). 8 The Commission adopted Lifeline/Link-Up prior to passage of the 1996 Act pursuant to its general authority under sections 1, 4(i), 201, and 205 of the Act. See Federal-State Joint Board on Universal Service, CC Docket No. 96-45, Report and Order, 12 FCC Rcd 8776, 8952-53, para. 329 (1997) (1997 Universal Service Order); 47 U.S.C. §§ 151, 154(i), 201, 205. 9 See 47 U.S.C. § 254(b)(3). 10 See 47 U.S.C. § 254(b)(6). 11 See 47 C.F.R. § 54.401(a)(2); 1997 Universal Service Order, 12 FCC Rcd at 8957, para. 341. 12 See 47 C.F.R. § 54.411(a)(1). 13 See 47 C.F.R. §§ 54.405(a)(4), 54.411(a)(3). 14 See Federal-State Joint Board on Universal Service, CC Docket No. 96-45, Order, 15 FCC Rcd 25257 (2000) (Referral Order). 4 FCC 03J-2 Federal Communications Commission review of Lifeline and Link-Up service for all low-income customers, including a review of the income eligibility criteria.”15 In response to the Referral Order, the Joint Board released a Public Notice seeking comment on its review of Lifeline and Link Up, including enhanced Lifeline and Link Up.16 The Public Notice specifically sought comment on the effectiveness of the current rules, possible modifications to the programs, and outreach efforts.17 III. DISCUSSION 6. As an initial matter, the Joint Board believes that the Lifeline/Link-Up programs generally are effective, and we are confident that the modifications proposed herein will improve them even further. As discussed below, on average, Lifeline/Link-Up assistance programs have produced increased telephone subscription among low income households, as was our goal. Nevertheless, our analysis also reveals that significant differences in low-income telephone penetration exist over time and among the states.18 Therefore we also recommend that the Commission seek more information as to the reasons for these differences in program effectiveness. 7. The data reviewed by the Joint Board suggests that there may be a strong connection between Lifeline/Link-Up assistance and telephone penetration. Between 1984 and 1997, the telephone penetration rate for low-income households in states with Lifeline/Link-Up assistance increased by an average of 0.5% per year. By comparison, the penetration rate for low-income households in states without Lifeline/Link-Up assistance increased by an average of 0.25% per year. The Federal-State Joint Board’s 1999 Monitoring Report found that “the Lifeline program has a positive and significant impact on telephone subscribership, implying that increases in the Lifeline discount would increase telephone penetration.”19 In addition, the Commission’s 1997 Universal Service Order concluded that “providing Lifeline support in all states, irrespective of state participation, will help increase subscribership in those states that presently do not participate in the Lifeline program.”20 8. Additionally, the data reviewed by the Joint Board indicates that, between 1997 and 2001, the penetration rate for low-income households in states with full Lifeline/Link-Up 15 Referral Order, 15 FCC Rcd at 25257. 16 See Federal-State Joint Board on Universal Service Seeks Comment on Review of Lifeline and Link-Up Service for All Low-Income Consumers, CC-Docket 96-45, Public Notice, 16 FCC Rcd 18407 (2001) (Public Notice). 17 See id. at 18409, 18411-14. 18 For example, between 1997 to 2001, low-income telephone penetration increased by 15.9% in Alaska, but decreased by 2.4% in Illinois. See Industry Analysis and Technology Division, Wireline Competition Bureau, Federal Communications Commission, Telephone Penetration By Income By State at 9, Table 3 (rel Apr. 2002). 19 Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission, Federal-State Joint Board December 1999 Monitoring Report at 6 - 7 (rel. Feb. 2000), available at . 20 1997 Universal Service Order, 12 FCC Rcd 8776, 8963-64, para. 353 (1997). 5 FCC 03J-2 Federal Communications Commission assistance increased by an average of 0.61% per year.21 By comparison, the penetration rate for low-income households in states with intermediate Lifeline/Link-Up assistance increased by 0.29%.22 States with basic Lifeline/Link-Up assistance saw a decrease in penetration rates by an average of -0.21% per year.23 Thus, telephone penetration increased at a greater rate, on average, for low-income households in those states where the maximum federal support was provided. 9. Despite significant success of the program in some states, Lifeline/Link-Up continues to serve only a small portion of the low-income households in this country. Lifeline/Link-Up take rates have been highest in states that provide matching funds and engage in proactive targeted efforts such as automatic enrollment, aggressive outreach and intrastate multi-agency cooperation. We agree with the vast majority of commenters that the current Lifeline/Link-Up program could be further improved, consistent with the statutory principle of affordability and the goal to promote universal service by increasing subscribership.24 The Joint Board must also ensure that the public interest is served by the efficient use of universal service support. We believe that the recommendations outlined in this Recommended Decision will achieve these goals. 10. Generally, the Joint Board recommends that the current program-based default federal eligibility criteria should be expanded to include an income-based standard of 135% of the Federal Poverty Guidelines (FPG), the Temporary Aid to Needy Families program (TANF), and the National School Lunch free lunch program (NSL).25 Although the Joint Board strongly encourages the states to adopt these measures, we do not recommend imposing a national eligibility standard on states that currently provide Lifeline/Link-Up support. The Joint Board also recommends that the Commission require states to adopt verification procedures, under 21 Industry Analysis Division, Common Carrier Bureau, Federal Communications Commission, Federal-State Joint Board October 2002 Monitoring Report, Table 6.4, 6 - 19 (rel. Oct. 2002), available at (2002 Monitoring Report). States with full Lifeline/Link-Up assistance provide “sufficient support to get the maximum federal matching support. The total federal and state support in these states was $11.35 or more.” Id. at 6 - 7. 22 2001 Monitoring Report at Table 6.4. States with intermediate Lifeline/Link-Up assistance provide “some support, but less than enough to qualify for the maximum federal support. The monthly level of support in such states was more than $6.10, but less than $11.35.” Id. at 6 - 7. 23 2001 Monitoring Report at Table 6.4. States with basic Lifeline/Link-Up assistance provide “no support, but receiv[e] the basic federal support of $6.10 per line per month.” Id. at 6 – 7. 24 47 U.S.C. § 254(b). 25 We note that the recommended modifications to the federal Lifeline/Link-Up program are general in nature and not specifically intended to affect consumers living on tribal lands. We note that several commenters who specifically addressed tribal needs made generally applicable suggestions, such as adding an income-based criterion and requiring more extensive outreach efforts targeting consumers living on tribal lands. See Confederated Tribes of the Umatilla Indian Reservation Comments at 1-2 (Umatilla); Gila River Telecommunications, Inc. Comments at 5, 12-15 (Gila River); Smith Bagley, Inc. Reply Comments at 3-5, 8-9 (Smith Bagley). We propose many of these suggestions in this Recommended Decision. 6 FCC 03J-2 Federal Communications Commission certain circumstances, to ensure that consumers receiving benefits are eligible. The Joint Board recommends that the Commission issue guidelines on which states and carriers can base their outreach practices, in order to more effectively target low-income consumers and increase participation in the Lifeline/Link-Up programs. The Joint Board believes that gathering data and information about state Lifeline/Link-Up programs will enable the Commission to make more informed decisions in any future Lifeline/Link-Up orders. In order to obtain feedback on the success of the revised Lifeline/Link-Up program, the Joint Board recommends that the Commission adopt a voluntary information collection from the states in its future order adopting changes to the Lifeline/Link-Up program. This voluntary survey, as contained in Appendix C, would ask states to provide information about any eligibility and verification criteria implemented as a result of the Commission’s changes to the Lifeline/Link-Up program.26 States would provide feedback on whether the changes improved telephone penetration rates, on any administrative burdens or inefficiencies the state has experienced, and on suggestions for additional changes to the Lifeline/Link-Up program. The Joint Board recommends that the Commission make submission of this survey voluntary for states with a due date of one year following the effective date of any changes made to the Lifeline/Link-Up program. The Joint Board recommends that the Commission seek comment on the survey’s format and questions asked. Eligibility A. Background 1. 11. Currently, Lifeline/Link-Up eligibility is based on participation in means-tested programs. In order to be eligible for Lifeline/Link-Up assistance under the default federal criteria for states that do not have their own Lifeline/Link-Up program, a consumer must certify, under penalty of perjury, that he/she participates in at least one of the following federal programs: Medicaid, Food Stamps, Supplemental Security Income (SSI), Federal Public Housing Assistance (Section 8), or the Low Income Home Energy Assistance Program (LIHEAP).27 In states that have their own Lifeline/Link-Up programs, the consumer must meet the eligibility criteria established by the state, consistent with sections 54.409 and 54.415 of the Commission’s rules, and follow the state’s certification procedures.28 The current rules allow states that have their own programs flexibility in establishing their own eligibility criteria to fit the unique characteristics of that state, although some of these states elect to use the federal criteria as their default standard.29 12. In the Twelfth Report and Order,30 the Commission adopted more expansive 26 See Appendix C. 27 See 47 C.F.R. §§ 54.409(b), 54.415(b). 28 See 47 C.F.R. §§ 54.409(a), 54.415(a). 29 See 47 C.F.R. §§ 54.409(a). 30 Federal-State Joint Board on Universal Service; Promoting Deployment and Subscribership in Unserved and Underserved Areas, Including Tribal and Insular Areas, CC Docket No. 96-45, Twelfth Report and Order, (continued....) 7 FCC 03J-2 Federal Communications Commission Lifeline/Link-Up eligibility criteria for low-income consumers living on tribal lands.31 For those consumers, the Commission established an enhanced Lifeline/Link-Up program. In order to qualify for enhanced Lifeline/Link-Up if the consumer lives in a state that does not have its own Lifeline/Link-Up program, the consumer must certify, under penalty of perjury, that he/she participates in one of the five programs listed above or any of the following additional federal programs: Bureau of Indian Affairs General Assistance, Tribally-Administered Temporary Assistance for Needy Families (Tribal TANF), Head Start (only for those meeting its income qualifying standard), or the National School Lunch Program’s free lunch program.32 In a state with its own enhanced Lifeline/Link-Up program, a consumer living on tribal lands may also meet the eligibility and certification criteria established by the state.33 But consumers living on tribal lands may still receive federal enhanced Lifeline/Link-Up support even if they do not meet the state’s eligibility criteria, as long as they meet the federal default eligibility criteria for the enhanced program. 13. In January 2002, there were an estimated 107 million total households in the U.S.34 An estimated 17.4 million of these households (or 16.3%) were eligible for Lifeline/Link-Up assistance under the current eligibility criteria.35 Based on Universal Service Administrative Company (USAC) data,36 an estimated 6.5 million low-income households (or 38% of eligible households) actually subscribed to the Lifeline service in 2002.37 USAC also reports that total 2002 expenditures for Lifeline were $647 million.38 (...continued from previous page) Memorandum Opinion and Order, and Further Notice of Proposed Rulemaking, 15 FCC Rcd 12208 (2000) (Twelfth Report and Order). 31 See Twelfth Report and Order, 15 FCC Rcd at 12215, 12243-49. 32 See 47 C.F.R. §§ 54.409(c), 54.415(c). 33 See 47 C.F.R. §§ 54.409(c), 54.415(c). 34 This estimate was based on March 2000 Current Population Survey of Household data (CPSH data), and adjusted for growth. 35 These figures were based on March 2000 CPSH data and adjusted for growth. See also . 36 USAC is a private, not-for-profit corporation that administers the universal service support mechanisms. 37 This data represents average quarterly Lifeline subscribers for the first three quarters of 2002 and estimated average quarterly Lifeline subscribers for the last quarter of 2002. See Federal Universal Service Support Mechanisms Fund Size Projections for the Second Quarter 2003, LI08, available at (filed January 31, 2003) (USAC Filing for Second Quarter 2003 Projections); USAC Lifeline Report for the Fourth Quarter of 2002 (unpublished report). 38 This data represents actual reported Lifeline expenditures for the first nine months of 2002 and estimated Lifeline expenditures for the last three months of 2002. See USAC Filing for Second Quarter 2003 Projections at LI07; Federal Universal Service Support Mechanisms Fund Size Projections for the Fourth Quarter 2002, LI01, available at (filed August 2, 2002) (USAC Filing for Fourth Quarter 2002 Projections). 8 FCC 03J-2 Federal Communications Commission 14. In the Public Notice, the Joint Board sought comment on whether an income-based standard should replace or be added to the current program-based criteria, and whether to add more assistance programs to the current list of default eligibility criteria in order to improve Lifeline/Link-Up participation.39 The Joint Board also sought comment on whether the Commission should adopt one national standard for purposes of determining eligibility for federal support.40 Finally, the Joint Board sought comment on whether the Commission should adopt a set duration period of eligibility for Lifeline/Link-Up.41 Discussion 2. Income-Based Criteria a. 15. The Joint Board recommends that the Commission add an income-based standard to the current default federal eligibility criteria. Specifically, the Joint Board recommends that a consumer be eligible for Lifeline/Link-Up when the consumer’s income is at or below 135% of the Federal Poverty Guidelines (FPG). The Joint Board finds that adding an income-based criterion of 135% of the FPG will increase low-income participation in the Lifeline/Link-Up program. This would enable, for example, a family of four whose annual income is at or below $24,840 to qualify for Lifeline/Link-Up support.42 We have included in Appendix D estimated income requirements for various sizes of households at or below 135% of the FPG.43 We estimate that adding an income-based criterion of 135% of the FPG could result in approximately one million new Lifeline subscribers. Of these new Lifeline subscribers, we project that approximately one quarter would be new subscribers to telephone service. Therefore, in addition to putting many low-income subscribers on the network for the first time, this additional criterion would also ensure that many other low-income subscribers will be better able to afford to remain on the network.44 We believe adding an income-based standard would promote universal service by increasing subscribership and overcoming certain barriers to participation, as described below. 39 See Public Notice, 16 FCC Rcd at 18411-12. 40 See Public Notice, 16 FCC Rcd at 18411. 41 See Public Notice, 16 FCC Rcd at 18411. 42 See 2003 Poverty Guidelines for the 48 Contiguous States and the District of Columbia, 68 Fed. Reg. 6456-58 (2003) (2003 FPG). 43 See Appendix D. 44 See Joint Board Staff Analysis set forth in Appendix F. The staff analysis assumes that all states choose to adopt the new federal default income-based standard in accordance with the Joint Board’s recommendation that the Commission encourage all states to adopt the new recommended federal default criteria. See infra para. 25. Accordingly, the estimates presented are likely to represent the upper limit of potential new Lifeline and telephone subscribers and estimated impact on the fund. If some states choose not to adopt the federal income-based criteria, the number of subscribers would be correspondingly lower. This analysis also assumes the following: states that already have an income criterion of 150% of the FPG or higher keep it; the FPG standards remain the same; there are no other changes to the Lifeline/Link-Up program or the qualifying Lifeline/Link-Up eligibility programs; and people quickly learn of the program change and rapidly act on that information. See Appendix F at 13. 9 FCC 03J-2 16. We note that, in the Twelfth Report and Order, the Commission declined to adopt an income-based criterion, but promised to further examine this possibility in the future because it might “reach more low-income consumers, including low-income tribal members, than the current method of conditioning eligibility on participation in particular low-income assistance programs.”45 The overwhelming majority of commenters support adding an income-based standard to the current program-based criteria.46 We agree with commenters that adding an income-based standard could capture some low-income consumers that are no longer eligible for Lifeline/Link-Up because they no longer participate in the qualifying assistance programs.47 In 1996, Congress passed “The Personal Responsibility and Work Opportunity Reconciliation Act,”48 also known by the acronym “PRWORA.” PRWORA instituted sweeping changes to federal public assistance programs, including time limits and work requirements backed by sanctions. Either as a direct or indirect result of PRWORA, participation is decreasing in many public assistance programs, including programs used to determine eligibility for Lifeline/Link- Up.49 We agree with BellSouth that participation in “the very programs that have been used to meet the needs-based eligibility requirements for Lifeline and Link-Up have been shrinking” as one consequence of PRWORA.50 A few commenters state that individuals that are no longer eligible to receive welfare benefits are still too poor to afford the cost of local telephone 46 See Public Utilities Commission of Ohio Comments at 2 (Ohio Commission); National Association of State Utility Consumer Advocates Comments at 13-18, 37 (NASUCA); NASUCA Reply Comments at 2-3; Gila River Comments at 5; Texas Office of Public Utility Counsel Reply Comments at 1-3 (Texas OPC); United States Conference of Catholic Bishops, Alliance for Community Media, Appalachian People’s Action Coalition, Center for Digital Democracy, Consumer Action, Consumer Federation of America, Edgemont Neighborhood Coalition, Migrant Legal Action Program Comments at 3-5 (U.S. Catholic Bishops); United States Conference of Catholic Bishops, Alliance for Community Media, Appalachian People’s Action, Consumer Federation of America, Edgemont Neighborhood Coalition, Migrant Legal Action Program Reply Comments at 6-7 (U.S. Catholic Bishop); Civil Rights Forum on Communication Policy Comments at 8-9 (Civil Rights Forum); Minnesota Department of Commerce Comments at 3 (Minnesota DOC); BellSouth Corporation Comments at 2 (BellSouth); BellSouth Reply Comments at 1-2; Utility, Cable & Telecommunications Committee of the City Council of New Orleans Comments at 3 (New Orleans Council); Colorado Department of Human Services, Office of Self-Sufficiency, Colorado Office of Consumer Counsel Reply Comments at 3 (Colorado DHS/OCC); Dollar Energy Fund, Inc. Comments at 1 (Dollar Energy Fund); North Dakota Comments Public Service Commissioner Susan Wefald at Attachment C (North Dakota Comments Public Service Commissioner); Oklahoma Corporation Commission Comments at 3-4 (Oklahoma Commission); Smith Bagley Reply Comments at 3-5. 47 BellSouth Reply Comments at 2; Civil Rights Forum Comments at 8; New Orleans Council Comments at 3; NASUCA Comments at 7-10; U.S. Catholic Bishops Comments at 4; U.S. Catholic Bishops Reply Comments at 4-5. 49 For example, participation decreased in Medicaid from 41.2 million enrollees in fiscal year (FY) 1996 to 40.8 million enrollees in FY 1999. See . Participation substantially decreased in Food Stamps from 25.5 million recipients in FY 1996 to 17.3 million recipients in FY 2001. See . Federal Communications Commission 45 Twelfth Report and Order, 15 FCC Rcd at 12247, para. 72. 48 Pub.L.No. 104-193, 110 Stat. 2105 (Aug. 22, 1996). 50 BellSouth Comments at 2. Accord NASUCA Comments at 7-10. 10 FCC 03J-2 Federal Communications Commission service.51 In addition, we believe that many otherwise qualified low-income individuals refuse to participate in public assistance programs because they wish to avoid the stigma that may be associated with such programs.52 In the 1997 Universal Service Order, the Commission promised to monitor the impact of PRWORA and revise eligibility criteria if participation in Lifeline/Link-Up qualifying programs “becomes an unworkable standard.”53 Although we have not reached an “unworkable standard,” the Joint Board now believes that decreasing enrollment in Lifeline/Link-Up qualifying programs requires revisions to the criteria to avoid reaching such a critical point. Therefore, adding an income-based standard could increase subscribership among individuals affected by PRWORA or who wish to avoid the stigma commonly associated with being on welfare. Moreover, adding an income-based standard should also help ensure that low-income subscribers will be better able to afford to remain on the network. Thus, adding this standard will further the goals of section 254 and enhance the value of the network.54 17. We believe that the selection of 135% of the FPG strikes an appropriate balance between increasing subscribership and not significantly burdening the universal service support mechanism. We also believe that it is in accord with recommendations from commenters, other federal welfare programs, and existing state rules. We note that most commenters support adoption of an income-based standard ranging from 125% to 150% of the FPG.55 In addition, many other federal welfare programs base eligibility on an income-based criterion within that range.56 Finally, many state Lifeline/Link-Up programs have an income-based criterion that falls within that range.57 Because the Public Notice in this proceeding did not include a specific FPG proposal, the Joint Board recommends that the Commission seek additional comment on whether 51 Civil Rights Forum Comments at 8 (“[A]s people have been taken off the welfare rolls and begun to work, many [have] incomes at or just above the poverty line and still remain too poor to afford basic telephone service.”). Accord Colorado DHS/OCC Reply Comments at 3. 52 New Orleans Council Comments at 3; Oklahoma Commission Comments at 5; U.S. Catholic Bishops Comments at 19. 53 1997 Universal Service Order, 12 FCC Rcd at 8974, para. 374. 54 47 U.S.C. § 254. 55 See, e.g., Civil Rights Forum Comments at 8; Gila River Comments at 5; Minnesota DOC Comments at 3; NASUCA Comments at 15-18; Ohio Comments at 2; Texas OPC Reply Comments at 2-3; U.S. Catholic Bishops Comments at 4-5. 