103 The Premier referred to the review that the Minister had requested and to my investigation, noting that the government would await their outcome. The Minister also expressed his regrets for Mr. Edmonds’ ordeal and indicated that OLG’s Chief Executive Officer would personally apologize to him. The Chief Executive Officer of the Corporation also responded:
We look forward to any recommendations to improve our customer service and to ensure that we continue to be a leader in lottery security. OLG is committed to full co-operation in the reviews that will be undertaken of our practices and to transparency in all of our activities.
104 The Corporation’s first act was to finally acknowledge that it had done wrong by Bob Edmonds. On Oct. 26, 2006, the Chief Executive Officer announced that the Corporation had, “in an effort to remove any potential barriers to the Ombudsman’s investigation,” decided to release Mr. Edmonds from his confidentiality agreement. The Chief Executive Officer stated:
This was a regrettable situation that should not have occurred, and one that was clearly unacceptable to everyone involved, and I have spoken with Mr. Edmonds and offered him a personal apology.
105 Conscious of how bad it looked that it had spent over half a million dollars battling Mr. Edmonds, the Corporation emphasized on its website that the retailers were ordered to pay restitution. In fact, this was not true – they had voluntarily agreed to pay the Corporation $200,000, and at the rate of $300 a month, it is doubtful that the debt will ever be extinguished. As well, apologizing to Mr. Edmonds, while appropriate and overdue, did not address the underlying systemic issue of the Corporation’s historic indifference to retailer misconduct. Meanwhile, the Corporation was about to devote considerable time, energy and funds to challenging the CBC’s statistical premise.
106 In April 2006, the fifth estate made a request under the Freedom of Information and Protection of Privacy Act for information from the OLG, including insider win forms and statistics showing the frequency of insider wins. This placed the Corporation in an awkward position, as it had not kept such statistics or monitored insider wins. The figures the Corporation provided to the CBC in July 2006 indicated that of roughly $2.6 billion in major prizes won from 1999 to July 2006, about $100 million had been won by insiders. This represented 214 inside wins out of 5,713 big jackpots. While the Corporation provided some insider win forms, about 60 were missing. Accordingly, it was not possible to determine how many of the “insider winners” were retailers.
107 The CBC provided the information obtained from the OLG, along with the results of a survey of 200 retailers it had conducted, to Prof. Jeffrey S. Rosenthal, a Professor of Statistics at the University of Toronto. Assuming 10,300 OLG retail locations, Prof. Rosenthal estimated that the total number of retail employees was 36,050. He also took into consideration evidence from a Corporation official at the Edmonds trial to the effect that retail insiders numbered 50,000-60,000. From all this, Prof. Rosenthal determined that retailers spent 1.5 times as much as the average adult on lotteries.
108 Prof. Rosenthal issued his report, Analysis of Insider Ontario Lottery Wins, for the fifth estate. In it, he concluded that retailers were winning prizes of $50,000 and above about 5.7 times more than expected, assuming a retail population of 36,050 – or, alternatively 3.5 times more than expected, assuming a retail population of 60,000. Either way, he said the probability of this happening by pure luck was less than one chance in a trillion trillion trillion trillion – i.e., impossible. In his opinion this suggested that, similar to Mr. Edmonds, other lottery customers had missed out on major lottery prizes that they had won. He concluded that store owners and employees won significantly more major lottery prizes than they possibly could by chance. He also found that retailers were winning instant-win games about 15 times more often than expected, assuming a retail population of 36,050 (or nine times more often than expected, assuming a retail population of 60,000), and that it was also statistically impossible for this to occur randomly.
109 Even before the fifth estate program was broadcast, the Corporation had been scrambling to dispute this hypothesis. An Oct. 17, 2006 Corporation e-mail stated:
110 Following Prof. Rosenthal’s initial analysis and days before the fifth estate program was to air, the Corporation advised the CBC that its estimate of the total number of retail employees was 140,217. Prof. Rosenthal said he believed the figure was inflated, but he concluded that it still did not explain the large number of insider major lottery wins.
111 My investigators met with Prof. Rosenthal, who said he is convinced that to truly understand the nature of the insider lottery wins, it is important to know how many of the 214 insider wins under investigation came from each of the various lottery retail business types (Convenience Independent, Convenience Corporate, Gas with Convenience, Supermarket, Health Care, etc.). He stood by his original analysis, and explained that he took into consideration that lottery retailers spend more on lottery tickets than the average Ontario adult. He does not believe even the OLG’s latest figures explain all of the recorded insider wins.
112 The Corporation has maintained that it is virtually impossible to have an accurate record of lottery retail employees, given the difference in operations and large turnover in the industry. The 140,217 figure, while appearing specific, was simply a best guess. Based on its own research, the Corporation concluded that there were 10,911 retailers and estimated the number of employees for each type of retailer. For instance, for each independently owned convenience store, it estimated there were four employees selling, with an annual turnover of one, but for each convenience store owned by a corporate chain, it estimated five employees with a turnover of two. For gas stations with convenience stores, it estimated eight employees with an annual turnover of 50%.
