IV. STAFF'S ANALYSIS AND FINDINGS

In its investigation, Staff analyzed 1) the results of AT&T's and Verizon's research to determine whether OSP's billings match their call records; 2) the declaration of OSP's billing aggregator attesting that OSP's call records were fictitious; and 3) OSP's revenues and excessive refund amounts (compared to its revenues) issued to California customers.

As discussed in FTC v. Inc21.com Corp, supra,⁶² third-party billing by local exchange carriers (LECs) is "fraud-friendly" and could lead to deceptive billing of tens of thousands of businesses and consumers via their telephone bills. As mentioned earlier, with Mr. Vogel's (OSP's sole officer) knowledge and familiarity with the industry's operations and practices, Staff suspects he took advantage of the third-party billing system to defraud California consumers. Some consumers may not scrutinize the bills

received from their own telephone companies and thereby miss noticing the unauthorized OSP charges. Staff believes that even though OSP made refunds to the consumers who complained, OSP stands to profit from many others who do not detect the unauthorized charges and thus, do not complain about them. As can be seen in Table 4 below, although OSP refunded well over $2 million, it still received approximately 65% of the over $8 million it generated in billing revenues.

Based on the analysis detailed below, none of OSP's collect call charges proved legitimate. Staff believes OSP crammed more than 250,000⁶³ California consumers and unlawfully collected over $8 million⁶⁴ from California consumers.

A. the lack of supporting call records from telephone companies calls into question the existence of osp's collect calls

Telecommunications companies utilize a standard format called Exchange Message Interface (EMI)⁶⁵ to exchange messages or information for billing purposes, among other things. Third parties (i.e., OSP) - who would like LECs (i.e., AT&T and Verizon) to bill and collect for them - have to use the EMI standard format to provide their customer billing information. Here, for billing purposes, first OSP had to provide the customer billing information (i.e., EMI "records") to its billing aggregator,⁶⁶ TBR, who generally ensured the required data fields were properly filled in. TBR would then forward the EMI "records" to the appropriate LECs.

Also, for billing purposes, the telecommunications companies' end office switch⁶⁷ automatically collects customer's phone call information. These call records are called Automatic Messaging Accounting (AMA) or switch records.

Both EMI and AMA records have numerous data fields that contain, among other things, the following information:⁶⁸

· Connect date and time

· Disconnect time

· Time zone

· Originating country and number

· Terminating country and number

· Dialed number

· Elapsed Minutes

· Called code

· Answer Indicator

A legitimate EMI record should have a matching AMA call record maintained by the LEC terminating the call. Therefore, a reliable way to verify whether OSP's collect calls in fact occurred is to check whether AT&T and Verizon have AMA call records on OSP's billings, as their switches would have automatically captured the call information for the collect calls OSP purportedly serviced. As discussed below, both AT&T and Verizon could not locate the accompanying AMA call records for the sample of EMI billing "records" Staff randomly selected. From this finding, Staff infers that none of the EMI "records" for which OSP based its collect call billings was valid.

1. AT&T call records do not support OSP's collect call charges.

As part of its review process, Staff requested TBR submit all OSP-related EMI "records" that served as TBR's basis for billing between October 9, 2008 and August 31, 2009.⁶⁹ In response, TBR provided an electronic file containing 252,694 EMI "records" covering the period October 9, 2008 through April 18, 2009.⁷⁰ These EMI "records" represent only TBR's records, excluding Integretel's records. (Staff is awaiting reply from TBR regarding Integretel's EMI "records" for OSP's billings.)

Staff asked TBR whether each EMI "record" represents one specific consumer. In response to Staff's inquiry, TBR clarified that more than one call could have been billed to a single California consumer and the total number of records does not necessarily equal to the total number of California consumers billed.⁷¹ Still, Staff believes it is likely OSP billed more than 250,000 California consumers during its 21 months of billing and collection operations. It is because the 252,694 EMI "records" represent only 7 months' of billing (from October 2008 through April 2009) and it is reasonable to assume that OSP illegally billed other consumers during its earlier 14 months of operations (from August 2007 through September 2008). This is because of the numerous complaints discussed above that resulted from OSP billings during that earlier period. Moreover, the 252,694 EMI "records" represent only 38% of OSP's over $8 million total billed revenues, compared to 62% that OSP collected during its earlier 14 months of operations.

From these EMI "records," Staff randomly selected a sample of 384 records⁷² (which represents a 95% confidence level),⁷³ that Staff initially forwarded to AT&T to verify the corresponding AMA records. According to AT&T, its analysis of the 384 EMI "records" indicated Verizon would have provided dial tone associated with 91 EMI "records" (in other words, they are EMI "records" related to Verizon customers' telephone numbers), and AT&T would have provided dial tone (or AT&T's customers) associated with 293 EMI "records."⁷⁴ Utilizing a search of switch records (AMA) associated with collect calls based solely on the date and time of each call identified in OSP's EMI "records," AT&T was unable to identify switch records (AMA) for the 293 EMI "records" that OSP submitted.⁷⁵

In response to Staff's inquiry, using the 252,694 EMI "records," AT&T reviewed the high volume originating telephone numbers (i.e., top 100)⁷⁶ from its network and again could not locate AMA records for these numbers. In some cases, the telephone line was not in service at the time of the calls.⁷⁷ This result substantiates Staff's claim that OSP fabricated its call records because there is no way a collect call could be made if the telephone line is not in service.