56 For example, the following federal programs use an income-based standard as an eligibility criterion: Medicaid (income at or below 133% of the FPG), Food Stamps (gross income at or below 130% of the FPG), LIHEAP (income at or below 150% of the FPG or 60% of state median income), National School Lunch program’s free lunch program (income at or below 130% of the FPG). We note that these programs may also use other eligibility criteria. 57 For example, BellSouth Florida, Sprint Nevada, Tennessee, and Texas have an income-based eligibility criterion of 125% of the FPG. Idaho, Qwest Oregon, and Qwest Utah have an income-based eligibility criterion of 133% of the FPG. Verizon Oregon has an income-based eligibility criterion of 135% of the FPG. Pacific Bell California, Kansas, Michigan, Minnesota, Sprint Missouri, Nebraska, Moapa Valley Nevada, Verizon Nevada, Sprint Pennsylvania, and Vermont have an income-based eligibility criterion of 150% of the FPG. We note these programs may also use other eligibility criteria. 11 FCC 03J-2 Federal Communications Commission 135% of the FPG is appropriate or whether a different FPG level should be used for the federal default eligibility criteria. 18. The Joint Board is cognizant that the addition of an income-based standard at 135% of the FPG would likely increase the cost of the Lifeline/Link-Up program.58 The Joint Board believes, however, that the benefits of adding new and maintaining existing low-income households on the network outweigh these potential increased costs.59 We also note that a few commenters express concern that adding an income standard may result in fraud because income levels may be difficult to determine, audit, and verify. Those concerns are addressed in section B of this document, where the Joint Board recommends more stringent verification of eligibility when using an income-based standard. 19. We do not recommend replacing the current default federal eligibility criteria altogether with an income-based standard. We believe that replacing the current program-based criteria with an income-based standard could cause many current subscribers to become ineligible.60 We also find that replacing program-based criteria with income-based criteria could be very disruptive to states that utilize the federal default standard to determine eligibility in their respective Lifeline/Link-Up programs. If the Commission adds an income-based criterion to the federal default criteria, these states can continue to rely on program-based criteria. As the Commission found in the 1997 Universal Service Order, the diversity of the qualifying programs’ eligibility criteria ensures that low-income individuals with disparate public assistance needs remain eligible for Lifeline/Link-Up support.61 Moreover, program-based criteria are easily verified. We also note that commenters overwhelmingly preferred adding an incomebased standard rather than replacing the current program-based criteria with an income-based standard.62 58See Appendix F. Although we estimate that increased funding requirements could be over $100 million, we recognize that it is difficult to predict with certainty how consumers may behave if program requirements change. In addition, the staff analysis assumes that all states choose to adopt the new federal default income-based standard in accordance with the Joint Board’s recommendation that the Commission encourage all states to adopt the new recommended federal default criteria. See infra para. 25. Accordingly, the estimates presented are likely to represent the upper limit of the estimated impact on the fund. If some states choose not to adopt the federal incomebased criteria, the costs would be correspondingly lower. This analysis also assumes the following: states that already have an income criterion of 150% of the FPG or higher keep it; the FPG standards remain the same; there are no other changes to the Lifeline/Link-Up program or the qualifying Lifeline/Link-Up eligibility programs, or to the TANF and NSL programs that we recommend here; and people quickly learn of the program change and rapidly act on that information. See Appendix F at 13. 59 See generally Appendix F. 60 See Appendix F at 12, 32 (Table 4.A). 61 See 1997 Universal Service Order, 12 FCC Rcd at 8974, para. 374. 62 Numerous commenters supported adding an income-based criterion to the current program-based criteria. See e.g., BellSouth Comments at 2; BellSouth Reply Comments at 1-2; Civil Rights Forum Comments at 8-9; Colorado DHS/OCC Comments at 3; Dollar Energy Fund Comments at 1; Gila River Comments at 5; Minnesota DOC Comments at 3; NASUCA Comments at 14-18, 37; NASUCA Reply Comments at 3; New Orleans Council Comments at 3; North Dakota Public Service Commissioner Comments at Attachment C; Ohio Commission Comments at 2; (continued....) 12 FCC 03J-2 Federal Communications Commission Program-Based Criteria b. 20. The Joint Board recommends adding two additional assistance programs to the current list of default federal eligibility criteria. Specifically, the Joint Board recommends that the Commission add the Temporary Assistance to Needy Families program (TANF)63 and the National School Lunch free lunch program (NSL) because we believe these programs will help to capture more low-income individuals and, in doing so, increase telephone subscribership among low-income households.64 21. Under the Commission’s current rules, Tribal TANF is an eligibility criterion for enhanced Lifeline/Link-Up.65 As discussed below, TANF eligibility varies by state. Adding TANF to the current list of eligibility criteria would permit more low-income individuals, not just those living on tribal lands, to qualify for Lifeline/Link-Up support, thereby increasing telephone subscribership. Although 2.1 million families currently participate in TANF,66 we cannot project how many additional persons may become eligible for Lifeline/Link-Up if TANF is adopted because many low-income households typically participate in more than one assistance program once they meet the qualifying criteria. Some people who are enrolled in TANF may already be participating in Lifeline/Link-Up through another qualifying assistance program. 22. The Joint Board believes that one benefit of adding TANF to improve telephone penetration among low-income subscribers is the broad discretion states are given to establish eligibility standards for each state’s respective TANF program.67 This broad discretion enables states to tailor the TANF program to their constituents’ needs. Another advantage of adding TANF is that verification of Lifeline/Link-Up eligibility would simply involve checking TANF (...continued from previous page) Oklahoma Commission Comments at 3-4; Smith Bagley Reply Comments at 3-5; Texas OPC Reply Comments at 1-3; U.S. Catholic Bishops Comments at 4-5; U.S. Catholic Bishop Reply Comments at 6-7. We note that only one commenter supported replacement of the current program-based criteria with an income-based standard. See Dollar Energy Fund Comments at 1. 63 TANF replaced the Aid to Families with Dependent Children program (AFDC). TANF is codified at 42 U.S.C. §§ 600 et seq. 64 We note that both TANF and NSL are subject to modification, as are all the means-tested programs that comprise Lifeline/Link-Up’s program-based criteria. 65 In Tribal TANF, participation is only open to those living on tribal lands, and tribes implement their own TANF programs with eligibility criteria and benefits that vary by tribe rather than by state. See . 66 In fiscal year 2001, there were approximately 2.1 million families receiving TANF support. This includes tribal and non-tribal families. See . Data for Tribal TANF is incomplete because not all tribes reported enrollment data. See . 67 We note that flexible state use of TANF funds has led to some controversy. Much of the controversy surrounds state partnerships with religious organizations, and the potential use of TANF funds for non-income-based purposes, such as preserving marriage and reducing teenage pregnancy rates. 13 FCC 03J-2 Federal Communications Commission program records. Furthermore, commenters suggest that TANF captures many low-income households that may not participate in other Lifeline/Link-Up qualifying public-assistance programs.68 We note that in the 1997 Universal Service Order, the Commission rejected a proposal to add TANF’s predecessor, AFDC, to the list of qualifying Lifeline/Link-Up programs.69 At the time, the Commission was concerned about the impact of PRWORA on that particular program.70 Although the Joint Board agrees that this program has been significantly impacted by PRWORA as evidenced by decreased TANF participation rates since fiscal year 1996, participation rates are still high.71 Furthermore, in the Twelfth Report and Order, the Commission extended Lifeline/Link-Up eligibility criteria to include the Tribal TANF program, concluding that the “household income thresholds for these newly added programs range from 100-130 percent of the [FPG]” and were therefore “consistent with the [income thresholds of those] programs included in our current federal default list.”72 Therefore, we believe that extending Lifeline/Link-Up benefits to TANF participants will increase telephone subscribership. 23. The Joint Board also recommends that the Commission add the NSL free lunch program.73 Under the Commission’s current rules, Tribal NSL is an eligibility criterion for enhanced Lifeline/Link-Up on tribal lands.74 To be eligible for NSL’s free lunch program, the household income must be at or below 130% of the FPG, which is $23,920 for a family of four.75 The child, not the parent, is the named applicant. Adding NSL to the current list of default federal eligibility criteria would permit more low-income individuals, not just those living on tribal lands, to qualify for Lifeline/Link-Up support, thereby increasing subscribership. There were approximately 16 million children enrolled in NSL in fiscal year 2001.76 As with TANF, however, we cannot project how many additional persons may become eligible for Lifeline/Link- Up if NSL is adopted because many low-income households typically participate in more than one assistance program once they meet the qualifying criteria. We also note that there is no data 68 See National Consumer Law Center on behalf of Massachusetts Union of Public Housing Tenants Comments at 11- 12 (NCLC); North Dakota Public Service Commissioner Comments at 3, Attachment C; U.S. Catholic Bishops Comments at 18-19. 69 See 1997 Universal Service Order, 12 FCC Rcd at 8974, para. 374. 70 See id. 71 See supra note 61. 72 Twelfth Report and Order, 15 FCC Rcd at 12245, para. 68. 73 See . NSL is codified at 42 U.S.C. §§ 1751 et seq. 74 In Tribal NSL, participation is only open to children living on tribal lands, and children living on tribal lands are automatically eligible if they or their household receives assistance under the Food Distribution Program on Indian Reservations. See . 75 See 2003 FPG, 68 Fed.Reg. at 6456-58. 76 This does not include Tribal NSL, as no tribal enrollment data is currently available. See . 14 FCC 03J-2 Federal Communications Commission on the total number of households in which NSL participants reside, because more than one NSL participant may reside in a single household. 24. The Joint Board agrees with the U.S. Catholic Bishops that adding NSL may improve telephone penetration among low-income subscribers because it may capture many low-income households that may not participate in other Lifeline/Link-Up qualifying public-assistance programs.77 Participation in the NSL program is increasing, unlike other assistance programs where PRWORA has prompted decreased enrollment.78 Also, adding NSL is consistent with the Commission’s determination in the Twelfth Report and Order that eligibility for Lifeline/Link- Up should be limited to those qualifying for free lunch.79 This captures only the neediest of families. Another advantage of adding NSL is that verification of eligibility would simply involve checking NSL program records. Accordingly, we believe that adding NSL will help to target Lifeline/Link-Up support to the appropriate low-income households. National Eligibility Standard for Receipt of Federal Support c. 25. The Joint Board recommends that the Commission strongly encourage states to incorporate the federal eligibility changes into state programs and to implement them accordingly. The Joint Board does not recommend, however, that the Commission mandate any federal criteria for states because we believe states should maintain the flexibility to respond to the needs of their constituents. For example, the Joint Board does not recommend that the Commission establish a mandatory national eligibility standard for receipt of federal Lifeline/Link-Up funds. A national standard set by the Commission would impose a uniform list of eligibility criteria on all states, including those with their own Lifeline/Link-Up programs. States would not be free to add or remove eligibility criteria. The Joint Board also does not recommend the establishment of a minimum federal floor upon which states would be able to expand their own eligibility criteria. A minimum floor set by the Commission would impose a uniform list of eligibility criteria on all states, but would leave states free to expand eligibility criteria to better target the needs of their constituents. Although the Public Notice sought comment on these proposals, at this time, the Joint Board recommends retention of the current rules that permit states to establish their own eligibility criteria if they have their own Lifeline/Link-Up program.80 26. Although a national eligibility standard or a minimum federal floor might bring a certain level of uniformity among states, we believe, as do many commenters, that, generally, states are in a better position than the federal government to target the needs of their own 77 U.S. Catholic Bishops Comments at 19. 78 For example, in 1997, there were 15.8 million children enrolled in the NSL free lunch program. In 2001, there were 16 million children enrolled in the NSL free lunch program. See . 79 Twelfth Report and Order, 15 FCC Rcd at 12215, 12245. 80 See Public Notice, 16 FCC Rcd at 18411. 15 FCC 03J-2 Federal Communications Commission consumers.81 As explained above, if a state has its own state Lifeline/Link-Up program, it may use its own eligibility criteria. The Joint Board believes that states should continue to have the flexibility to permit low-income consumers to receive federal and state Lifeline/Link-Up support if the consumer participates in state-administered public assistance programs, such as Alaska’s Heating Assistance Program,82 in addition to broader-based federal public assistance programs such as Food Stamps. For states that choose to adopt an income-based standard, as is recommended here for the federal default eligibility criteria, this will also permit the states flexibility to tailor that standard to economic variables unique to the state’s cost of living, per capita income, and demographics.83 27. The recommended modifications to the federal Lifeline/Link-Up eligibility criteria, if adopted, will only affect a state that uses the default federal eligibility criteria. This includes states that do not have their own Lifeline/Link-Up program, as well as states that have their own programs but have chosen to use the federal default eligibility criteria. If the Commission adopts these recommendations, a low-income consumer residing in such a state will be eligible to receive federal Lifeline/Link-Up support if that consumer’s income is at or below 135% of the FPG, or participates in TANF or NSL, in addition the current default federal eligibility criteria. States that have adopted their own Lifeline/Link-Up programs and use their own criteria would not be affected by the proposed changes to the federal default eligibility criteria unless they choose to adopt such criteria. Duration of an Individual’s Eligibility for Lifeline/Link-Up d. 28. The Joint Board recommends retention of the current rule that permits consumers to participate in Lifeline only as long as they meet the eligibility criteria. This rule also requires all Lifeline subscribers to notify their carriers when the eligibility criteria are no longer met, so that only qualifying consumers receive Lifeline support.84 In the Public Notice, the Joint Board sought comment on whether Lifeline enrollment should be guaranteed for a specified minimum period of time even if a subscriber subsequently becomes ineligible to receive Lifeline benefits.85 29. Although the Joint Board recommends that consumers be removed from the Lifeline program if they no longer meet the eligibility standards, we also believe that consumers should be given an opportunity to appeal a finding of ineligibility. We agree with commenters that allowing Lifeline benefits to continue prior to a final decision to terminate enrollment may help to ensure uninterrupted telephone service necessary for finding and keeping a job, and may help 81 See BellSouth Reply at 2; People of the State of California, California Public Utilities Commission Comments at 5-6 (California PUC); Regulatory Commission of Alaska Comments at 2-3 (Alaska Commission); Verizon Telephone Companies Reply Comments at 4 (Verizon). 82 See . 83 See BellSouth Reply Comments at 2. 84 See 47 C.F.R. § 54.409(b). 85 See Public Notice, 16 FCC Rcd at 18411. 16 FCC 03J-2 Federal Communications Commission if a subscriber’s financial situation temporarily fluctuates.86 Accordingly, we recommend that the Commission adopt a federal rule that requires carriers to notify consumers of their impending termination of Lifeline benefits, and to implement an appeals process. For example, carriers could send a termination of Lifeline benefits notice in a letter separate from the consumer’s monthly bill.87 If a consumer receives such a termination notice, the consumer could have up to 60 days in which to appeal to their carrier before Lifeline support is discontinued. The 60 day time period may ensure that consumers have ample notice to make arrangements to pay the full cost of local service, should they wish to continue telephone service, and permit consumers ample time to appeal if they believe their Lifeline benefits have been wrongly terminated. We recommend that the Commission obtain more information on how such an appeals process could work in order to balance the needs of Lifeline recipients with the administrative burden that an appeals process may impose on carriers. We recommend an appeals process only in circumstances in which the carrier has initiated the termination of benefits. We do not believe that an appeals process is necessary where the Lifeline subscriber himself has notified the carrier that he is no longer eligible. 30. Because a specific time period for appealing termination of Lifeline benefits was not specified in the Public Notice, we recommend that the Commission obtain more information on an appropriate time period for such an appeal. Although the Joint Board’s recommendation of a federal notice and appeals process will only affect carriers in states that have not adopted their own program and in states with their own programs that have adopted the federal default criteria, the Joint Board recommends that the Commission encourage all states to require carriers to implement termination and appeal procedures. Verification of Eligibility B. Background 1. 31. In the Public Notice, the Joint Board sought comment on the efficacy of application, certification, and verification procedures, including whether automatic enrollment or other verification methods should be adopted.88 Currently, in a state that has instituted its own Lifeline/Link-Up program, an individual must follow that state’s certification procedures, if any, in order to enroll in that state’s Lifeline/Link-Up program. In a state that has not instituted its own Lifeline/Link-Up program, an individual must self-certify to his/her carrier, under penalty of perjury, that he/she is enrolled in a qualifying assistance program. Certification occurs at the time an individual is applying to enroll in Lifeline/Link-Up, while verification occurs on a periodic basis after the subscriber has already been certified. There is no current federal verification requirement to check on a Lifeline subscriber’s continued eligibility, although an 86 See Indiana Utility Regulatory Commission Comments at 6 (Indiana URC); Minnesota DOC Comments at 4; Umatilla Comments at 2. 87 See, e.g., Indiana URC Comments at 6. We note that the Indiana URC suggests that a termination notice could be sent either in the form of a bill insert or by direct notification 30 days prior to the date of termination. See Indiana URC Comments at 6. 88 See Public Notice, 16 FCC Rcd at 18412-13. 17 FCC 03J-2 Federal Communications Commission individual is required to notify his/her carrier when he/she no longer meets the eligibility standards.89 Discussion 2. 32. The Joint Board does not recommend modifying the current certification procedures for enrollment using program-based eligibility. The Commission should continue to require selfcertification, under penalty of perjury, as the federal default rule, while allowing states the necessary flexibility to require more strict measures for certification as they deem appropriate. The Joint Board does recommend, however, that consumers eligible for federal or state Lifeline/Link-Up support under an income-based criterion should be required to present documentation of income eligibility prior to being enrolled in the program. 33. The Joint Board agrees with commenters that the current federal self-certification rules for enrollment should be retained with respect to program-based criteria. We believe that the ease of self-certification encourages eligible consumers to participate in Lifeline/Link-Up.90 The absence of a documentation requirement alleviates the burden on consumers to prove eligibility, and eliminates administrative costs associated with certifying documentation, ensuring that consumers receive telephone service as soon as possible.91 We also note that participation in public assistance programs is easily verified and the record contained no evidence of fraud and abuse resulting from the use of self-certification.92 Furthermore, the Joint Board believes that the safeguard currently in place, certification under penalty of perjury, serves as an effective disincentive to abuse the system.93 34. The Joint Board does recommend, however, that consumers eligible for federal Lifeline/Link-Up support under an income-based criterion be required to present documentation of income eligibility before enrolling in Lifeline/Link-Up. The Joint Board is concerned that there may be a greater potential for fraud and abuse when an individual self-certifies his/her income eligibility than there appears to be when an individual is enrolled in a qualifying program because program enrollment is more easily verified. Accordingly, the Joint Board recommends that the Commission require all states, including states that use the federal default criteria, to adopt certification procedures to document income-based eligibility for Lifeline/Link-Up enrollment in order for the carriers in that state to continue to receive federal Lifeline/Link-Up support. The Joint Board believes that states should be given the flexibility to determine the certification procedures and the carriers should be required to perform the certification. The 89 47 C.F.R. § 54.409(b). 90 See Oklahoma Commission Comments at 2. 91 See Indiana URC Comments at 7; NASUCA Comments at 19-21, 23; New Orleans Council Comments at 3-4; Western Wireless Corporation Comments at 3 (Western Wireless). 92 See Colorado DHS/OCC Reply Comments at 5; California PUC Comments at 7; NASUCA Reply Comments at 7-8; U.S. Catholic Bishops Comments at 14; U.S. Catholic Bishops Reply Comments at 10-11. 93 See 47 C.F.R. § 54.409(b). 18 FCC 03J-2 Federal Communications Commission Joint Board believes affording states this degree of flexibility will be especially useful for states that have differing sources of income data. Such flexibility will also provide multiple means of proving income levels by recognizing that no single source of income data would capture all consumers eligible for Lifeline/Link-Up. 35. States could access the documentation via an online database, if available in that state, or could require consumers to provide one or more forms of documentation from the following list: a tax return from the prior year, a current income statement from an employer or a paycheck stub, a Social Security statement of benefits, a Veterans Administration statement of benefits, a retirement/pension statement of benefits, an Unemployment/Workmen’s Compensation statement of benefits, a divorce decree or child support document, or other official governmental agency documents.94 We note that there are important factors to consider for each form of documentation. For example, although tax returns are widely available, they only reflect income earned one year ago. As another example, a current income statement from an employer or a paycheck stub would be an indicator of current income, but might be incomplete because the consumer might have more than one job. States that choose to include these types of documentation as acceptable forms of proof of income-eligibility should additionally require consumers to certify, under penalty of perjury, that the income identified for eligibility purposes includes all income currently being received by all members of the consumer’s household. 36. For states that use the federal default criteria or states that do not have jurisdiction over carriers, we recommend that the Commission adopt federal default criteria for documentation of income eligibility. We recommend that the federal default criteria require consumers to provide one or more forms of documentation from the list above, with the exception of “other official governmental agency documents.” We also recommend that the federal default criteria also require consumers to certify, under penalty of perjury, that the income identified for eligibility purposes includes all income currently being received by all members of the consumer’s household. 37. The Joint Board also recommends that all states, including states that use the federal default criteria, require Lifeline/Link-Up consumers that are qualifying under the income criteria to self-certify, under penalty of perjury, the number of individuals in their household. Such a measure is required for determining income-eligibility at or below 135% of the FPG, but the number of people in a household may not be readily apparent depending on the type of documentation presented.95 In addition to documentation, random auditing can also be used as an effective method of certifying income eligibility.96 38. The Joint Board also recommends that the Commission encourage all states, including states that use the federal default criteria, to adopt automatic enrollment as a means of 94 See Minnesota DOC Comments at 4; New Orleans Council Comments at 3-4. 95 For example, a tax return lists the number of people in a given household, but a current income statement from employer or paycheck card does not. 96 See Colorado DHS/OCC Reply Comments at 6; Missouri Commission Comments at 3; NASUCA Comments at 25; Texas OPC Reply Comments at 4; U.S. Catholic Bishops Reply Comments at 12. 19 FCC 03J-2 Federal Communications Commission certifying that consumers are eligible for Lifeline/Link-Up and also to encourage enrollment in Lifeline/Link-Up. Automatic enrollment is an electronic interface between a state agency and the carrier that allows low-income individuals to automatically enroll in Lifeline/Link-Up following enrollment in a qualifying public assistance program. The Joint Board agrees with commenters who state that automatic enrollment will increase participation and aid administrative efficiency by identifying eligible consumers.97 39. We believe, as do many commenters, that states who wish to implement automatic enrollment procedures should follow the lead of other states with similar procedures in place, and treat public assistance enrollment data in a confidential manner. 98 This will alleviate the privacy concerns that some commenters raise.99 In addition, consumers should have the opportunity to decline enrollment in Lifeline/Link-Up if they choose.100 For example, in the Massachusetts automatic enrollment program, households that qualify for LIHEAP must give permission to release their personal information before they may be enrolled into Lifeline/Link- Up.101 We note that no commenter provided any specific data that demonstrated consumers’ displeasure with automatic enrollment.102 40. The Joint Board recognizes the additional administrative burden and cost associated with implementing automatic enrollment procedures.103 To provide guidance and minimize administrative burdens, the Joint Board has included an appendix that lists states that commenters have identified as having successfully implemented automatic enrollment procedures.104 The appendix describes the automatic enrollment programs of three states as examples of different ways that a state can implement such a program.105 97 See Civil Rights Forum Comments at 3-4; Colorado DHS/OCC Reply Comments at 4; Indiana URC Comments at 7; NASUCA Reply Comments at 6; U.S. Catholic Bishops Comments at 16-17. 98 See NASUCA Comments at 21, n.52 (carriers have generally been able to secure and execute agreements for the exchange of sensitive data in states with automatic enrollment procedures); U.S. Catholic Bishops Comments at 16-17; U.S. Catholic Bishops Reply Comments at 12, 14-15 (states with automatic enrollment procedures have been successful in ensuring confidential treatment of public assistance data). 99 See BellSouth Comments at 2-3; BellSouth Reply Comments at 3 (stating that the Privacy Act may place some constraints on the type of automatic enrollment program that can be implemented); SBC Comments at 5. 100 See NASUCA Reply Comments at 6. 101 See NCLC Reply Comments at 5, n.11. 102 See Beacon Telecommunications Advisors, LLC Comments at 2 (Beacon) (questioning the appropriateness of carriers, who are not typically in the position of administering social service programs, verifying the eligibility of participants in these programs). 103 See BellSouth Reply Comments at 3; Verizon Comments at 6-7. 104 See Appendix E. 105 See id. 20 FCC 03J-2 Federal Communications Commission 41. The Joint Board recommends that the Commission should require all states, including those that use the federal default criteria, to establish procedures to verify consumers’ continued eligibility for the program in order for the state to continue to receive federal Lifeline/Link-Up support. The Joint Board believes that states should be given the flexibility to determine the verification procedures and the carriers should be required to perform the verification. This flexibility will permit states to determine what verification procedures best accommodate their own Lifeline participants, based on the available resources of carriers and state commissions, each states’ eligibility criteria, and local conditions. As commenters have indicated, states have strong incentives to control fraud and prevent abuse in their programs.106 Prevention of fraud and abuse will maintain the integrity of this universal service support mechanism and will help to control costs. 42. Verification procedures could include random beneficiary audits, periodic submission of documents, or annual self-certification.107 States may wish to implement an on-line verification database for verification of program-based eligibility and adopt verification procedures such as documentation for income-based eligibility.108 The Joint Board strongly encourages states to adopt an on-line verification process, where states can obtain and provide data to allow carriers real-time access to a database of low-income assistance program participants. Numerous commenters support this form of verification as a streamlined process for both consumers and carriers.109 Some states already utilize some form of on-line verification because it provides quick, easy, and accurate information. On-line verification allows a carrier to immediately verify that a consumer receives public assistance, whether or not the consumer is a current telephone subscriber.110 An on-line verification database will also inform carriers about those customers no longer enrolled in qualifying public assistance programs.111 To assist state efforts in determining appropriate verification procedures to ensure continued eligibility, we have included an appendix that describes various Lifeline state verification procedures.112 These state procedures include on-line verification databases used in Illinois, Minnesota, and an on-line verification database used in Tennessee in conjunction with a documentation requirement.113 106 See BellSouth Reply Comments at 4; Indiana URC Comments at 7; Ohio Commission Comments at 4; Oklahoma Commission Comments at 4; Staff of Washington UTC Comments at 2, 4-5. 107 See Colorado DHS/OCC Reply Comments at 5; NASUCA Comments at 25; Public Service Commission of the State of Missouri Comments at 3 (Missouri Commission); Texas OPC Reply Comments at 4; U.S. Catholic Bishops Reply Comments at 10-12. 108 See also Appendix E. 109 See Colorado DHS/OCC Comments at 6; NASUCA Comments at 22-23; SBC Communications Inc. Comments at 2 (SBC); Texas OPC Reply Comments at 3; Staff of Washington UTC Comments at 2. 110 See NASUCA Comments at 23. 111 See SBC Comments at 6-7. 112 See Appendix E. 113 See Appendix E. 21 FCC 03J-2 Federal Communications Commission 43. For states that use the federal default criteria or states that do not have jurisdiction over carriers, the Joint Board recommends that the Commission adopt federal default criteria for verification of program-eligibility. Therefore, we recommend that the Commission adopt the following federal default verification procedure. In order to verify a consumer’s continued participation in one of the Lifeline/Link-Up qualifying programs, we recommend that carriers be required to send annual verification forms to a certain percentage of Lifeline subscribers or to a statistically valid sample of Lifeline subscribers. Subscribers who are subject to this verification must prove they are still eligible for Lifeline, by sending a copy of their Medicaid card or other Lifeline/Link-Up qualifying public assistance card. If subscribers do not return the form or cannot prove they are eligible, the aforementioned appeals process would be triggered permitting subscribers 60 days to prove they are once again eligible. We recommend that any fraud discovered through this verification process be reported to the Commission. 44. With regard to verifying income-eligibility, the Joint Board recommends using verification methods similar to those recommended for certifying consumer income for initial enrollment in the Lifeline/Link-Up program. For states that use the federal default criteria or states that do not have jurisdiction over carriers, we recommend that the Commission adopt the same federal default verification criteria as for certification of income-eligibility, including selfcertification, under penalty of perjury, that the income identified for eligibility purposes includes all income currently being received by all members of the consumer’s household.114 The Joint Board also recommends that all states, including states that use the federal default criteria, require Lifeline/Link-Up consumers that are qualifying under the income criteria to self-certify, under penalty of perjury, the number of individuals in their household. Such a measure is required for verifying income-eligibility at or below 135% of the FPG, because the number of people in a household may not be readily apparent depending on the type of documentation presented.115 45. We disagree with those commenters who state that verification would be expensive or administratively burdensome or that the cost of verification would exceed losses resulting from fraud and abuse.116 Nor do we think that it would be unduly burdensome to require states to undertake periodic verification of customers’ eligibility for the Lifeline/Link-Up program.117 Verification is an effective way to prevent fraud and abuse and ensure that only eligible consumers receive benefits. 46. The Joint Board recommends that states be given one year to implement new verification procedures. We recognize that states use a variety of verification procedures and may need time to review and evaluate the efficacy of these procedures. Programs such as on-line 114 See para. 35, supra. 115 For example, a tax return should list the number of people in a given household, but a current income statement from employer or paycheck does not. 116 See California PUC Comments at 7; Colorado DHS/OCC Reply Comments at 5; Gila River Comments at 8-9; Smith Bagley Reply Comments at 5-6; U.S. Catholic Bishops Comments at 14. 117 See Smith Bagley Reply Comments at 6 (explaining that on tribal lands, customers are spread across a wide area). 22 FCC 03J-2 Federal Communications Commission verification may require extensive coordination between agencies and carriers and may be expensive to implement. Some states may not have verification procedures at present and will need to determine what type of verification is most effective and allocate resources accordingly. The Joint Board also recommends, however, that the Commission seek comment on whether a one-year implementation period is appropriate. Outreach C. 1. Background 47. In the Public Notice, the Joint Board sought comment on whether more extensive consumer education and outreach efforts were necessary to increase participation in Lifeline/Link-Up.118 In the Public Notice, the Joint Board also invited commenters to describe state Lifeline/Link-Up procedures that have been successful at increasing participation in the Lifeline/Link-Up program in that state.119 Currently, there are no specific federal outreach standards. Eligible telecommunication carriers are, however, required to publicize the availability of Lifeline/Link-Up in a manner reasonably designed to reach those likely to qualify for the service.120 48. Effective outreach programs have been shown to improve Lifeline/Link-Up participation. According to an August 2000 report by the Telecommunications Industries Analysis Project, the Lifeline/Link-Up take rate almost triples from 13.1% to 39.6% when states implement outreach initiatives designed to increase telephone penetration and participation.121 For example, Maine reports that its penetration rate among low-income households increased from 90.5% in March 1997 to 97.6% in March 2001 due to its aggressive outreach program, which includes coordinating with social service agencies and sending flyers and personal letters to eligible customers.122 49. In July 2002, the Commission’s Consumer and Governmental Affairs Bureau (CGB) announced the kick-off of “Get Connected-Afford-A-Phone,” a national campaign designed to educate consumers, including tribal consumers, about the Lifeline/Link-Up program.123 CGB 118 See Public Notice, 16 FCC Rcd at 18413. 119 See Public Notice, 16 FCC Rcd at 18414. These state procedures are described in detail in Appendix E. 120 See 47 C.F.R. §§ 54.405(b), 54.411(d). See also Twelfth Report and Order, 15 FCC Rcd at 12250, para. 78 (amending sections 54.405 and 54.411 of the Commission’s rules). 121 Carol Weinhus, Tom Wilson, Gordon Calaway, et al., Telecommunications Industries Analysis Project, Calculations and Sources for Closing the Gap: Universal Service for Low-Income Households, August 1, 2000. 122 Telephone Penetration Report at table 4 (Ind. Anal. and Tech. Div. rel. Apr. 2002), available at . 123 FCC Kicks Off Campaign To Educate Consumers About Phone Service Programs For Low-Income Consumers Lifeline and Link-Up Programs Provide Discounted Phone Service To Eligible Consumers, News Release, July 22, 2002. 23 FCC 03J-2 Federal Communications Commission also engages in extensive outreach to tribal populations for certain federal programs, such as the availability of discounts for obtaining wireless licenses on tribal lands, in addition to Lifeline/Link-Up benefits. 2. Discussion 50. The Joint Board recommends that the Commission provide outreach guidelines to states and carriers. The Joint Board agrees with commenters that more vigorous outreach efforts are necessary to improve Lifeline/Link-Up subscribership.124 We agree with the majority of commenters who believe the Commission should not require specific outreach procedures, but should instead provide guidelines for states and carriers so that they can adopt their own specific standards and engage in outreach themselves.125 We also recommend that the Commission encourage states to establish partnerships with other state agencies and telephone companies in order to maximize public awareness and participation in the Lifeline/Link-Up program.126 The guidelines would provide states and carriers with examples of how to reach those likely to qualify, but would still allow states and carriers to retain authority to determine the most appropriate outreach mechanisms for their consumers. 51. The Joint Board recommends the following guidelines: (1) states and carriers should utilize outreach materials and methods designed to reach households that do not currently have telephone service; (2) states and carriers should develop outreach advertising that can be read or accessed by any sizeable non-English speaking populations within the carrier’s service area; and (3) states and carriers should coordinate their outreach efforts with governmental agencies/tribes that administer any of the relevant government assistance programs. These guidelines are described in detail in the paragraphs below. An appendix compiling state practices is also included in this document.127 State practices include establishing marketing boards to devise outreach materials, providing multi-lingual customer support, and implementing innovative tribal outreach practices. 52. The first recommended guideline is that states and carriers should utilize outreach materials and methods designed to reach households that do not currently have telephone service. 124 See Alaska Commission Comments at 6-7; BellSouth Reply Comments at 4-5; Civil Rights Forum Comments at 4; Colorado DHS/OCC Comments at 9-10; Colorado DHS/OCC Reply Comments at 6; Dollar Energy Fund Comments at 2; Indiana URC Comments at 8-9; Katherine Keller Reply Comments at 1-2; Minnesota DOC Comments at 5; New Orleans Council Comments at 5; Oklahoma Commission Comments at 5-6; Staff of Washington UTC Comments at 6- 7; Western Wireless Comments at 4-5. 125 See AT&T Corp. Comments at 4 (AT&T); BellSouth Comments at 4; BellSouth Reply Comments at 4; California PUC Comments at 8-9; Colorado DHS/OCC Comments at 9-10; Colorado DHS/OCC Reply Comments at 6; Florida PSC Comments at 8-9; Indiana URC Comments at 8-9; NASUCA Reply Comments 8-9; Texas OPC Reply Comments at 6. 126 See AT&T Comments at 4; BellSouth Comments at 4; BellSouth Reply Comments at 4; California PUC Comments at 8-9; Colorado DHS/OCC Comments at 9-10; Colorado DHS/OCC Reply Comments at 6; Florida PSC Comments at 8-9; Indiana URC Comments at 8-9; NASUCA Reply Comments 8-9; Texas OPC Reply Comments at 6. 127 See Appendix E. 24 7. Comments at 6-7. FCC 03J-2 States or carriers may wish to send frequent mailings to eligible households in the form of letters or brochures.128 Posters could be placed in locations where low-income individuals are likely to visit, such as shelters, soup kitchens, and public assistance agencies.129 Commenters suggest multi-media outreach approaches such as newspaper advertisements, articles in consumer newsletters, press releases, radio commercials, and radio and television public service announcements.130 For low-income consumers that live in remote areas, including those living on tribal lands, going door-to-door or setting up an information booth at a central location may be more suitable outreach methods.131 States and carriers should ensure that outreach materials and methods accommodate low-income individuals with sight, hearing, and speech disabilities by producing brochures, mailings and posters in Braille, and providing customer service through telecommunications relay services (TRS), text telephone (TTY), and speech-to-speech (STS) services.132 States and carriers should also take into consideration that some low-income consumers may be illiterate or functionally illiterate, and therefore should consider how to supplement outreach materials and methods to accommodate those individuals. We note that some commenters suggest disseminating Lifeline/Link-Up information over the Internet as a means of providing information to low-income individuals.133 The Joint Board believes, however, that although websites are helpful in providing information generally, the Internet should not be relied on as a primary means of Lifeline/Link-Up outreach because many lowincome individuals may lack Internet access.134 We also note that one commenter suggests that Lifeline/Link-Up should be prominently advertised in carriers’ telephone phone books.135 128 See Colorado DHS/OCC Comments at 9-10; Colorado DHS/OCC Reply Comments at 6; Florida PSC Comments at 130 Alaska Commission Comments at 6-7; Minnesota DOC Comments at 5; Staff of Washington UTC Comments at 6- 131 Alaska Commission Comments at 6-7; Smith Bagley Reply Comments at 2, 7-8; Staff of Washington UTC 132 See Katherine Keller Reply Comments at 1-2. TRS are “telephone transmission services” that enable individuals with a hearing or speech disability to communicate “by wire or radio with a hearing individual in a manner that is functionally equivalent to the ability of an individual” without a hearing or speech disability to communicate over wire or radio. Examples of TRS include TTY and STS services. 47 C.F.R. § 64.601(7). TTY is “a machine that employs graphic communication in the transmission of coded signals through a wire or radio communication system.” 47 C.F.R. § 64.601(8). STS “allows people with speech disabilities to communicate with voice telephone users through the use of specially trained [communications assistants (CAs)] who understand the speech patterns of persons with disabilities and can repeat the words spoken by that person.” 47 C.F.R. § 64.601(10). 133 See Alaska Commission Comments at 6-7; BellSouth Comments at 5; Katherine Keller Reply Comments at 1. 134 See Civil Rights Forum Comments at 6 (outreach efforts using information on Lifeline/Link-Up on websites are insufficient); Oklahoma Commission Comments at 6. Useful website information may include the amount a consumer can save on their telephone bill, eligibility requirements, program restrictions, and instructions on how to apply for Lifeline/Link-Up. See BellSouth Comments at 5. We note that a lot of this information is currently available at . Federal Communications Commission 5; Indiana URC Comments at 8-9; Minnesota DOC Comments at 5. 