With regard to the thirteen complainants who provided declarations (see Section III above), Staff requested that AT&T provide their billing statements that contained an OSP charge and the corresponding AMA record for the charge made by OSP. AT&T provided the billing statements for only twelve complainants. AT&T found no billing statements or AMA record for one of the thirteen complainants.⁷⁸ (Complainant has three telephone numbers. AT&T provides service to two of the telephone numbers and Verizon provides service to the 3^rd telephone number, which is a wireless line. Staff is awaiting reply from Verizon.) As to the corresponding AMA records, AT&T found no switch records (AMA) associated with the OSP charges that appeared on the billing statements of the twelve declarants. Again, Staff believes this demonstrates that these customers did not receive or authorize the collect call charges.

2. Verizon call records do not support OSP's collect call charges

Staff requested Verizon verify its AMA records for 91 Verizon telephone numbers out of the 384 randomly selected EMI "records."⁷⁹ According to Verizon, it attempted to locate Verizon-recorded terminating usage data for these calls and it conducted three different sets of matching criteria. For the first search, it attempted to match four attributes, (To number, Date, Time, and From number), and found no matching AMA call records. For the second search, it attempted to match three attributes, (To number, Date, and Time), and found no matching AMA call records. For the third search, it attempted to match two attributes, (To number and Date), and found 72 matching AMA call records; however, the "Time" and "From Number" attributes substantially differed even with time zone adjustment based on the originating point of the call.⁸⁰

In addition, Staff requested Verizon recheck the 91 Verizon telephone numbers using OSP's access numbers (877-487-9455 and 866-697-2198) as the originating numbers, in case these numbers may have been signaled as the originating point rather than the "From Number" from the EMI "records." According to Verizon, it found no changes from its earlier response.⁸¹ Since the "connect time" in the EMI "records" is based on Pacific Time, Verizon stated that the EMI "records" would match the local switch time in any of its terminating AMA records.⁸² Using a search that attempted to match three attributes, (Date, Time and Terminating Number), Verizon did not find any match in its records. From the lack of validating AMA records for any of the sample of OSP's EMI "records," Staff infers that none of AT&T's nor Verizon's customers received (and therefore could not have authorized) the collect calls for which OSP billed them.

3. Approximately all of OSP's billing transactions for collect calls are invalid.

As discussed above, all of the 384 EMI "records" randomly selected from billings submitted by OSP do not have supporting AMA call records from AT&T and Verizon. This suggests with a 95% confidence level that the collect calls did not occur. Thus, extrapolating the sample to the total population, Staff concludes with 95% confidence that OSP fabricated approximately all of the 252,694 EMI "records" used to bill California consumers.

B. complainants claim they neither received nor authorized osp's collect calls

As detailed earlier in Section III, consumers claim they did not receive or authorize the collect calls because OSP charged their phone line dedicated to a DVR⁸³ or TIVO,⁸⁴ or a fax machine with no handset plugged into it, or they were at work. Moreover, Staff observed that the CAB/FTC/internet complainants raised only one issue or concern: OSP's collect calls were not received or authorized. Once more, Staff believes this demonstrates OSP fabricated its call records and defrauded consumers.

C. osp experienced excessively high refund rates

As part of its investigation, Staff also reviewed the billing revenues and refunds associated with OSP's California billings provided by TBR. The following Table 4 shows that OSP experienced an approximate 30% refund rate from August 2007 to March 2010. Further, adding the $544,000 in refunds made by the LECs brings the refund rate to approximately 35%, for a total of nearly $2.9 million in refunds. (See Section V.)

Table 4

OSP Communications

Revenues, Refunds and Refund Rate

((2007-2009)⁸⁵

**

**Source: The Billing Resource Response dated May 10, 2010.

In 2009 alone, OSP experienced an abnormally high annual refund rate of 53%. With regard to OSP's monthly refund rates - except for the month of August 2007 - it ranged from 14% to 43% (see Table 4 above). According to TBR, operator service providers, such as OSP, should only experience a 4% to 6% refund rate; a refund rate beyond this range raises the suspicion of fraudulent activities.⁸⁶ (See Declaration of Nelson Gross in Section V.)

Based on the discussion above regarding the absence of supporting call records from AT&T and Verizon, OSP's significantly high refund rate, and the discussion below regarding TBR's declaration that its sampling of OSP's collect calls proved fictitious (i.e., fabricated), Staff suspects that none of the over $8 million in billings OSP generated are legitimate because the collect calls apparently never happened. Therefore, the remaining approximate 65% of revenue collected by OSP should be returned to California customers whom OSP appeared to have defrauded.