129 See Minnesota DOC Comments at 5. 135 See Katherine Keller Reply Comments at 1. 25 FCC 03J-2 Federal Communications Commission Although this may be effective in reaching some low-income individuals, it will not be effective for those without established phone service because carriers only distribute telephone books after phone service is established.136 Similarly, states and carriers should not rely on hotlines as a primary outreach method because many low-income individuals may not have access to a telephone from which to initiate an inquiry on Lifeline/Link-Up benefits.137 53. The second recommended guideline is that states and carriers should develop outreach advertising that can be read or accessed by any sizeable non-English speaking populations within the carrier’s service area. For example, many of the suggestions in the above paragraph can be implemented in languages other than English, including mailings, print advertisements, radio and television commercials, and posters. States with a large ethnically diverse population should have a toll-free call center to answer questions about Lifeline/Link-Up in the low-income population’s native languages.138 Similarly, enrollment applications should be made available in other languages.139 54. The third recommended guideline is that states and carriers should coordinate their outreach efforts with governmental agencies that administer any of the relevant government assistance programs.140 Commenters also suggest cooperative outreach efforts with state commissions, carriers, social service agencies, community centers, nursing homes, public schools, and private organizations that may serve low-income individuals, such as American Association for Retired Persons and the United Way.141 Cooperative outreach among those most likely to have influential contact with low-income individuals will help to target messages about Lifeline/Link-Up to the low-income community.142 For example, state agencies that conduct outreach efforts for a state’s “earned income tax credit,” an income tax credit for low-income working individuals and families, could conduct simultaneous outreach efforts for Lifeline/Link- Up. Commenters also suggest that establishing a marketing or consumer advisory board with state, carrier, non-profit and consumer representatives may be an effective way of developing outreach materials.143 The Joint Board also recommends that states and carriers should also issue 136 See Civil Rights Forum Comments at 6 (explaining that Lifeline/Link-Up information in telephone books is an insufficient outreach effort). 137 See Alaska Commission Comments at 6-7; Florida PSC Comments at 8-9. 138 See Civil Rights Forum Comments at 4 (describing California’s toll-free call center which answers questions about Lifeline in Spanish, Korean, Laotian, Cambodian, Vietnamese, Tagalog, and Hmong, in addition to English). See also Appendix C. 139 See Minnesota DOC Comments at 5. 140 See Gila River Comments at 12. 141 See Alaska Commission Comments at 6-7; BellSouth Comments at 4; BellSouth Reply Comments at 4; Colorado DHS/OCC Comments at 9-10; Colorado DHS/OCC Reply Comments at 6; Indiana URC Comments at 8-9. 142 See Bell South Reply Comments at 4-5. 143 See California PUC Comments at 8-9; NASUCA Reply Comments at 8-9; Ohio Commission Comments at 5. 26 FCC 03J-2 Federal Communications Commission a joint report to the Commission as to the state’s outreach practices. 55. The Joint Board also recommends that the Commission encourage states to utilize USAC as a resource for outreach to states and carriers, similar to USAC’s outreach efforts with regard to the Rural Health Care and Schools and Libraries programs. USAC currently engages in limited outreach for the Lifeline/Link-Up program. Its primary means of outreach is its website, www.lifelinesupport.org, which has information about state Lifeline/Link-Up programs, eligibility criteria, and information for carriers. USAC speaks about Lifeline/Link-Up at public events such as the National Association of Regulatory Utility Commissioners (NARUC) conference and the National Conference of American Indians. USAC distributes letters to consumer groups, tribal leaders, and social service organizations to publicize the availability of Lifeline/Link-Up and also sends letters to carriers to remind them of their outreach obligations. USAC also frequently takes phone calls from consumers and others with questions about the Lifeline/Link-Up program. 56. In addition, the Joint Board recommends that USAC assist in outreach efforts for Lifeline/Link-Up similar to what USAC currently does for the Rural Health Care and Schools and Libraries Programs. For example, USAC could work with various organizations that have contact with low-income individuals, such as state welfare agencies, tribal leaders, places of worship, community centers, the National Association for the Advancement of Colored People (NAACP), the Urban League, and others, to assist them in marketing the Lifeline/Link-Up program to eligible members of their organizations using materials and methods best suited for that particular organization. USAC could hold conference calls with these parties to assist them with the Lifeline/Link-Up application process or with other concerns they may have, comparable to what USAC currently does with both the Rural Health Care and Schools and Libraries programs. USAC could also host conferences for these parties, giving them an overview of the Lifeline/Link-Up program, updating them on news about the program, and providing them with presentations and speakers, in geographically strategic areas. Other Issues D. 57. Commenters proposed several additional changes to the current Lifeline/Link-Up program. Some of these suggestions included recommending that the Commission adopt rules: (1) governing the disconnection of Lifeline/Link-Up support for failure to pay toll charges; (2) imposing additional toll blocking requirements; (3) permitting non-eligible telecommunications carriers to receive federal Lifeline/Link-Up support; and (4) restricting the purchase of vertical services by Lifeline customers. 58. NASUCA proposes that the Joint Board recommend rules governing disconnection of Lifeline/Link-Up consumers for non-payment of toll charges.144 The Joint Board declines to recommend such a rule. We note that, in the 1997 Universal Service Order, the Commission adopted the Joint Board’s recommendation that carriers should be prohibited from disconnecting 144 See NASUCA Comments at 32, 36 (suggesting the Commission revisit its earlier position on this matter before it was overruled in the 5th Circuit Court). NASUCA and Ohio both support a prohibition on disconnection of local service for nonpayment of toll charges. See NASUCA Reply Comments at 9; Ohio Commission Comments at 7. 27 FCC 03J-2 Federal Communications Commission Lifeline consumers’ local service for nonpayment of toll charges, and adopted section 54.401(b) stating that “[e]ligible telecommunications carriers may not disconnect Lifeline service for nonpayment of toll charges.”145 In Texas PUC v. FCC, however, the Fifth Circuit found that the Commission lacked jurisdiction to take such actions without additional justification and that the Commission had “failed to show why allowing states to control disconnections from local service” would interfere with federal objectives.146 In light of the Fifth Circuit’s ruling, section 54.401(b) was repealed in 1999.147 Accordingly, the Joint Board recommends that the Commission take no action on disconnection requirements at this time. 59. Although we decline to make the recommendation proposed by NASUCA regarding our earlier position on disconnection for nonpayment of toll charges, the Joint Board does acknowledge that carriers often prohibit consumers who have prior outstanding balances for local and/or long distance services, but who otherwise qualify for Lifeline/Link-Up, from signing up for local telephone service. As a result, these outstanding balances stand as a barrier to expanding subscribership among low-income consumers. Therefore, the Joint Board recommends that the Commission encourage states to implement rules that require carriers to offer Lifeline service to consumers who may have been previously disconnected for unpaid toll charges.148 For example, Florida has a provision within its state Lifeline program that requires carriers to provide Lifeline even when the consumer has been disconnected for non-payment of toll charges.149 The Lifeline service for such a consumer requires the consumer to accept tollblocking and to commit to pay back the prior long-distance balance on a monthly basis.150 The Joint Board also recommends that the Commission seek comment on whether it would be possible to modify the Link-Up program to directly address outstanding balances for local and long distance service without conflicting with the Fifth Circuit’s prior decision. For example, one option might be to modify the Link-Up program to include provisions to assist low-income consumers in payment of such outstanding balances. 145 1997 Universal Service Order, 12 FCC Rcd at 8983, para. 390; 47 C.F.R. § 54.401(b) (repealed in 1999 and reserved). 146 Texas PUC v. FCC, 183 F.3d 393, 421-425 (5th Cir. 1999) (finding that the Commission had no “‘unambiguous or straightforward’ grant of authority to override the limits set by [section] 2(b) [of the Communications Act of 1934], and, accordingly, it has no jurisdiction to adopt the ‘no disconnect’ rule on the basis of the vague, general language of [section] 254(b)(3).”). 147 See Federal State Joint Board on Universal Service Access Charge Reform, Sixteenth Order on Reconsideration, CC Docket No. 96-45, Eight Report and Order, CC Docket No. 96-45, Sixth Report and Order, CC Docket No. 96-262, 15 FCC Rcd 1679, 1693, para. 34 (1999) (repealing, inter alia, section 54.401(b) of the Commission’s rules). 148 Pursuant to the Commission’s Link-Up rules, a consumer shall only receive the benefit of the Link-Up program for a second or subsequent time for a principal residence that is different from the one where the Link-Up assistance was previously provided. See 47 C.F.R. § 54.411(c). 149 See Florida PSC Comments at 4. See also FLA. STAT. § 364.604(4) (“A billing party shall not disconnect a customer's Lifeline local service if the charges, taxes, and fees applicable to basic local exchange telecommunications service are paid.”). 150 See id. 28 FCC 03J-2 Federal Communications Commission 60. Contrary to the suggestion of the U.S. Catholic Bishops,151 the Joint Board does not recommend that the Commission adopt any additional rules governing toll blocking. The U.S. Catholic Bishops suggest that the rules be revised to prohibit carriers who do not offer toll blocking from requiring deposits from Lifeline/Link-Up customers in order to receive local service.152 Currently, the Commission’s rules prohibit carriers from requiring a Lifeline/Link-Up subscriber to pay a service deposit in order to initiate service if the consumer elects to receive toll blocking.153 Carriers may collect service deposits from consumers who do not elect to receive toll blocking or if the carrier does not have the technical ability to offer toll blocking.154 We believe that, at this time, all carriers have the technological capability to offer toll blocking, and that therefore this proposal is moot. In addition, the Joint Board recommends that states should remain free to decide whether to adopt rules for deposits and/or credit checks of lowincome consumers in their respective jurisdictions.155 61. Contrary to the suggestion of AT&T and NALA/PCA,156 the Joint Board does not recommend that telephone companies who are not eligible telecommunications carriers should receive federal support for providing Lifeline/Link-Up service. AT&T suggests that the Commission allow telephone companies to receive federal Lifeline/Link-Up support when they have qualified for state support under parallel state programs.157 NALA/PCA believes that allowing resellers who are not eligible telecommunications carriers to receive federal Universal Service support will increase competition and choice for low-income consumers.158 The Joint Board believes, however, that only eligible telecommunications carriers should receive universal service funds, including federal Lifeline/Link-Up support. We note that the Commission found in the 1997 Universal Service Order that “[although] we have the authority under sections 1, 4(i), 201, 205 and 254 to extend Lifeline to include carriers other than eligible telecommunications carriers…we decline to do so at the present time.”159 The Joint Board 151 See U.S. Catholic Bishops Comments at 25-27; U.S. Catholic Bishops Reply Comments at 17; accord Colorado DHS/OCC Comments at 8 (suggesting that toll blocking as a deposit alternative will enable more customers to obtain phone service regardless of their credit history). 152 See U.S. Catholic Bishops Comments at 25-27; U.S. Catholic Bishops Reply Comments at 17. 153 See 47 C.F.R. § 54.401(c). 154 See id. 155 See e.g. Minnesota DOC Comments at 5. Minnesota explains that its credit and deposit rules do not require a deposit for a customer with good credit. A customer who has not had service disconnected in the last twelve months for nonpayment and has not been liable for disconnection for nonpayment is deemed to have good credit. For a customer who does not have good credit, a deposit of no more than two months of the gross bill is required but the deposit must be refunded after twelve consecutive months of prompt payment. See id. 156 See AT&T Comments at 2-4; NALA/PCA Comments at 4-5. 157 See AT&T Comments at 2-4. 158 See NALA/PCA Comments at 4-5. 159 1997 Universal Service Order, 12 FCC Rcd at 8971, para. 369. 29 FCC 03J-2 Federal Communications Commission continues to believe that only eligible telecommunications carriers should receive universal service funds, including federal Lifeline/Link-Up support. We agree with the Commission that “a single support mechanism with a single administrator following similar rules will have significant advantages in terms of administrative convenience and efficiency.”160 Furthermore, eligible telecommunications carriers are carriers that agree to certain obligations in order to receive universal service support.161 Existing rules ensure that Lifeline/Link-Up funding goes only to eligible telecommunications carriers that have committed to these obligations, including the provision of supported services in accordance with section 54.101 of the Commission’s rules.162 Finally, to alleviate concerns that pure resellers may not be able to offer Lifeline service to their low-income customers, we note that incumbent LECs are required to offer Lifeline service at wholesale rates, pursuant to section 251(c)(4), to those carriers that provide service purely by reselling another carrier’s services, so that these pure resellers can offer Lifeline discounts to qualifying low-income consumers.163 62. Contrary to the suggestion of the Ohio Commission,164 the Joint Board does not recommend that the Commission adopt rules prohibiting Lifeline/Link-Up customers from purchasing vertical services (e.g., call waiting, call forwarding, voice mail, caller identification). The Ohio Commission suggests that the Commission should prohibit a Lifeline/Link-Up customer from ordering vertical services and should also prohibit the marketing of vertical services to Lifeline/Link-Up customers.165 The Joint Board believes that any restriction on vertical services is outside the scope of the Lifeline/Link-Up program, and, in addition, this issue may be entirely within the purview of the states.166 The Joint Board recognizes, however, that restrictions on the purchase of vertical services may discourage qualified consumers from enrolling in the Lifeline/Link-Up program, effectively serving as a barrier to participation.167 Accordingly, the Joint Board recommends that the Commission encourage states not to adopt rules that would restrict Lifeline/Link-Up customers from purchasing vertical services. For the same reason, the Joint Board does not recommend imposing restrictions on the marketing of vertical services to Lifeline/Link-Up customers. 160 1997 Universal Service Order, 12 FCC Rcd at 8971, para. 369. 161 See 47 C.F.R. § 54.201(d)(1)-(2). 162 47 C.F.R. § 54.101. 163 See 1997 Universal Service Order, 12 FCC Rcd at 8972, para. 370. 164 See Ohio Commission Comments at 6-7. But see NASUCA Reply Comments at 6 (supports prohibiting carriers from marketing vertical services to Lifeline/Link-Up customers but believes those customers should not be barred from ordering vertical services). 165 See Ohio Commission Comments at 6-7. 166 We also note that, as more telecommunications services are sold in bundled form, it may be more difficult to maintain a restriction on the purchase of only vertical services. 167 See Civil Rights Forum Comments at 9 (stating that customers should have choices in their telephone service). 30 FCC 03J-2 Federal Communications Commission IV. CONCLUSION 63. In conclusion, the Joint Board recommends that the Commission adopt the abovementioned modifications to the Lifeline/Link-Up program. Adding an income-based criterion and the TANF and NSL programs to the current federal default eligibility criteria will allow the Lifeline/Link-Up program to adapt to the changes resulting from PROWRA and otherwise address issues associated with receiving public assistance. Adding certification and verification requirements will ensure that only eligible low-income individuals receive benefits, thereby preventing fraud and abuse. Adopting outreach guidelines will facilitate the marketing of the Lifeline/Link-Up program to eligible individuals and increase telephone subscribership among low-income households. Finally, issuing a survey form like the one contained in Appendix C, will enable the Commission to gather data and information from states regarding the administration of Lifeline/Link-Up programs.168 The Joint Board believes that the proposed modifications to the Lifeline/Link-Up program will increase Lifeline/Link-Up subscription rates and make phone service affordable to more low-income individuals and families. V. RECOMMENDING CLAUSE 64. For the reasons discussed herein, the Federal-State Joint Board on Universal Service, pursuant to sections 254 and 410(c) of the Communications Act of 1934, as amended, 47 U.S.C. §§ 254, 410(c), recommends that the Commission consider the Joint Board’s recommendations to modify the federal Lifeline/Link-Up program, including positions relating to state Lifeline/Link-Up programs. FEDERAL COMMUNICATIONS COMMISSION Marlene H. Dortch Secretary 168 See Appendix C. 31 FCC 03J-2 Federal Communications Commission APPENDIX A PARTIES FILING INITIAL COMMENTS Commenter Abbreviation Office of Self-Sufficiency American Public Communications Council APCC AT&T Corp. AT&T Beacon Telecommunications Advisors, LLC Beacon BellSouth Corporation BellSouth Civil Rights Forum on Communication Policy Civil Rights Forum Colorado Department of Human Services Colorado DHS/OCC Colorado Office of Consumer Counsel Dollar Energy Fund Florida PSC Gila River Telecommunications Lake Superior Chippewa Indians Minnesota Office of the Attorney General- Confederated Tribes of the Umatilla Indian Reservation Umatilla Dollar Energy Fund, Inc. Florida Public Service Commission Gila River Telecommunications, Inc. Indiana Utility Regulatory Commission Indiana URC Lac du Flambeau Band of Lake Superior Chippewa Indians Legal Services Advocacy Project LSAP Minnesota Department of Commerce Minnesota DOC Minnesota Department of Human Services Residential and Small Business Utilities Division National ALEC Association/Prepaid Communications NALA/PCA Association National Association of Regulatory Utility Commissioners NARUC National Association of State Utility Consumer NASUCA Advocates National Congress of American Indians NCAI National Consumer Law Center on behalf of NCLC Massachusetts Union of Public Housing Tenants North Dakota Public Service Commissioner Susan Wefald North Dakota Public Service Commissioner Oklahoma Corporation Commission Oklahoma Commission People of the State of California California PUC California Public Utilities Commission Public Service Commission of the State of Missouri Missouri Commission Public Utilities Commission of Ohio Ohio Commission Regulatory Commission of Alaska Alaska Commission SBC Communications Inc. SBC Sprint Corporation Sprint Staff of the Washington Utilities and Transportation Staff of Washington UTC Commission Tennessee Regulatory Authority Tennessee Regulatory Authority 32 FCC 03J-2 Federal Communications Commission United States Department of Agriculture, USDA FNS Food and Nutrition Service United States Conference of Catholic Bishops U.S. Catholic Bishops Alliance for Community Media Appalachian People’s Action Coalition Center for Digital Democracy Consumer Action Consumer Federation of America Edgemont Neighborhood Coalition Migrant Legal Action Program Universal Service Administrative Company USAC Utility, Cable & Telecommunications Committee New Orleans Council of the City Council of New Orleans Verizon Telephone Companies Verizon Western Wireless Corporation Western Wireless WorldCom, Inc. WorldCom 33 FCC 03J-2 Federal Communications Commission APPENDIX B PARTIES FILING REPLY COMMENTS Commenter Abbreviation BellSouth Corporation BellSouth Colorado Department of Human Services Colorado DHS/OCC Office of Self-Sufficiency Colorado Office of Consumer Counsel Katherine Keller National Association of State Utility Consumer Advocates NASUCA National Consumer Law Center on behalf of NCLC Massachusetts Union of Public Housing Tenants Smith Bagley, Inc. Smith Bagley Texas Office of Public Utility Counsel Texas OPC United States Conference of Catholic Bishops U.S. Catholic Bishops Alliance for Community Media Appalachian People’s Action Consumer Federation of America Edgemont Neighborhood Coalition Migrant Legal Action Program Verizon Telephone Companies Verizon 34 Federal Communications Commission APPENDIX C LIFELINE/LINK-UP STATE SURVEY 1. What changes, if any, has the state implemented in its Lifeline/Link-Up program due to changes in the federal Lifeline/Link-Up program? Of those changes, which have been most effective in increasing the state’s telephone penetration rate? 2. Please provide any additional information the state wishes to submit regarding positive or negative results experienced due to adoption of new Lifeline/Link-Up procedures during the past 12 months. 3. Please provide any additional information the state wishes to submit regarding any administrative burdens or inefficiencies that the state has experienced due to adoption of new Lifeline/Link-Up procedures during the past 12 months. 4. Describe the state’s Lifeline/Link-Up eligibility requirements. 5. Describe the state’s Lifeline/Link-Up procedures for enrollment and certification, including documentation requirements. Do any state agencies qualify applicants for the Lifeline/Link-Up program? 6. Describe the state’s Lifeline/Link-Up procedures for verification, including documentation requirements. If the state plans to implement a verification program, please describe. 7. List suggestions for improvements to the federal Lifeline/Link-Up program. 8. Does the state require all incumbent LECs to provide Lifeline/Link-Up Service? 9. Does the state require all competitive LECs to provide Lifeline/Link-Up Service? 35 FCC 03J-2 Federal Communications Commission APPENDIX D ESTIMATED INCOME REQUIREMENTS FOR A HOUSEHOLD AT OR BELOW 135% OF THE FEDERAL POVERTY GUIDELINES Size of Family Unit 1 2 3 4 5 6 7 8 For each additional person, add 48 Contiguous States and D.C. Alaska Hawaii $ 12,123 16,362 20,601 24,840 29,079 33,318 37,557 41,796 4,239 36 FCC 03J-2 $15,134 $13,946 20,439 18,819 25,745 23,693 31,050 28,566 36,356 33,440 41,661 38,313 46,967 43,187 52,272 48,060 4,874 5,306 FCC 03J-2 Federal Communications Commission APPENDIX E LIFELINE/LINK-UP STATE PROCEDURES: Examples of State Programs Submitted by Commenters I. Eligibility Paperless Enrollment Application A. Self-Certification of Eligibility for Enrollment 1. California169 In California, telephone companies must “immediately enroll” a customer who verbally certifies that he or she is eligible to participate in the Lifeline program. The company then sends the customer a self-certification form on which the customer affirms in writing that he or she is eligible for Lifeline and agrees that the company may verify his or her income. If the customer does not return the form within 30 days or if the company determines that the customer is not in fact eligible, the customer is removed from the program. B. Automatic Enrollment 1. Colorado170 Colorado has implemented a paperless application process that allows potential recipients, after being notified of eligibility, to call their local telephone company to receive the discounts. There is no written application. This paperless application process makes it easier for the consumer to get the needed assistance and also enables low-income consumers to choose a competitive LEC that offers the assistance to eligible subscribers using the same paperless application process as the incumbent LEC. There is no paper application to keep track of and transfer from company to company. C. 1. Massachusetts171 In Massachusetts, households that qualify for LIHEAP can voluntarily give their permission, at the time of application, for the LIHEAP-administering agency to disclose information to Verizon that allows the household to be enrolled in Lifeline. Thus, enrollment is not “automatic” in the sense of being done without the household’s permission, but it is done electronically in most cases. This facilitates enrollment, and the results are evident in the relatively high Lifeline subscription rate in Massachusetts. 169 See NCLC Comments at 5-6. 170 See Colorado DHS/OCC Comments at 4. 171 See NCLC Comments at 6. 37 FCC 03J-2 Federal Communications Commission 2. New York172 In New York State, the Public Utility Law Project (PULP) has spent several years working to increase participation rates in the Lifeline/Link-Up programs. PULP represents low-income and rural consumers in utility, telephone and energy related matters. PULP worked with the New York Public Service Commission (NYPSC), the New York Department of Family Assistance (NYFDA), and NYNEX (now Verizon) to create an automatic enrollment database. The data transferred between the NYDFA and Verizon is confidential and cannot be used by Verizon or the state for any reason other than Lifeline assistance. Anytime an individual enrolls for a program administered by NYDFA they are automatically enrolled in Lifeline/Link-Up, but are also given the option to opt-out of the Lifeline/Link-Up program. Individuals who are not Verizon customers but have been identified by NYDFA as being eligible because of enrollment in a program administered by NYDFA are notified of their eligibility and given the opportunity to request Lifeline service by returning a pre-printed form. This system increased the number of people participating in Lifeline from 197,339 in 1987 to 703,001 in 1998. Lifeline consumers who have ceased receiving other assistance through NYDFA for four consecutive months are removed from Lifeline. 3. North Dakota173 In North Dakota, when consumers go to the county office of North Dakota Department of Human Services (NDHS) and are determined eligible for any of the qualifying programs in the North Dakota Lifeline and Link-Up program, they receive an information sheet about Lifeline/Link-Up or enhanced Lifeline/Link-Up. Each qualifying individual receives a certificate of eligibility in the mail from NDHS which states that the individual must return this certificate to the telephone company in order to receive Lifeline/Link-Up. Once a year, all eligible North Dakotans receive a new qualifying certificate from the NDHS. The annual mailing of this certificate to eligible parties helps increase participation in Lifeline and Link-Up programs by providing an additional opportunity to sign up with the local telephone company. Qwest and some other North Dakota companies use a different method of verification. Through arrangements with NDHS, these companies receive an annual list of eligible participants to verify against their current participation list and delete unqualified participants based on this list. Participants with these companies do not need to send in a qualifying certificate annually. D. Paper-proof Verification of Continued Eligibility 1. Tennessee174 172 See Civil Rights Forum Comments at 3. 173 See North Dakota Public Service Commissioner Comments at 1. 174 See Tennessee Regulatory Authority Comments at 11-12. 38 FCC 03J-2 Federal Communications Commission The process used in Tennessee initially requires the applicant requesting Link-Up and Lifeline to provide proof of the public assistance program they receive. Proof of benefits may be demonstrated by providing a copy of the approval letter to receive Food Stamps, Medicaid or TANF from Tennessee Department of Human Services (TDHS) or a copy of the SSI benefit letter from the Social Security Administration. On-line Verification of Continued Eligibility E. Illinois175 1. In Illinois, carriers can perform on-line verification of a consumer’s eligibility by obtaining realtime access to a database of state low-income assistance program participants. The result is a streamlined process for both consumers and carriers. 2. Minnesota176 Minnesota verifies the income and/or disability of all applicants. An enrollee’s continued participation in the program is also verified on an annual basis. Minnesota verifies 85% of its Telephone Assistance Program participants by the use of computer interfaces with the Minnesota Department of Revenue, public assistance databases, and LIHEAP databases. The remainder are contacted by mail and asked to provide proof of continuing eligibility. Due to these verification procedures, Minnesota is not aware of problems with ineligible or fraudulent individuals being enrolled in the Telephone Assistance Program. 3. Tennessee177 In Tennessee, Lifeline applicants are required to certify eligibility by presenting documentation to their carrier of their participation in Food Stamps, Medicaid, TANF, or SSI. Documentation can be demonstrated by a copy of their approval letter to receive benefits through one of those programs. Self-certification is not permitted. Once the documentation is received by the carrier, the carrier then verifies the accuracy of the documentation with the Tennessee Department of Human Services (TDHS) client database. Verification of continued eligibility is also accomplished utilizing this electronic system. This has been the most efficient and effective way in which to verify and re-verify that a consumer is receiving public assistance. Tennessee requires re-verification of consumers on Lifeline no less than twice a year or every six months. 175 See SBC Comments at 2. 176 See Minnesota DOC Comments at 4. 177 See Tennessee Regulatory Authority Comments at 11-12. 39 FCC 03J-2 Federal Communications Commission II. OUTREACH A. Multi-Lingual Assistance Florida179 1. California178 On December 11, 2001, the California PUC approved a one-year, $5 million contract to design and implement a competitively neutral public awareness and outreach program in order to increase universal Lifeline telephone service subscribership. On the same date, the California PUC approved a three-year, $1.5 million contract for a multi-lingual toll-free call center that provides customer service information about Lifeline in Spanish, Korean, Laotian, Cambodian, Vietnamese, Tagalog and Hmong, as well as English. As a result of California’s outreach efforts, Lifeline participation rates have increased from 1,467,859 in 1989 to 3,196,661 in 2000. 2. The Florida Public Service Commission sends eligible Florida consumers a postcard-size flier about the Lifeline/Link-Up program. Approximately 35,000 of the fliers, which were written in English on one side and Spanish on the other, were mailed to consumers in 2000. 1. 3. Minnesota180 To accommodate the state’s increasingly diverse community, the Minnesota Department of Human Services currently makes Lifeline/Link-Up applications available in Arabic, Hmong, Cambodian, Lao, Russian, Somali, Spanish and Vietnamese. 4. Tennessee181 The Tennessee Regulatory Authority (TRA) has created four color posters in English and Spanish and posted them in locations frequented by low-income individuals, such as health care facilities, legal offices, churches, charitable organizations, and Human Services offices. To support this campaign, the TRA has established a toll-free hotline. The TRA has produced public service announcements for radio and television. B. Tribal Outreach Arizona and New Mexico182 178 See Civil Rights Forum Comments at 4; NCLC Comments at 5. 179 See Florida PSC Comments at 7. 180 See Minnesota DOC Comments at 5. 181 See Civil Rights Forum Comments at 5. 182 See Smith Bagley Reply Comments at 2, 7-8. 40 FCC 03J-2 Federal Communications Commission In Arizona and New Mexico, Smith Bagley, a wireless carrier, conducts intensive advertising campaigns on tribal reservations in service areas where they are designated as an eligible telecommunications carrier. One of its most successful forms of outreach is its day-long event. Smith Bagley moves its storefront into town for a day and hosts a sign-up event where customers can learn about wireless service, determine their eligibility for Lifeline/Link-Up, sign up for service, have car installations done, obtain training on using a cell phone, and ask Smith Bagley’s staff any questions they may have about Lifeline/Link-Up or wireless service. This unique outreach event has led to an increase of 14,000 new Lifeline subscribers. C. Agreement with Carriers 1. Florida183 The Florida Public Service Commission (Florida PSC) has recently approved a joint stipulation between the Florida Office of Public Counsel and BellSouth that established a Community Service Fund for use in educating customers and promoting BellSouth’s Lifeline/Link-Up services. As part of the stipulation, BellSouth agreed to contribute $250,000 in 2002 and $150,000 in 2003. D. “Warm Transfer Line” 1. Florida184 The Florida PSC has made consumer education about Lifeline a priority. The Florida PSC operates an innovative “warm transfer line” which allows consumers who call the agency with Lifeline/Link-Up questions to be automatically transferred to the appropriate eligible telecommunications carrier providing phone service in their service area. The warm transfer line assures consumers that they will be in touch directly with the company who can initiate the service. E. Coordination with Organizations and Other Agencies 1. Florida185 The Florida PSC also works closely with key state agencies, such as the Florida Department of Children and Families (DCF) and Department of Community Affairs, to ensure that the materials are received by the target population. For example, the Florida PSC created a postcard-sized flier to be sent to eligible Florida consumers using the DCF’s mailing lists and mail system. Approximately 35,000 of the fliers, which were written in English on one side and Spanish on the other, were mailed to consumers in 2000. Finally, the Florida PSC is partnering with the 183 See Florida PSC Comments at 4. 184 See Florida PSC Comments at 7. 185 See Florida PSC Comments at 7. 41 FCC 03J-2 Federal Communications Commission American Association of Retired Persons (AARP), the Florida Association of Counties, and the Florida League of Cities to further promote Lifeline/Link-Up. Lifeline/Link-Up Seminars F. 1. Rhode Island186 In Rhode Island, consumer advocates hold annual forums and conferences, often consisting of panels in which local telephone company representatives speak about Lifeline and distribute brochures. 2. Tennessee187 The Tennessee Regulatory Authority (TRA) has implemented several methods to promote Lifeline and Link-Up. It has created a Manager of Consumer Outreach position that concentrates on providing consumer information. This Manager conducts three or four Lifeline/Link-Up seminars per month at nursing homes across Tennessee. At the seminar, brochures and applications are distributed, leading to numerous applications for Lifeline/Link-Up. Brochures are also distributed at various public affairs events. G. Direct Mailings 1. Connecticut188 The Connecticut Department of Social Services works in conjunction with carriers to target eligible low-income consumers through the mail. 2. Idaho189 The state of Idaho sends flyers and brochures printed by the Idaho Public Utilities Commission to eligible state residents. 3. Maine190 In late 1999, the Maine State Housing Authority and the Maine Community Action Programs jointly carried out two major mass mailings to all eligible LIHEAP recipients notifying those 186 See Universal Service Administrative Company Comments at 10 (USAC). 187 See Civil Rights Forum Comments at 5. 188 See USAC Comments at 14. 189 See USAC Comments at 14. 190 See USAC Comments at 9. 42 FCC 03J-2 Federal Communications Commission consumers that they were also eligible for Lifeline. An estimated 134,000 letters and flyers were mailed, paid for by the Maine Telecommunications Education Fund. 4. New York191 The Public Utility Law Project of New York sends annual personalized letters to all persons eligible for Lifeline, informing them about the program. 5. North Carolina192 In North Carolina, an ad hoc committee comprised of staff members from the North Carolina Utilities Commission, the Attorney General’s Office, major telecommunications industries, and social services organizations have made major strides since 1998 in their Lifeline/Link-Up outreach efforts with direct mailings and other forms of outreach. Since the committee’s first meeting, 200,000 brochures have been printed and distributed to various organizations across the state that works with low-income families. The North Carolina Public Service Commission sent notices to everyone in North Carolina who was eligible for the programs. 6. Tennessee193 The Tennessee Regulatory Authority (TRA) works with the Tennessee Department of Human Services (TDHS) database to determine eligible individuals and then mails Lifeline/Link-Up information to those people. H. Lifeline/Link-Up Notification on Every Call 1. Maine194 Maine’s public assistance agencies explain the Lifeline/Link-Up program whenever a household applies for public assistance and the state’s telephone companies mention Lifeline/Link-Up whenever a customer applies for telephone service. This way, a household can apply for Lifeline/Link-Up by phone by simply stating that they receive one of the listed public benefits and providing either a social security number or welfare identification number. Maine credits its high penetration rates to this combination of innovative outreach and easy application methods. I. Tax Break for Lifeline/Link-Up Telephone Companies 1. North Carolina195 191 See USAC Comments at 12. 192 See Civil Rights Forum Comments at 4-5. 193 See Civil Rights Forum Comments at 5. 194 See NCLC Comments at 7. 195 See North Carolina Utilities Commission Comments at 4-5. 43 FCC 03J-2 Federal Communications Commission North Carolina provides for a tax break to Lifeline/Link-Up telephone companies equal to the amount of money they are required to contribute for Lifeline/Link-Up. According to FCC data, Lifeline enrollment in North Carolina increased from 29,640 in 1998 to 62,475 in 2000. J. Lifeline/Link-Up Marketing Board California196 1. California created a Lifeline Marketing Board which promotes the Lifeline program beyond the typical telephone company policy of including information in their telephone bills. 196 See Civil Rights Forum Comments at 4-5. 44 FCC 03J-2 Federal Communications Commission Appendix F Lifeline Staff Analysis Quantifying the effects of adding an income criterion to the Lifeline eligibility criteria A Study for the Federal-State Joint Board on Universal Service Prepared by Craig Stroup Industry Analysis & Technology Division Wireline Competition Bureau Federal Communications Commission FCC 03J-2 FCC 03J-2 Federal Communications Commission Executive Summary Lifeline Staff Analysis April 2003 New Lifeline households = New Lifeline-eligible households times predicted Lifeline subscription rate among newly-eligible households. Additional federal Lifeline expenditures = number of additional households that would take Lifeline times the amount of federal expenditures per household that takes Lifeline. Introduction The Federal-State Joint Board on Universal Service (Joint Board) recommends that the Federal Communications Commission (FCC) add a federal default income-based criterion of at least 1.35 times the Federal Poverty Guidelines . a 1.35 Poverty Level Criterion (PLC). This would allow many additional low-income citizens in those states that utilize the federal default criteria to take the Lifeline program. The Joint Board also recommends that the FCC encourage all states to adopt the recommended federal income-based criteria. There is a benefit to increasing the number of participants, and also a cost. The obvious benefit would be the increase in the number of telephone subscribers. The cost at the federal level would be the additional federal dollars spent on the additional Lifeline enrollees. Methodology This study uses the economic method of forecasting baseline, change and new policy impact. This means that first we estimate the number of Lifeline subscribers and the costs of the program to form the baseline, also known as the status quo. Second, we estimate the changes that would result from a nationwide implementation of a 1.35 PLC, assuming that all states adopt this criterion.1 Third, we add (or apply) the changes to the baseline to the time period when the policy is expected to be implemented. This step provides an estimate of the number of Lifeline subscribers and costs under the new policy. We have chosen to estimate the baseline and changes for 2004 because that is the timeframe in which the proposed changes will likely be made. This study uses a combination of statistical regression analysis and simple math in a series of spreadsheet tables. The following equations form the basic structure of the spreadsheet model. 1 Some states have a 1.5 PLC. This study assumes that those states with a 1.5 PLC keep it. FCC 03J-2 Federal Communications Commission In sum, the results of two regression models are used to predict the impact of a policy change, and these predictions are applied to the baseline to calculate the new level of Lifeline subscription and federal Lifeline expenditures. Results The results are summarized below: Summary information for 2004 if states adopted a 1.35 PLC: Additional households that would take Lifeline: 967,000 to 1,136,000 Of the additional Lifeline subscribers, the number that would subscribe to telephone service because of the 1.35 PLC: 259,000 Of the additional Lifeline subscribers, the number that would already have telephone service: 708,000 to 877,000 Additional federal expenditures in 2004: Amount that federal expenditures would increase $105,000,000 to $123,000,000 Additional federal expenditures per new telephone subscriber: $405 to $475 2 FCC 03J-2 Federal Communications Commission Lifeline Staff Analysis Introduction States use different criteria for determining whether a household qualifies for Lifeline. Some states use the federal eligibility criteria (set by the FCC), which enable households receiving Federal Public Housing Assistance (Section 8), Food Stamps, Low-Income Home Energy Assistance Program, Medicaid, or Supplemental Security Income to receive Lifeline. Other states have set their own criteria. States setting their own criteria often use one or more of the programs from the federal criteria and sometimes include one or more of their own state-wide programs. Some states also use an income-based criterion, which is based on some multiple of the Federal Poverty Guidelines. In all cases, a household need meet only one of a state’s criteria to be eligible for Lifeline. The Joint Board recommends that the FCC add an income-based criterion to the federal eligibility criteria for Lifeline. The Joint Board also recommends that the income-based criterion be set at 1.35 times the Federal Poverty Guidelines. Thus, households with incomes at or below 1.35 times the Federal Poverty Guidelines would be eligible for Lifeline. This study assumes that all states (not just those that currently utilize the federal default criteria) add an income-based criterion of at least 1.35 times the Federal Poverty Guidelines (poverty level).a 1.35 Poverty Level Criterion (PLC).which would increase the overall number of eligible households.2 This would enable additional low-income citizens in many states to take the Lifeline program. (Households meeting at least one eligibility criterion are eligible for Lifeline, so adding an additional eligibility criterion increases the number of households that are eligible for Lifeline.) There is a benefit to increasing the number of participants, and also a cost. The obvious benefit would be the increase in the number of telephone subscribers. The cost at a federal level would be the additional federal dollars spent on the additional Lifeline enrollees. Because the study assumes that all states choose to adopt the recommended federal income-based eligibility criteria, the estimates presented are likely to represent the upper limit of potential new Lifeline and telephone subscribers and estimated impact on the fund. If some states choose not to adopt the federal income-based standard, the number of new Lifeline and telephone subscribers, and additional cost would be correspondingly lower. The relationship between Lifeline eligibility, Lifeline subscribership, and telephone subscribership is as follows. A portion of newly-eligible households (because of a 1.35 PLC) will take Lifeline service. Of those households that subscribe to Lifeline because of the 1.35 PLC, a portion will start taking telephone service because they would then qualify for Lifeline. The other portion would already have telephone service, and would be taking the Lifeline just because they are newly-eligible. See the graphs below. 