D. osp began operating a similar collect call scheme shortly after link systems ceased operations because of cpsd's investigation.

As discussed and illustrated in Section II.C. above, Mr. Vogel held key positions as Link's president, vice president, CFO and director. It appears that he also acted as Link's sole officer, similar to OSP. In addition, as with OSP, California consumers filed complaints against Link for charging them for collect calls, which they neither received nor authorized.⁸⁷ When CPSD's investigation led to Link ceasing its operations in 2007, OSP shortly thereafter began providing what appears to be the same collect call service. This time John Vogel utilized different billing agents. CPSD believes that OSP simply took over where Link left off. Approximately a month after OSP started operating, its billing agent began receiving complaints of unauthorized collect call charges. Consumers also began complaining about OSP's collect call charges to CAB. See Table 2.

E. osp failed to provide a legible contract that details the precise services performed by ekc, its supposed underlying carrier.

As mentioned earlier (Section II.B.), Mr. Vogel - acting as the sole officer and employee of OSP Communications - indicated that EKC served as OSP's underlying carrier providing dial tone. Although Mr. Vogel submitted a copy of its contract with EKC in response to Staff's request, he provided an illegible copy that prevented Staff from confirming its exact relationship with EKC. Furthermore, Staff gave Mr. Vogel several opportunities to produce a legible copy but he failed to provide one.⁸⁸ Therefore,

Staff suspects EKC never provided any service to OSP and they never had any valid contract. Once again, this confirms Staff's belief that none of the collect calls billed by OSP in fact occurred.

⁶² FTC v. Inc21.com Corp., supra,745 F. Supp. 2d 975, 982 (N.D. Cal. 2010).

⁶³ Attachment 25C (TBR CD Response dated June 15, 2010 to CPSD Data Request No. TEL-277-002 dated May 11, 2010).

⁶⁴ Attachment 9C (TBR Response dated May 10, 2010 to CPSD Data Request No. TEL-277-001 dated April 1, 2010).

⁶⁵ Formerly called Exchange Message Record (EMR). The EMI adheres to certain standards established by the Alliance for Telecommunications Industry Solutions (ATIS) Ordering and Billing Forum (OBF). Full copies of the EMI standards are available through ATIS at http://www.atis.org

⁶⁶ Attachment 37 (OSP Response dated August 15, 2010 to CPSD Data Request OSP TEL-277-002 dated August 13, 2010).

⁶⁷ An end office switch is a telephone central office switch that connects directly to the customer. Most of the call recording and billing are performed in end office switches. ( http://pcmag.com/encylopedia).

⁶⁸ Attachment 26C (AT&T e-mail Response dated March 4, 2011).

⁶⁹ Attachment 25C (CPSD Data Request No. TEL-277-002 dated May 11, 2011).

⁷⁰ Ibid. (TBR Response dated June 15, 2011; TBR provided a CD).

⁷¹ Attachment 5C (TBR e-mail Response dated April 1, 2011).

⁷² Attachment 27 (Printout of 384 EMI records and method employed by Staff to randomly select the sample).

⁷³ A confidence level is a measure of the reliability of a result. A confidence level of 95 per cent or 0.95 means that there is a probability of at least 95 per cent that the result is reliable. ( http://www.thefreedictionary.com/confidence+level)

⁷⁴ Attachment 28C (AT&T Response dated September 10, 2010 to CPSD Data Request No.TEL-277-002 dated June 30, 2010).

⁷⁵ Id.

⁷⁶ Ibid. (See AT&T Response dated September 10, 2010).

⁷⁷ Id.

⁷⁸ Id.

⁷⁹ Attachment 29C (CPSD Data Request No. TEL-277-002 dated September 20, 2010).

⁸⁰ Ibid. (Verizon Response dated October 7, 2010).

⁸¹ Ibid. (Verizon Response dated October 25, 2010).

⁸² Id.

⁸³ A DVR stands for "Digital Video Recorder." A DVR is basically a VCR that uses a hard drive instead of video tapes. It can be used to record, save and play back television programs. Unlike a VCR, however, a DVR can also pause live TV by recording the current show in real time. The user can choose to fast forward (often during commercials) to return to live television. (Source: http://www.iwebtool.com/what_is_dvr.hmtl)

⁸⁴ TiVo is a digital video recorder developed and marketed by TiVo, Inc. TiVo provides an electronic television programming schedule, whose features include "Season Pass" schedules which record every new episode of a series, and "Wish List" searches which allow the user to find and record shows that match their interests by title, actor, director, category or keyword. (Source: http://en.wikipedia.org/wiki/TiVo)

⁸⁵ The grand total refund amount of $** ** includes $** ** total refunds in 2010 (January, February and March).

⁸⁶ Attachment 30 (Declaration of Nelson Gross).

⁸⁷ Attachment 15 (Complaint Data from CAB about Link Systems).

⁸⁸ Attachment 2 (OSP Response, received on March 8, 2011, to CPSD Data Request No. 4 dated January 11, 2011); Attachment 11 (OSP e-mail response dated April 14, 2011).