2 This study assumes throughout that states with a 1.5 PLC continue to use a 1.5PLC. 3 17% take Lifeline Households that are newlyeligible for Lifeline service due to 83% do not take Lifeline a 1.35 PLC Federal Communications Commission Lifeline Eligibility with a 1.35 PLC, households taking Lifeline, and households taking telephone service due to a 1.35 PLC Currently Eligible for Lifeline Newly-eligible for Lifeline due to 1.35 PLC Eligible under current criteria or 1.35 PLC 25% of households that take Lifeline service were phoneless 75% of households that take Lifeline service already had phone service 4 Households that take Lifeline service due to a 1.35 PLC FCC 03J-2 FCC 03J-2 Federal Communications Commission Methodology Summary This study uses the economic method of forecasting baseline, change and new policy impact. This means that first we estimate the number of Lifeline subscribers and the federal expenditures of the program to form the baseline numbers. Second, we estimate the changes that would result from a nationwide implementation of a 1.35 PLC. Third, we add (or apply) the changes to the baseline in the time period when the policy is expected to be implemented. This step provides an estimate of the number of Lifeline subscribers and costs under the new policy. For the first step, we estimate Lifeline subscribership in Year 2000 and update those estimates using data for Year 2002. The 2002 estimates are used as a base from which to forecast 2004 baseline Lifeline subscribership. We have chosen to estimate the baseline and changes for 2004 because that is the timeframe in which the proposed changes will be made. For the second step, the Year 2000 subscribership estimates are used to predict the change in Lifeline subscribership due to a 1.35 PLC. The study uses the plethora of demographic data available from the Year 2000 to model the effects that a 1.35 PLC would have had on Lifeline subscribership and telephone penetration in 2000. For Lifeline subscribership, a regression model is constructed that predicts the increase in Lifeline subscribers as a function of increasing multiples of the Federal Poverty Guidelines. For example, the model predicts that if Texas— which has a 1.25 PLC—adopted a 1.35 PLC, Lifeline subscribers in 2004 would increase by 16,669 to 19,576 (See Table 2.F). For telephone subscribership, a logistic regression is constructed that predicts the increase in telephone subscribership as a function of increasing multiples of the Federal Poverty Guidelines and other important factors, such as income and home ownership. If all states adopt a 1.35 (or higher) PLC for Lifeline, the model predicts that 259,000 households would take telephone service because of that change. In the third step, the estimated additional number of Lifeline subscribers is added to the baseline in Year 2004 to get the forecasted number of Lifeline subscribers that would exist in 2004 under a nationwide implementation of the new policy. This study forecasts the additional Lifeline subscribers that would result from the implementation of a 1.35 PLC (baseline plus change). These steps are exhibited in the following graphs. The first graph shows the steps for predicting the number of Lifeline subscribers, and the second graph shows the amount of federal Lifeline expenditures. 5 9 8 7 6 5 4 3 2 1 0 Millions of Lifeline subscribers Millions of dollars $1,000 $900 $800 $700 $600 $500 $400 $300 $200 $100 $0 A (Baseline) 1998 Federal Lifeline Expenditures A (Baseline) 1998 E (Policy change) B (Baseline) 2000 Baseline Change due to 1.35 PLC 6 C (Baseline) D (Baseline) 2002 Federal Communications Commission Lifeline Subscribers E (Policy change) B (Baseline) C (Baseline) 2000 Baseline Change due to 1.35 PLC 2002 F (New level with change) D (Baseline) F (New level with change) FCC 03J-2 2004 2004 FCC 03J-2 Federal Communications Commission Modeling Process The modeling process is outlined below. The word “produce” is used below when the FCC did not have the actual data, and so the quantities were estimated based on a sound methodology. The word “forecast” is used when data are predicted for a future time period. • Create baseline o Produce baseline Lifeline subscription rates for 2000. o Produce baseline Lifeline subscription rates for 2002. o Forecast baseline Lifeline subscription rates for 2004. o Forecast baseline federal Lifeline expenditures for 2004. • Estimate change from new policy o Produce change to Lifeline eligibility resulting from a 1.35 PLC. o Forecast change to Lifeline subscription rates in 2000 resulting from a 1.35 PLC. o Forecast change to Lifeline subscription rates for 2004. o Forecast for Years 2000 and 2004, change to telephone subscribership resulting from a 1.35 PLC. o Forecast change to federal Lifeline expenditures for 2004. • Apply new policy to baseline to compute new level o Apply forecasted changes to forecasted baseline to determine the new number of Lifeline subscribers in 2004. o Apply forecasted changes to forecasted baseline to determine the new federal Lifeline expenditures in 2004. Methodology Detail The above steps will now be discussed in more detail. A series of tables is constructed that show the computations for the three steps outlined above. This study combines data from three sources: 1) Current Population Survey of Households (CPSH) provided by the Bureau of Labor Statistics; 2) the website www.lifelinesupport.org; and 3) Universal Service Administrative Company (USAC). The CPSH data contain the results from over 50,000 households that were surveyed around January 2000. The website www.lifelinesupport.org provides the Lifeline eligibility requirements for each state, and USAC’s website provides actual Lifeline subscribers in 2000 and 2002. This study uses a combination of statistical regression analysis and simple math in a series of spreadsheet tables. Two regression models are constructed. • Lifeline Regression Model - A regression analysis model is constructed that correlates higher Lifeline subscription rates to the use of higher multiples of the Federal Poverty Guidelines for income criteria. Many states have income-based Lifeline eligibility criteria, and in general, the states with a higher multiple of the Federal Poverty Guidelines have higher Lifeline subscription rates. The results from this model are then used to predict the number of households that would take Lifeline in 2000 and 2004 as a result of a nationwide implementation of a 1.35 PLC. 7 FCC 03J-2 Federal Communications Commission • Telephone Regression Model - Another regression model, this time a logistic regression, is used to predict increased telephone participation that would have resulted in 2000 had a 1.35 PLC been implemented. This model incorporates several factors, including the 1.35 PLC, income, and other demographic information. Many states have income-based Lifeline eligibility criteria, and in general, the states with a higher multiple of the Federal Poverty Guidelines have higher telephone subscription rates. The results from this model are then used to determine the number of households that would take telephone service in 2004 as a result of a nationwide implementation of a 1.35 PLC. The spreadsheet tables use a series of equations which simply add or multiply the contents of various columns in the table to produce a final column (to the right) which is of the most interest. The results of the regression analysis are incorporated into several columns in the tables. The following equations are used in the tables: • Number of additional households taking Lifeline = number of newly-eligible households times the Lifeline subscription rate (the percentage of those households that would take Lifeline, which is determined by the Lifeline Regression Model). • Additional federal Lifeline expenditures = number of additional households that would take Lifeline times the amount of federal expenditures per household that takes Lifeline. In sum, the results of two regression models are used to predict the impact of a policy change, and these predictions are applied to the baseline to calculate the new level. The data and analysis is discussed in more detail below. 8 FCC 03J-2 Federal Communications Commission Step 1: Create Baseline The tables in this section examine the number of Lifeline subscribers, the number of households that are eligible for Lifeline and the Lifeline subscription rate. Each table reflects data for a different year. Baseline Lifeline subscription rates for Year 2000. Nationally, 16.3% of households are estimated to be eligible for Lifeline. Of these eligible households, an estimated 33.1% subscribe to Lifeline. The CPSH data contain demographic data from which the eligibility for each household in the sample can be determined. So, if a state uses Food Stamps as an eligibility criterion, then those households in that state that received Food Stamps are marked as being eligible for Lifeline. Each household is analyzed according to its state’s eligibility criteria, as reported by www.lifelinesupport.org.3 Only those households that meet at least one of the eligibility criteria are deemed eligible for Lifeline, the rest are deemed ineligible. This is accomplished electronically using Visual Basic for Applications for Microsoft Access. From these data, statewide estimates for the number of Lifeline eligible households are created. USAC data from the Monitoring Report are then used to create the Lifeline subscription rate, which is the percentage of eligible households that subscribe to Lifeline. See Table 1.A. Baseline Lifeline subscription rates for 2002. Nationally, 16.3% of households are estimated to be eligible for Lifeline. Of these households, an estimated 37.5% subscribe to Lifeline. USAC Lifeline data from 2002 are used to create a new baseline subscribership rate, using the same methodology as for Year 2000 described above. The number of households in each state in Year 2002 is forecasted based on the growth rate of households between 1998 and 2000. It is assumed that the same percentage of households that qualified for Lifeline in 2000 qualified for Lifeline in 2002. See Table 1.B. Forecasted Baseline Lifeline subscription rates for 2004. There will be an estimated 110.1 million households in the Year 2004, and 6.8 million of those households are expected to take Lifeline under existing rules. The results from the previous tables are used to forecast the number of households, the number of Lifeline-eligible households, and the number of Lifeline subscribers in 2004. The number of households in 2004 is calculated in the same manner as it was in Table 1.B. The number of households qualifying for Lifeline in 2004 (July 1, 2004, to be exact) is simply calculated by multiplying the percentage of all households that are eligible for Lifeline in 2000 by the forecasted number of households in 2004. This calculation assumes that the same percentage of households will qualify for Lifeline in 2004 as did in 2000. The number of households that would take Lifeline in 2004 is calculated by multiplying the percentage of eligible households that took Lifeline in 2002 by the forecasted number of eligible households in 2004. This 3 The website was viewed in early 2002. 9 FCC 03J-2 Federal Communications Commission This section quantifies the number of additional households that would become eligible calculation assumes that the same percentage of Lifeline-eligible households will take Lifeline in 2004 as did in 2002. These predictions make two implicit assumptions: the number of households in each state increases at a constant rate, and the economy continues to grow at the same rate it did in 2002. See Table 1.C. Forecasted Baseline federal Lifeline expenditures for 2004. Forecasted federal Lifeline expenditures under existing rules in Year 2004 are $709 million. The forecasted federal Lifeline expenditures are calculated by multiplying the forecasted number of Lifeline subscribers in each state times the expected federal expenditures per line in that state. The state-by-state federal expenditures are then summed to form the national total. See Table 1.D. Step 2: Estimate Change from New Policy for Lifeline, the number of households that would subscribe to Lifeline, and the number of additional households that would subscribe to telephone service due to the implementation of a 1.35 PLC. (This analysis assumes that states without a PLC for Lifeline and states with a PLC below 1.35 adopt a 1.35 PLC. This analysis also assumes that states with a 1.5 PLC keep it.) This section then calculates the increased federal Lifeline expenditures resulting from the increased number of households taking Lifeline due to the 1.35 PLC. CPSH data are used to determine the number of additional households that would become eligible for Lifeline. Two regression analyses are used to determine the number of additional households that would subscribe to Lifeline and the number of households that would take telephone service due to a 1.35 PLC. Change to Lifeline eligibility in 2000 and 2004 resulting from a 1.35 PLC. We predict that an additional 6.1 percent of total households would qualify for Lifeline under the 1.35 PLC, and this would qualify an additional 6.6 million households in Year 2004. The demographic data from each household in the CPSH data are examined to determine whether it would have become eligible for Lifeline with a 1.35 PLC. The estimates from the CPSH data are then used to determine the number of households in each state that would become eligible for Lifeline with a 1.35 PLC. Table 2.A presents the information for the Year 2000 and 2.B presents the information for the Year 2004. Change to Lifeline subscribership in 2000 resulting from a 1.35 PLC. We predict that states without a PLC and states with PLCs at 1.25 or lower would see a significant increase in the number of low-income households that take Lifeline if they adopted a 1.35 PLC. Nationwide, the number of Lifeline takers would increase between 928,000 to 1,090,000 if all states adopted a 1.35 PLC. Different states have different Lifeline eligibility criteria, so regression analysis can be employed to quantify the correlation between the use of a higher multiple of the poverty level (i.e., a higher PLC) and the resulting higher Lifeline subscription rate. The Lifeline Regression Model predicts 10 FCC 03J-2 Federal Communications Commission The new levels of subscribership and costs are shown in several tables. First, the new increased Lifeline subscribership that would have resulted from a nationwide 1.35 PLC in 2000. See Tables 2.C and 2.D. (At the end of this study is a technical appendix that more thoroughly discusses the regression analysis used for this model.) Tables 2.E and 2.F show the number of additional Lifeline subscribers on a state-by-state basis for 2000 and 2004. Forecasted change to telephone subscribership for 2004. We predict that if all states adopted a 1.35 PLC, 259,000 households that do not have telephone service would take telephone service. The Telephone Regression Model uses logistic regression to predict the increased telephone subscribership that would have resulted from a nationwide 1.35 PLC in 2000. See Tables 2.G and 2.H. (At the end of this study is a technical appendix that more thoroughly discusses the logistic regression analysis used for this model.) Table 2.H also uses these results to quantify the number of households that would take telephone service in 2000 and 2004 because of a 1.35 PLC. For 2000 and 2004 respectively, Tables 2.I and 2.J. break down the number of new Lifeline subscribers into two groups: those that would be taking telephone service because of the 1.35 PLC, and those that are already had telephone service, and who are subscribing to Lifeline just because they would then be eligible for it. Change to federal Lifeline expenditures for 2004 is forecasted. We predict that federal Lifeline expenditures would increase by $105 million to $123 million if all states implemented a 1.35 PLC. The forecasted change to federal Lifeline expenditures is calculated by multiplying the forecasted change to the number of Lifeline subscribers in each state times the expected federal expenditures per Lifeline subscribers in that state. The state-by-state change in the amount of federal expenditures is then summed to form the national total. See Table 2.K. Step 3: Apply New Policy to Baseline to Compute New Level total of Lifeline subscribers is calculated, and then the increased federal Lifeline expenditures are calculated. Forecasted New Policy Levels for Lifeline subscribership in 2004. We predict that if all states implement a 1.35 PLC for Lifeline, an estimated 8 million households will subscribe. Here the forecasted increase in Lifeline subscribers is added to the forecasted baseline number of subscribers to create the new forecasted number of Lifeline subscribers in 2004 with the 1.35 PLC. See Table 3.A. Forecasted New Policy Levels for federal Lifeline expenditures. We predict that if all states implement a 1.35 PLC for Lifeline, federal Lifeline expenditures are forecasted to be in the range of $814 million to $832 million. 11 FCC 03J-2 Federal Communications Commission Here the forecasted increase in federal Lifeline expenditures is added to the forecasted baseline federal Lifeline expenditures to create the new forecasted federal Lifeline expenditures in 2004 with the 1.35 PLC. See Table 3.B. Additional request Finally, this study examines, at the Joint Board’s request, the effects of replacing the current federal default Lifeline eligibility criteria with a single income-based criterion (Table 4.A). For administrative ease, the model assumes that all states (even those that do not presently utilize the federal default criteria) would adopt a single criterion of 1.35 PLC, except that states with a 1.5 PLC would keep it. Therefore, these estimates may overstate the results of the policy change. If current criteria were replaced with a 1.35 PLC, then some current Lifeline participants would no longer be eligible, so there would be decreases in Lifeline subscribership resulting from the discontinued criteria. There would also be offsetting increases from the new 1.35 PLC. The net impact is that fewer households would take Lifeline if the 1.35 PLC were the only eligibility criterion. The calculations are as follows. The baseline number of households taking Lifeline is the same as calculated above in Section Three. CPSH data are examined to determine the percentage of households that would no longer qualify for Lifeline due to the removal of all other eligibility criteria. The number of newly-eligible households that would take Lifeline as a result of the 1.35 PLC criteria change is derived in Section Three. Thus, the new policy level of Lifeline subscribers is the baseline number of Lifeline subscribers less those subscribers that could not remain due to the change, plus those Lifeline subscribers that would take it because of the change. See Table 4.A. Other Factors This study cannot take several important factors into consideration, such as economic conditions and state outreach programs because there are not enough data to do so. Properly accounting for a fluctuating economy would require five or more decades of data. The Lifeline program started only about 20 years ago, so an analysis incorporating a fluctuating economy is not attempted in this study. Further, there are no comprehensive estimates quantifying state spending on outreach programs, or the effects the outreach programs have on Lifeline subscribership. By not accounting for these factors explicitly, this study assumes that these factors will remain constant between 2000 and 2004. Although changes in these factors can affect the forecasted baseline number of Lifeline subscribers (and therefore, baseline federal expenditures), those factors should have a relatively smaller effect on the forecasted number of households that will take Lifeline as a result of a 1.35 PLC. The number of households that would take Lifeline because of a 1.35 PLC is about 1/6th of those that already take Lifeline. So, as the economy fluctuates, and more or less households take Lifeline, the number of households that would take Lifeline due to a 1.35 PLC will go up and down by 1/6th as much as the number of households that would take Lifeline based on other eligibility criteria. Thus, the number of households 12 FCC 03J-2 Federal Communications Commission taking Lifeline due to a 1.35 PLC will have 1/36th the variance that the number of households taking Lifeline will have.4 Additional assumptions In addition to the factors discussed above, this study makes several assumptions that are needed to estimate the impact of the program: 1) All other Lifeline/Linkup eligibility criteria (and the qualifications for the underlying programs) stay constant over time. Aside from the addition of a 1.35 PLC, this model assumes that between 2000 and 2004, no other changes are made to the Lifeline/Linkup programs or to the programs that are frequently used as qualifying criteria for Lifeline between 2000 and 2004; 2) Data can be substituted. Several states have a 1.33 PLC in effect. This study treats states that have a 1.33 PLC as having a 1.35 PLC. This assumption is reasonable because the effects of a 1.33 PLC are statistically indistinguishable from a 1.35 PLC. 3) Rapid adoption and continuity. This model assumes that all states rapidly adopt a 1.35 PLC (and that states with a 1.5 PLC keep it). The model also assumes that households rapidly learn of the changes to the Lifeline program and expeditiously act on this new information. 4 See Henry Scheffe, The Analysis of Variance, at 8 (1959). 13 FCC 03J-2 Federal Communications Commission Results The results are summarized below: Summary information for 2004: Household information: Forecasted households on Lifeline without 1.35 PLC: 6,827,000 Forecasted additional households on Lifeline with 1.35 PLC: 967,000 to 1,136,000 Forecasted households on Lifeline with 1.35 PLC: 7,974,000 to 7,961,000 Lifeline subscriber information: Households that would take telephone service due to the 1.35 PLC: 259,000 Households taking Lifeline that already have telephone service: 708,000 to 877,000 Federal Lifeline expenditures: Forecasted federal Lifeline expenditures without 1.35 PLC: $709,000,000 Forecasted amount federal expenditures would increase: $105,000,000 to $123,000,000 Forecasted federal Lifeline expenditures with 1.35 PLC: $814,000,000 to $832,000,000 Additional federal expenditures per new telephone subscriber: $405 to $475 14 FCC 03J-2 Federal Communications Commission 15 Federal Communications Commission 16 FCC 03J-2 FCC 03J-2 Federal Communications Commission 17 FCC 03J-2 Federal Communications Commission 18 Federal Communications Commission 19 FCC 03J-2 Federal Communications Commission 20 FCC 03J-2 FCC 03J-2 Federal Communications Commission 21 FCC 03J-2 Federal Communications Commission 22 FCC 03J-2 Federal Communications Commission 23 FCC 03J-2 Federal Communications Commission 24 FCC 03J-2 Federal Communications Commission 25 FCC 03J-2 Federal Communications Commission 26 FCC 03J-2 Federal Communications Commission 27 FCC 03J-2 Federal Communications Commission 28 Federal Communications Commission 29 FCC 03J-2 Federal Communications Commission 30 FCC 03J-2 FCC 03J-2 Federal Communications Commission 31 Federal Communications Commission 32 FCC 03J-2 FCC 03J-2 Federal Communications Commission Technical Appendix 1 Background information for Table 2.C (Would Lifeline take rates increase due to a nationwide implementation of a 1.35 PLC?) Below are the two regression results that are used to determine the effect that a nationwide implementation of a 1.35 poverty level criterion would have on Lifeline subscribership. Regression 1 – Lifeline specification 1. The regression model calculated from the data is %HHBelowOnePtFiveTakingLifeline = -0.02 + 0.58 x IncElgAbv125 + 1.04 x California + 0.0167 x TotSup. Explanation of variables for Lifeline regression specification 1. The dependent variable is the number of households taking Lifeline divided by the number of households that are at or below 1.5 times the poverty level1 (%HHBelowOnePtFiveTakingLifeline). For example, Texas had 263,934 Lifeline subscribers in 2000, and 1,575,172 households at or below 1.5 times the poverty line. The dependent variable data point for Texas therefore equals 0.15 (=263,934/1,348,089). The first Independent Variable is IncEligAbv125. For each state, IncEligAbv125 equals that state’s income eligibility level (if it has one) minus 1.25. So, for California, which has an income eligibility criterion of 1.5 times the poverty level, IncEligAbv125 equals 0.25 (= 1.5 – 1.25). For states with an income eligibility criterion at or below 1.25 times the poverty level, or for states without an income criterion, IncEligAbv125 equals 0. So, for Texas, which has an income eligibility criterion of 1.25 times the poverty level, IncEligAbv125 equals 0. The coefficient on this variable allows us to predict the number of households that would take Lifeline if a 1.35 PLC were adopted. 1 The federal government establishes the poverty level threshold, which is based on the number of people living in the household, and whether the household is in the mainland United States, Alaska, or Hawaii. The Current Population Survey of Households (CPSH) data conveniently list the poverty level for each family in the family record portion of the data. 33 FCC 03J-2 Federal Communications Commission So for Texas, and other states with a 1.25 PLC (and for states without an incomebased criterion), the new policy would increase the independent variable from 0.25 to 0.35, or by 0.1, and the dependent variable would increase 5.8 percentage points. The percentage point increase in percentage of households at or below 1.5 times the poverty level that take Lifeline because of a 1.35 PLC were implemented would be 5.8%. = 0.58 * 0.1 = 0.058 or 5.8%. 2 The second Independent Variable is “California”. In statistical terms, this is called a “dummy” variable, and equals 1 if the state is California, and is 0 otherwise. A dummy variable is often used in regression analysis to quantify specific effects. California is the only state using self-certification with an income-based criterion, and it appears to have more households taking Lifeline than the CPSH data would indicate are eligible for it. Therefore, singling out California with a dummy variable to measure a California-specific effect is warranted. The variable “TotSup” is the amount of monthly telephone service support that Lifeline subscribers in each state receive (TotSup). The amount of total support that households receive varies with the local telephone carrier. For each state, TotSup is the amount of support from the largest carrier in that state. For example, in Texas, Lifeline subscribers in Southwestern Bell territories pay $11.35 per month less for telephone service than regular telephone subscribers. Therefore, the TotSup datapoint for Texas is $11.35. The more support that eligible households can receive, the more incentive they have to take Lifeline. 2 The coefficient 0.58 is used to calculate the number of additional households that would take Lifeline with a 1.35 PLC. It is multiplied by the number of households at or below 1.5 times the poverty level (i.e., from 0.0 to 1.5 times the poverty level). Even though those households between 1.35 and 1.5 times the poverty level would not actually qualify for Lifeline, the model coefficient is estimated in such a way that a correct prediction is made. 34 FCC 03J-2 Federal Communications Commission Regression 2 – Lifeline specification 2. %HHBelowOnePtFiveTakingLifeline = 0.14 + 0.68 x IncElgAbv125 + 1.04 x California When comparing the two specifications, this one suggests that more households would take Lifeline because the coefficient 0.68 is greater than the 0.58 coefficient in Regression 1. So for Texas, and other states with a 1.25 PLC, and for states without an income criterion, the percentage point increase in the percentage of households at or below 1.5 times the poverty level that would take Lifeline because of a 1.35 PLC is 6.8%. = 0.68 *x 0.1 = 0.068 or 6.8%. Additional information about Lifeline regression specifications 1 and 2: Data sources The data are from the Current Population Survey of Households (CPSH) (March 2000 data), the Universal Service Monitoring Report (Oct. 2002) and www.lifelinesupport.org. The CPSH data are used to determine the number of households at or below 1.5 times the poverty level in each state. The Universal Service Monitoring Report was used to determine the number of households on Lifeline and the total support (number of dollars) that Lifeline subscribers received in each state. The website www.lifelinesupport.org was used to determine which states had income criteria for Lifeline, and the multiple of the Federal Poverty Guidelines that was required to be eligible for Lifeline in those states. Data are aggregated to the state level. CPSH has data for thousands of households, including whether the household has telephone service or not. If it were possible to do so, it would be best to conduct the analysis at the household level to maximize the number of observations and to account for several demographic factors. Unfortunately, CPSH data do not report whether the household is receiving the Lifeline subsidy. Therefore, individual data observations could not directly be used for the estimation. These regressions 35 FCC 03J-2 Federal Communications Commission therefore use data that have been aggregated to the state level. This means there is a single data point constructed for each state. The number of Lifeline subscribers for each state is available from the Universal Service Monitoring Report, however, so the CPSH data are aggregated to the state level. The number of households that are at or below 1.5 times the poverty level in a particular state is determined by summing the statistical weight of each household at or below 1.5 times the poverty level (the statistical weight for each household is determined by the Bureau of Labor Statistics), and dividing by 100. (The statistical weights add up to 100 times the number of households in the state, so dividing by 100 is a necessary step.) Additional information on regression specification The dependent variable: %HHBelowOnePtFiveTakingLifeline. As mentioned above, the dependent variable is the number of households taking Lifeline divided by the number of households that are at or below 1.5 times the poverty level. The dependent variable should be a measure of participation rate, and this requires a measure of takers and a measure of eligibility. An ideal measure would have been the number of households taking Lifeline divided by the total number of households that are eligible. Obtaining a measure of number of eligible households in each state is not possible, as will be explained below, so a surrogate measure “number of households that are at or below 1.5 times the poverty level” is used in its place. As long as the resulting surrogate participation rate is consistent across states, and used properly, the resulting analysis is correct. The surrogate is necessary because of a measurement problem. There are several states where it is difficult to measure the number of households that are eligible for Lifeline. This happens most often when states use state-specific programs as eligibility criteria. Because the CPSH survey does not ask about every possible welfare program, the CPSH data cannot always be used to determine if a household is eligible for Lifeline or not. 36 FCC 03J-2 Federal Communications Commission Therefore, an alternative dependent variable was needed. The number of households below 1.5 times the poverty level is a reasonable proximate measure of support need. So, instead of dividing the number of households taking Lifeline by the number of households eligible for Lifeline, the dependent variable in this analysis is the number of households taking Lifeline divided by the number of households that are at or below 1.5 times the poverty level. The 1.5 times the poverty level threshold was chosen because it was the highest poverty level criterion used by any state, and it was used by several states. The principal independent variable: IncEligAbv125 As mentioned above, IncEligAbv125 equals that state’s income eligibility level (if it has one) minus 1.25. If the state has no income eligibility criterion, or if it has one that is less than 1.25 times the poverty level, then the datapoint equals zero for that state. The main objective of the regression analysis is to quantify the number of additional households that will subscribe to Lifeline with the implementation of an income-based eligibility criterion. Generally, states using higher multiples of the poverty level as an eligibility criterion have higher Lifeline participation rates than states using lower multiples of the poverty level criteria (or states using no income based criterion at all). The coefficient on IncEligAbv125 is used to predict the number of households that would take Lifeline due to a 1.35 PLC. Preliminary modeling indicated that a nationwide implementation of an income criterion set at or below 1.25 times the poverty level would not increase the number of households taking Lifeline by a statistically significant amount. Because some states use lower multiples of the poverty level to determine Lifeline eligibility, one would expect that using a higher multiple of the poverty level would increase the number of households eligible for Lifeline in those states. However, basing this independent variable on lower multiples of the poverty level did not produce statistically significant results. 37 FCC 03J-2 Federal Communications Commission Discussion Discussion of independent variables: “California” is significant in both regressions (indeed, it was significant for all regression specifications in which it is included). “TotSup” is positive, but is not significant. It is nearly significant, however. Further, there is strong economic reason to include it, because it measures a household’s incentive to take Lifeline, so it should not be eliminated from the model without good reason. “IncEligAbv125” is significant in both regressions, but the size of the coefficient varies, and it is just barely significant (at the 10% level) when TotSup is included. Other specifications of the model were run that included whether each state had a particular program as an eligibility criteria. Throughout most of the trial specifications, the coefficient of IncEligAbv125 ranged between the two values presented in this report and remained significant. Therefore, the analyses presented in this report are very robust. Low-income Home Energy Assistance Program (LIHEAP) Other trial variables are tested in the regression analysis, but for the reasons listed below, these trials are not adopted. However, when the regression included whether the state had energy assistance as a method for qualifying for Lifeline, the coefficient on IncEligAbv125 dropped 40% and was not even close to being significant. This trial regression model is contrary to sound economics for two reasons. First, if the results were accurate, it would indicate that there would be no significant additional Lifeline subscribership with the implementation of a 1.35 PLC. This is not plausible, because the logistic regression analysis (see Appendix 2) indicates that a 1.35 PLC would significantly increase the number of households taking telephone service. Because we find strong evidence that a 1.35 PLC would increase telephone subscribership, a similar impact on Lifeline subscribership is also expected. Second, if the coefficient on IncEligAbv125 from the Lifeline Regression were plugged into the model, it would indicate that just 10% of those households that 38 FCC 03J-2 Federal Communications Commission would become eligible would take Lifeline service, which seems far too low. Currently, over 30% of eligible households take Lifeline service. While the percentage of eligible households that would take Lifeline would surely decrease as eligibility requirements were eased, there is no reason to believe that it would drop by more than 2/3. Thus, adding a variable quantifying whether the state has energy assistance as an eligibility requirement leads to results that are not consistent with economic theory. That trial regression is therefore not used.3 Given that the coefficient on IncEligAbv125 ranges between 0.582 and 0.682 in all the other trial regressions, that range is used in this study. Table 2.D uses the results from the regression analysis to quantify the number of households that would take Lifeline as a result of a 1.35 PLC. OLS regression was used using the statistical computer program Stata 7.0. The regression outputs (below) show the significance of each coefficient. 3 We note that there is some multicollinearity between the energy assistance variable and TotSup. As a practical matter, if energy assistance is included in the regression and TotSup is removed, then the coefficient on IncElgAbv125 returns to normal levels and is significant. 39 FCC 03J-2 Federal Communications Commission ============== Regression output ================== . reg %HHBelowOnePtFiveTakingLifeline IncElgAbv125 California TotSup Source | SS df MS Number of obs = 51 -------------+------------------------------ F( 3, 47) = 19.64 Model | 1.5914109 3 .530470301 Prob > F = 0.0000 Residual | 1.26974251 47 .027015798 R-squared = 0.5562 -------------+------------------------------ Adj R-squared = 0.5279 Total | 2.86115341 50 .057223068 Root MSE = .16436 ------------------------------------------------------------------------------ | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- IncElgAbv125 | .5815073 .3422222 1.70 0.096 -.106955 1.26997 California | 1.040881 .1825073 5.70 0.000 .6737233 1.408038 TotSup | .0166981 .0102551 1.63 0.110 -.0039326 .0373288 Constant | -.0220947 .1013846 -0.22 0.828 -.2260543 .1818648 reg %HHBelowOnePtFiveTakingLifeline IncElgAbv125 California Source | SS df MS Number of obs = 51 -------------+------------------------------ F( 2, 48) = 27.19 Model | 1.51978515 2 .759892577 Prob > F = 0.0000 Residual | 1.34136826 48 .027945172 R-squared = 0.5312 -------------+------------------------------ Adj R-squared = 0.5116 Total | 2.86115341 50 .057223068 Root MSE = .16717 ------------------------------------------------------------------------------ | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- IncElgAbv125 | .682112 .3423391 1.99 0.052 -.006207 1.370431 California | 1.045145 .1856009 5.63 0.000 .6719696 1.418321 Constant | .1380132 .0251194 5.49 0.000 .0875073 .1885192 ------------------------------------------------------------------------------ 40 FCC 03J-2 Federal Communications Commission Technical Appendix 2 Background information for Table 2.G (Would a 1.35 PLC for Lifeline increase telephone penetration?) Below are the results of two logistic regressions. They show the effects that a 1.35 PLC for Lifeline has on telephone subscribership. Logistic regression 1 was used for the study. Logistic regression 2 tested whether California’s self-certification process for income-based eligibility increased telephone penetration among lowincome households. Logistic regression 1 — Telephone Specification 1: Y = 1 / (1 + e – [1.09 + 0.189*X1 + -.753* X2 + .728* X3 + .521* X4 + .032* X5 +- 0.326*X6]) Explanation of variables for Telephone Specification 1. Dependent variable: Does the household have telephone service? (Y = H_TELHHD) The dependent variable is whether the low-income household has telephone service. The data point for a household equals one if the household has telephone service, and equals zero otherwise. The dataset is comprised of data from only those households with incomes at or below 1.5 times the poverty level. Independent variables: Is the household in a state with a 1.35 or less restrictive poverty level criterion? (X1 = SH135BET) If the household is in a state that uses a 1.35 PLC for Lifeline (or if the state uses a higher multiple of the poverty level), then SH135BET equals one for that data point; otherwise, it equals zero. Because the sample is restricted to only those households that are at or below 1.35 times the poverty level, all data points for this 41 FCC 03J-2 Federal Communications Commission variable will be either a “0” or “1”. Of these low-income households, 19.1 percent live in a state with a 1.35 to 1.5 PLC, and the independent variable SH135BET equals 1 for these households. For the other 80.9 percent, the independent variable SH135BET value equals 0. This is the only independent variable used in the cost/benefit analysis, and therefore the accuracy of its coefficient is of most concern. The coefficient on this variable (0.189) is later used to quantify the increased probability that a lowincome household will take telephone service (or fraction of) as the result of a 1.35 PLC. This quantification is accomplished as follows: When X1 is changed, Y will change. For an individual household, the change of X1 from 0 to 1 models the effect of implementing a 1.35 PLC for that particular household. When modeling the change nationally, X1 is changed from .191 (19.1%, which reflects the fact that 19.1 percent of the sample households already live in a state with a 1.35 PLC) to 1. As a result, Y changes according to Logistic regression 1 above (Y is interpreted as a percentage—or probability—of households with telephone subscribership, and ranges from 0 to 1). When we change the “baseline” 19.1 percent of low-income households (living in a state with a 1.35 PLC) to the “new policy” 100.0 percent, then predicted telephone subscribership among sample households increases from 88.3 percent to 89.8 percent. Is the household a mobile home? (X2 = MOBHOME) If the household is a mobile home, then the MOBHOME equals one for that datapoint; otherwise, it equals zero. Is the household owned by the householders? (X3 = OWNHOME) If the householders own the home themselves, then OWNHOME for that data point equals 1; otherwise, it equals zero. Percentage of households who lived at that address for at least one year. (X4 = PCTONEYEAR) 42 FCC 03J-2 Federal Communications Commission The data points for PCTONEYEAR equal the percentage of the adults in that household that have lived at that address for at least one year. Total value of household income (X5 = HTOTVAL) The data points for each household equal the household’s entire annual income including the value of transfer (e.g., welfare) payments. Is someone in the household on food stamps? (X6 = HFOODSP) If someone in the household is on food stamps, then HFOODSP equals one for that data point; otherwise, it equals zero. For the results of this specification, see page 51, below. Logistic regression 2 — Telephone Specification 2: Telephone Specification 2 includes all the variables from specification 1, and includes the variable California. California. (CALIF) If the household is in California, the variable equals one, otherwise, it equals zero. For the results of this specification, see page 52, below. Additional information about specifications 1 and 2 Price None of the logistic regression specifications include the price of telephone service. This is because the price that each household faces is unknown. Different carriers offer service at different prices, and even within the same carrier, the price of telephone service varies from city to city. Because the carrier that would serve each household is unknown, price cannot be included in the logistic regressions. Earlier research has shown that omitting the price of telephone service does not affect the coefficients of the other variables in this logistic regression. This is 43 FCC 03J-2 Federal Communications Commission because the coefficient on price would be tiny, so any “missing variable” bias would also be tiny.4 Data sources The data in this analysis are from the Current Population Survey of Households (CPSH) from March 2000. CPSH data contain information on over 50,000 households. From these data, the relevant demographic information are extracted for analysis, including: 1) whether the household has telephone service, 2) household’s total income (including the value of transfer payments), 3) the poverty level for that household (i.e., household earnings divided by state definition of poverty-level income), 4) the state the household lives in, 5) whether the household dwelling is owned or rented, 6) the number of adult members that live in the household for at least one year, 7) the number of adults living in the household, and 8) the list of subsidies the household receives, which included Federal Public Housing Assistance (Section 8), Food Stamps, LIHEAP, Medicaid, and Supplemental Security Income. The CPSH data also includes information on whether or not the household has telephone service. Household-level data are used All the information is available for each household, so the analysis is conducted at the household level; aggregating to the state level is unnecessary. Logistic regression preferred to “standard” OLS regression Because the dependent variable is binary (a household either has telephone service and is thereby assigned a values of one (1), or it does not and is thereby assigned a value of 0 (zero), logistic regression analysis is preferred to a Linear Probability model using Ordinary Least Squares (OLS). With binary dependent variables, 4 The formula for calculating the missing variable bias can be found in many textbooks, including William H. Greene, Econometric Analysis, at 402 (3rd ed. 1997). Observation of the equation shows that if the missing variable is uncorrelated with an independent variable, then the coefficient on that independent variable is unbiased. A regression was run to see if telephone prices are correlated with the variable SH135BET. The weighted average price for each of the 41 states for which price data are available was created. The variable price was then regressed on the variable SH135BET. There was no correlation. (See Industry Analysis and Technology Division, Wireline Competition Bureau, Federal Communications Commission, Reference Book, at 7-8 (2002). 44 FCC 03J-2 Federal Communications Commission linear regressions can produce erroneous results, such as a household having more than a 100% probability of taking telephone service, or a household having a negative probability of taking telephone service. Both of these situations are impossible. Logistic regression analysis avoids this problem, and is appropriate for measuring saturation concepts such as telephone penetration. The following graph illustrates the difference between the two approaches. In the following graph (taken from the Internet), “linear probability model” refers to OLS regression results, and Y (ranging from 0 to 1) refers to probability. 5 Unfortunately, logistic regressions produce coefficients that are more difficult to interpret than the coefficients that OLS produces. A few additional computations are needed to use the coefficients in the cost-benefit analysis. Therefore, Table 2.H is created, which uses the coefficients from the logistic regression to determine the number of households that would have taken phone service in 2000 and 2004 if a 1.35 poverty level criterion were instituted nationally. The number of households that would take telephone service because of a 1.35 PLC is then compared to the number of households that would take Lifeline in Table 2.I. 5 For more information on logistic regression analysis, see Damodar Gujarati, Basic Econometrics at 481-491 (2nd ed. 1998). 45 FCC 03J-2 Federal Communications Commission Quantifying logistic regression coefficients In a standard regression analysis, the effect that a change in the independent variable has on the dependent variable is relatively easy to measure because it is linear. When using standard linear regression, a model is often expressed as follows: Y = a + b*X. In this equation, Y represents the dependent variable, ¡°a¡± represents a constant, and ¡°b¡± is the coefficient from the regression which is multiplied by the size of the independent variable X. The symbol . is often used to represent the change in a variable. The change in Y caused by a change in X is then represented like this: .Y = b*.X. Thus, the change in Y for a change in an independent variable is simply the coefficient on the independent variable times the amount of the change in that independent variable. Because logistic regression analysis is not linear, however, the above calculation cannot be made directly. Instead, two intermediate calculations must be made. The first calculation quantifies the dependent variable using the mean values of the independent variables. The second calculation quantifies the dependent variable using the same means as in the first calculation, except that one of the independent variables is set to the new policy level. The second calculation replaces the mean of the independent of the variable in question (e.g., a policy variable) with an appropriate value representing the change in the variable. If all states adopted a 1.35 PLC, then the percentage of low income households living in a state with a 1.35 PLC would move from 19.1% to 100%. So, in this case, the mean of SH135BET (which equals 0.191) would be replaced with 1.00. For both calculations, Y is calculated by the following equation: Y = 1 / (1 + e .[1.09 + 0.189*X1 + -.753* X2 + .728* X3 + .521* X4 + .032* X5 +- 0.326*X6 ]) Table 2.H explains the calculations. The coefficient values from the logistic regression are in column a. The means of the independent variables are in column b. Column c multiplies columns a and b. These products are often called the ¡°partial effects¡±. The partial effects are then summed to create a Z score. The Z score is simply a shorthand way of representing a +b1*x1 + b2*x2 + ¡¦. When evaluating the independent variables at their mean values, the Z score equals 2.025. Y (the probability that a household will take telephone service) is then calculated: Y = 1/(1+e-z), which equals 88.3%. This means that, nationwide, households with 46 FCC 03J-2 Federal Communications Commission incomes below 1.35 times the poverty level have an 88.3% chance of having telephone service. The second calculation is identical to the first, with one exception. Instead of using the mean value of SH135BET, the mean is replaced by a 1. As discussed above, this would be the case if all states have a 1.35 PLC. Just as before, the coefficients (column a) are multiplied by the means (column d) to produce the new partial effect. Notice that for SH135BET, the mean value of 0.191 was replaced with 1.00. The new partial effects are listed in column E. These partial effects are then summed to form the new Z score, which equals 2.178. This new Z score is then used in the calculation as before: Y= =1/(1+e-z). The new value for Y is 89.8%. This means that if all states adopted a 1.35 PLC, then 89.8% of households with incomes at or below 1.35 times the poverty level would have telephone service. This represents a 1.5 percentage point increase (89.8% - 88.3%) in telephone subscription rates. To determine the number of households in 2004 that would take phone service due to a 1.35 PLC, the difference in the Y’s (1.5%) is multiplied by the number of households that are at or below 1.35 times the poverty level. Projections made using the CPSH data indicate that in 2004, there will be 17,433,000 households at or below 1.35 times the poverty level. Thus, multiplying 1.5% (which equals 0.015) times 17,433,000 households equals 259,000 households. Thus, 259,000 households would take telephone service due to a 1.35 PLC in 2004. Restricted use of observations and variables The logistic regression analyses uses only selected observations and variables for good reason. One reason is to address a specific policy proposal from the Joint Board. The Joint Board is recommending using a 1.35 PLC. In order to determine how such a plan would affect households at or below 1.35 times the poverty level, only those households with incomes at or below 1.35 times the poverty level are included in this analysis.6 There are 8,358 usable observations. 6 Alternatively, the sample could be restricted to households at or below 1.33 times the poverty level because there are three states that have a 1.33 PLC. By including households at 1.34 and 1.35 times the poverty level, we are implicitly assuming that those households are eligible for Lifeline even though they just miss qualifying for it. On the other hand, restricting the sample to households at or below 1.33 times the poverty line would exclude many more households from the sample in other states with a 1.5 PLC. It is not clear whether a 1.33 PLC restriction is better than a 1.35 PLC. Fortunately, the results are the same in either case. For both models, the coefficient on SH135BET is virtually identical with either sample restriction. 47 FCC 03J-2 Federal Communications Commission The number of state specific variables that can be included in the analysis is limited because only 8 states have SH135BET equal to one. Therefore, including additional state specific variables reduces the accuracy of the coefficient SH135BET, the important policy variable used to quantify costs and benefits. Discussion of variables in the specifications Assumption that effects of a 1.33 PLC are indistinguishable from a 1.35 PLC As mentioned earlier, this study assumes that the effects of a 1.33 PLC are statistically indistinguishable from a 1.35 PLC. Therefore, SH135BET equals one for the states that have 1.33 or 1.5 PLCs. There is no alternative to measuring the effect of a 1.35 PLC because no states use a 1.35 PLC. Further, the fact that this analysis treats states with a 1.5 PLC the same as states with a 1.33 PLC is not problematic. This is because the households in the sample are restricted to those that are at or below 1.35 times the poverty level. Thus, all the households in the sample will make the same economic choice whether the state in which they live has a 1.33 (or 1.35) or 1.50 PLC, because the households qualify for Lifeline under either criterion. Inclusion of independent variables HFOODSP was included because it captures the concept of “poverty” in a way that income alone does not. Participation in the Food Stamps Program is an indicator of special household needs. Without a variable like HFOODSP to capture poverty in a way that income alone does not, the coefficient on SH135BET is negative and insignificant, which is counter to a reasonable economic theory of Lifeline effects. CALIFORNIA-Unique Effects. The CALIF (California) variable was tested as a separate variable in the second logistic regression because it was included in the Lifeline Model. The hypothesis is that California’s policy of using self-certification for income-eligibility could possibly have a unique impact on telephone subscribership that is different than other states. Just as California was singled out in the Lifeline subscribership regressions, one might reason that the unique policy of California should also be 48 FCC 03J-2 Federal Communications Commission reflected in the telephone subscribership analysis. The second logistic regression examines the effects of accounting for California separately. The results indicate that living in California does not have a unique effect on telephone subscribership. The second specification shows that the coefficient on CALIF is not significant, which suggests that California’s self-certification policy does not statistically significantly increase telephone subscribership among Californians (compared to other states) with incomes at or below 1.35 times the poverty level.7 The inclusion of the variable California in the logistic regression has a large erroneous impact on the primary variable of interest, SH135BET (whether or not the household is in a state with at least a 1.35 poverty level criterion). If the logistic regression includes the variable California, then the coefficient on SH135BET is smaller and not statistically significant.8 If the variable California is not included in the logistic regression, then the coefficient on SH135BET is larger and statistically significant, as expected. This larger SH135BET coefficient is found because the Lifeline program has a somewhat larger impact on low-income households in California than in other states. Furthermore, including a CALIF variable would compromise statistical accuracy. Including the CALIF variable would lower the statistical accuracy of the income criterion effect. Half of all households that live in a state with at least a 1.35 poverty level criterion for Lifeline are in California, so accounting for California separately would wrongly remove any influence California observations have on the “national” coefficient for the variable SH135BET. The influence from California observations should be included in the coefficient for SH135BET, and so the 2nd model excluding the California influence (by including a CALIF variable) is not used. 7 Because California has above-expected Lifeline subscribership, one might expect it to have above-expected telephone subscribership among households at or below 1.35 times the poverty level. However, the data does not support this. When responding to the CPSH survey, households have no incentive to misreport their income, so those households in California that report their income as being below the 1.35 times the poverty line most likely really are below that threshold. The result is that California telephone penetration follows that of the other states. 8 Although the coefficient on SH135BET is still positive, it is not statistically significant. If SH135BET is not statistically significant, then it would be difficult to conclude that states having a 1.35 PLC (or less restrictive poverty level criterion) have any impact on telephone penetration. 49 FCC 03J-2 Federal Communications Commission Because there is no compelling reason to account for California separately, and because the coefficient on the variable California is not significant, households in California are not singled out in the analysis by including a separate CALIF variable. Thus, the California variable should not be included in the logistic regression. Total Lifeline support The variable Total Lifeline support for the household is not included in the final model for two reasons. (See discussion of “TotSup” from Technical Appendix 1.) First, the total support that individuals within a state receive depends on the carrier that would potentially serve them. Thus, although the amount of total support from the largest carrier in the state was chosen, there would be a large number of households for which the variable “TotSup” would contain the wrong amount of support. For the majority of households in the CPSH data, the location of the household is unidentifiable, so the carrier that would potentially serve that household is also unidentifiable. Second, when the variable “TotSup” was tried in the logistic regression, it proved not significant. When “TotSup” was included, the coefficient on SH135BET was smaller, but was still significant. The logistic regression was run using the statistical computer program SPSS version 10. The regression analysis computer printouts are displayed below: 50 Percentage Correct .0 100.0 87.1 FCC 03J-2 Exp(B) .471 2.072 1.684 1.208 1.000 .722 2.978 Federal Communications Commission Percent 100.0 .0 100.0 .0 100.0 df Sig. .000 .000 .000 6 6 6 Nagelkerke R Square .064 H_TELHHD .00 .00 1.00 0 0 Predicted 1.00 1079 7279 df 1 1 1 1 1 1 1 S.E. .109 .081 .077 .089 .000 .072 .086 Classification Tablea Variables in the Equation Wald 47.273 81.442 45.929 4.523 30.847 20.325 160.887 MOBHOME OWNHOME PCTONEYR SH135BET HTOTVAL HFOODSP Constant a. Variable(s) entered on step 1: MOBHOME, OWNHOME, PCTONEYR, SH135BET, HTOTVAL, HFOODSP. Logistic Regression Case Processing Summary Unweighted Casesa Selected Cases Included in Analysis Missing Cases Total Unselected Cases Total N 8358 0 8358 0 8358 a. If weight is in effect, see classification table for the total number of cases. Omnibus Tests of Model Coefficients Step 1 Step Block Model Chi-square 291.862 291.862 291.862 Model Summary Cox & Snell R Square .034 Step 1 -2 Log likelihood 6138.251 Step 1 Observed H_TELHHD Overall Percentage a. The cut value is .500 Step 1a B -.752823 .728299 .521155 .189162 .000032 -.326141 1.091223 Sig. .000 .000 .000 .033 .000 .000 .000 51 Percentage Correct .0 100.0 87.1 FCC 03J-2 Exp(B) .473 2.084 1.678 1.087 1.000 .724 1.249 2.989 Federal Communications Commission Percent 100.0 .0 100.0 .0 100.0 df Sig. .000 .000 .000 7 7 7 Nagelkerke R Square .064 H_TELHHD .00 .00 1.00 0 0 Predicted 1.00 1079 7279 df 1 1 1 1 1 1 1 1 S.E. .110 .081 .077 .116 .000 .072 .162 .086 Classification Tablea Variables in the Equation Wald 46.727 82.599 45.218 .520 29.676 19.905 1.887 161.649 MOBHOME OWNHOME PCTONEYR SH135BET HTOTVAL HFOODSP CALIF Constant a. Variable(s) entered on step 1: MOBHOME, OWNHOME, PCTONEYR, SH135BET, HTOTVAL, HFOODSP, CALIF. Logistic Regression Case Processing Summary Unweighted Casesa Selected Cases Included in Analysis Missing Cases Total Unselected Cases Total N 8358 0 8358 0 8358 a. If weight is in effect, see classification table for the total number of cases. Omnibus Tests of Model Coefficients Step 1 Step Block Model Chi-square 293.757 293.757 293.757 Model Summary Cox & Snell R Square .035 Step 1 -2 Log likelihood 6136.356 Step 1 Observed H_TELHHD Overall Percentage a. The cut value is .500 Step 1a B -.748590 .734328 .517551 .083355 .000032 -.322910 .222716 1.095058 Sig. .000 .000 .000 .471 .000 .000 .170 .000 52 FCC 03J-2 Federal Communications Commission SEPARATE STATEMENT OF CHAIRMAN BOB ROWE, MONTANA PUBLIC SERVICE COMMISSION I am pleased that the Joint Board gave this referral the same close attention it has Re: Federal-State Joint Board on Universal Service, Recommended Decision on Low-Income Programs. This inquiry developed a wealth of good ideas: Good ideas to increase awareness of Lifeline and Link-up; to better match eligibility requirements with need; to increase participation; and to lower transaction costs while preserving accountability. Ultimately, all of these ideas are intended to ensure that the programs better achieve Congress’s goals for them. In very many instances, the Joint Board recommends that this compendium of good ideas be used by the states to tailor programs most appropriate to their specific circumstances. This is very much a prudential, “cooperative federalist” approach to achieving the programs’ purposes. It encourages state creativity. To succeed, it will require greater effort and engagement from many states, including my own. Specifically, it will require close coordination between state public utility commissions, state and local human services agencies, the industry, and other stakeholders. Consistent with cooperative federalism, I hope this recommendation will also stimulate a multi-directional dialogue, with states sharing successful strategies, and reporting back through some efficient medium on their implementation of this recommendation in ways that will provide useful information to the FCC and to others interested. afforded the other important issues with which it has recently dealt. FCC 03J-2 Federal Communications Commission SEPARATE STATEMENT OF COMMISSIONER LILA A. JABER, FLORIDA PUBLIC SERVICE COMMISSION Re: Federal-State Joint Board on Universal Service, Recommended Decision on Low-Income Programs. An important aspect of the Federal Communications Commission’s (FCC) mission is to ensure that telecommunications services are available to “all the people” of the United States. To that end, the Low-Income Program has been designed to assist eligible economically disadvantaged households that want, but cannot afford, telephone service by discounting services provided by local telephone companies. I believe that this recommended decision, if implemented, will improve the effectiveness of the program by addressing issues relating to sustainability and accountability. I wish to thank my colleagues on this Joint Board for a balanced and well-reasoned recommended decision. I am optimistic that this recommended decision will ensure that those customers that need assistance will be eligible to receive it by expanding the list of federal eligibility criteria; I support their inclusion. The long-term sustainability of the program requires effective accountability. Several states have taken such steps to ensure program integrity by utilizing automated enrollment procedures both to add eligible households and to remove them when they no longer qualify. I am pleased that this recommended decision has been used to highlight successful strategies that states may consider implementing to improve participation in the program. I am especially encouraged by the recommended decision’s proposal that would require states to establish a verification plan. While I have doubts about the use of selfcertification as a means of verification, I trust that the flexibility recommended for state implementation will successfully root out any waste, fraud, and abuse that may exist in the program. FCC 03J-2 Federal Communications Commission SEPARATE STATEMENT OF BILLY JACK GREGG, DIRECTOR OF THE CONSUMER ADVOCATE DIVISION, PUBLIC SERVICE COMMISSION OF WEST VIRGINIA Re: Federal-State Joint Board on Universal Service, Recommended Decision on Low-Income Programs. According to the Commission’s latest report on telephone subscribership, 95.1% of the 109 million households in the United States have telephone service.1 This is a remarkable achievement, but it still falls short of the goal of universal availability and affordability of service set forth in the 1996 Telecommunications Act. The fact that 95.1% of homes are connected to the telecommunications network means that over 5 million households in our country do not have telephone service. Moreover, this number has remained persistent. Since 1990, the overall number of households and the percentage of households with telephone service have grown, while the number of households without telephone service has continued to range between 4.8 and 6.4 million.2 The Joint Board and the Commission took action in 1997 to address the large number of unconnected households in our nation by expanding the federal Lifeline and Link-Up programs. Since that time 7.4 million households have been added to the telephone network and the percentage of households without phone service has dropped.3 Unfortunately, in spite of these efforts and the efforts of the individual States, the number of households without phone service remains high. Poverty is obviously the primary factor limiting the ability of unconnected households to join the telephone network. Low-income customers are significantly less likely to have telephone service than are other consumers.4 The federal Lifeline and Link-Up programs provide numerous options to low-income individuals and families to overcome the cost of obtaining and maintaining phone service. The Link-Up program will pay the lion’s share of local connection charges and provides for the waiver of all deposit requirements if a customer opts for toll-blocking service. Once a household is connected to the phone system, the Lifeline program provides substantial federal discounts off of normal monthly recurring charges, and encourages states to add discounts of their own. In some cases, these discounts can represent 90% of a regular phone bill. However, federal and state programs to assist in the payment of phone bills are of no use if a low-income customer cannot get phone service because of an outstanding balance for unpaid local and/or long distance service. 1 Telephone Subscribership in the United States, FCC Wireline Competition Bureau, IAD (Feb. 12, 2003), Table 1, http://www.fcc.gov/Bureaus/Common_Carrier/Reports/FCC-State_Link/IAD/subs0702.pdf. 2 The number of households without telephone service last topped 6 million in November 2000 and has only dipped below 5 million once, in March 2002. Id., Table 1. 3 Id., Table 1. 4 The telephone penetration rate in households with annual incomes below $5,000 is 78.9%, rising to 99.3% in households with annual incomes above $75,000. Id., Table 4. FCC 03J-2 Federal Communications Commission I believe a large number of the 5 to 6 million households that do not currently have phone service, do indeed want phone service and can afford the discounted Lifeline monthly charges on a going-forward basis. However, these customers cannot be connected to the network because they have previously had phone service, lost it for non-payment of local and/or long distance charges, and cannot afford payment of the unpaid balance. In short, the outstanding balances from previous phone service for these low-income customers stand as a barrier to these customers reconnecting to the telephone network. In taking further action on modifying the Lifeline and Link-Up programs, I urge the Commission to solicit data from interested parties to document the number of customers that remain disconnected because of prior balances, and the number of qualifying Lifeline and Link- Up customers who are precluded from obtaining service because of outstanding balances for local and/or long distance service. The Commission should also investigate whether changes can be made to the Link-Up program to address these prior balances for local and/or long distance service. 5 Such changes could include reconnection upon agreement by the qualifying customer to pay off the outstanding balances over a period of months – for example, six months or twelve months – in equal monthly payments. In return, the customer would be provided with Lifeline service with mandatory toll blocking until the past due balance was paid off. The Commission could also invite comment on whether it would be appropriate for the Link-Up program to pay a set percentage of the past outstanding balances for local service, and whether such payments should be contingent on state matching payments. I applaud the work of the States, the Commission and the Joint Board in attempting to make the Lifeline and Link-Up programs more effective. I sincerely hope that the Recommended Decision which we issue today will move these efforts forward. However, we must never lose sight of the fact that our goal is to connect the unconnected and to keep phone service affordable for everyone. We must continue to search out and eliminate programmatic and structural impediments to greater participation in the telecommunications network by all of our citizens. I believe expanding Lifeline and Link-Up assistance to address the issue of past balances will go a long way toward eliminating a major hurdle faced by low-income customers in attempting to become full participants in our globally connected society. 5 I recognize that the Fifth Circuit has previously held that a rule prohibiting disconnections of local service for nonpayment of long distance bills exceeded the Commission’s jurisdiction, absent additional justification. Texas Office of Public Utility Counsel v. FCC, 183 F.3d 393, 424 (5th Cir. 1999). The issue I raise now is the different but related issue of whether the Commission may properly design a program to assist in reconnecting low-income customers to the network. Such a program could involve partnering with States or providing inducements to the States to reconnect such customers.