113 It is difficult to put much stock in the retailer estimate the OLG gave to the CBC. At one point, a staff member suggested that since supermarkets don’t allow everyone to validate tickets, the number of retail insiders could be reduced to 106,185. At another point, the figure of 175,000 was suggested to capture all those working in retail stores, even if they did not handle lottery tickets.
114 Since the Corporation did not have the statistical information and analysis necessary to refute Prof. Rosenthal’s findings, it set out to create some – but it still couldn’t get the numbers straight. It told the CBC that the number of major wins between 1999 and July 2006 was 5,713, but after it calculated the insider win frequency and found it was 2.4 times the general population for 1999 to 2005, it used a figure of 5,931 major wins. The figure for major wins was later recalculated as 5,385, and still later as 5,337. My office was told that the difference in the later figures likely related to a game being overlooked. However, the Corporation maintains that this does not change the end result of a win rate of 2.4 times that of the general population – as opposed to the substantially greater rates suggested by Prof. Rosenthal’s analysis.
115 The Corporation had a little more luck in identifying the classes of insiders that had won big. For the period from January 1999 to Nov. 6, 2006, it identified 220 insiders as big winners: 78 of those were retail owners, 131 were retail employees, nine were Corporate employees and two Corporate family members. No major wins had been reported by employees of other areas covered by the Insider Win Policy. The Corporation also looked to a market research firm to verify its theory that retailers play more than the general public (and hence should be expected to win more). The firm’s survey of 380 retailers was conducted on Oct. 23 and 24, 2006, and the results reported on Oct. 25 and 26, 2006. It found that on a per capita basis, lottery retailers spent an average of 2.8 times more money on lottery tickets in a four-week span than ordinary consumers ($33.90 vs. $12.30). When different spending rates for owners, managers and employees were factored in, the retailer rate was reduced to 1.9 times that of the general population ($23.30 vs. $12.20). The research company concluded:
116 Just after the Oct. 25 fifth estate aired, the Corporation was contacted by a mathematician. According to his analysis, the probabilities of winning over $50,000 were different depending on the game played, and 200 wins among 140,000 retailers was actually quite possible. He considered two assumptions – that retailers are heavier gamblers and heavier gamblers spend a higher proportion of their money on games that offer a higher probability of winning over $50,000.
117 The Corporation also sought the input of a statistician it had retained in the past to review the odds of games and provide advice. He reported to the Corporation on Oct. 29, 2006. He suggested that it was not possible to render an accurate judgment without a good estimate of the number of tickets purchased by insiders.
118 Without any definitive answer, the Corporation proceeded to engage another market research company to look into data related to the number of insider wins from 1999 to 2006. The company issued its report on Nov. 3, 2006. It noted:
119 The research report indicated that if retailers’ per capita spending is approximately twice as much as the general population, all else being equal, their win rate would be twice that of consumers. It concluded:
120 The Corporation also retained a statistician from a U.S. university, who reported his findings on Dec. 1, 2006. He commented on the uncertainty of the data used in Prof. Rosenthal’s calculation, and suggested that the real questions were whether insiders adjust their playing behaviour as a result of contact with lottery players, and if so, whether they win more frequently than would be explained because of such a learning experience.
121 Yet another professor from a U.S. university reported to the Corporation on Jan. 23, 2007 that there was a substantial degree of uncertainty with respect to the actual number and spending habits of retailer insiders. He suggested “perhaps insiders are better informed and play the games with better chances more often than the general public.” He concluded that “the difference between that expected number and the actual number of wins by insiders may well lie within the limits of normal chance variation.”
122 Finally, one more Canadian university professor of mathematics and statistics gave the OLG his assessment of the fifth estate’s analysis of insider wins on Jan. 29, 2007. He stated:
My overall conclusion is that from the available data, the assertion that the retailers are winning at a higher rate cannot be justified because such an inference is dependent on the number of retailers and their yearly lottery expenditures, which are not known with a high degree of certainty. [….]
Since not all lotteries are played with equal frequencies, this limits the applicability of [the model used by Prof. Rosenthal].
He also opined that “the atmosphere has been sufficiently poisoned since the CBC telecast, that future survey data would not be reliable.”
123 For my Office’s investigation, we retained our own expert, a professor from the Department of Statistical and Actuarial Sciences at an Ontario university, to assist with our analysis of the various competing expert opinions. He echoed the comments of some of the others on the unreliability of the data on which Prof. Rosenthal’s assessment was based. He noted that generally if lottery wins were distributed fairly, one would expect to see that the proportion of prizes won by insiders would be equal to the proportion of money spent by insiders. In his view, the type of probability analysis carried out by Prof. Rosenthal is useful and could signal to the Corporation the need to take steps to prevent insider fraud.
124 Our expert also made a number of recommendations. He suggested that the Corporation develop methods to obtain a reliable estimate, or an actual count, of the insider population on a regular basis. He noted that this could be done by way of a regular census of all lottery sales outlets. For this to work, each lottery outlet would be required to file the names of the owner and current employees selling tickets. He also suggested that the Corporation develop a clear working definition of “insider” as it applies to retailers across all sales channels. He concluded that the Corporation should develop a method to obtain a valid estimate of the total amount spent on lottery games by insiders. Finally, he concluded that as a quality control procedure the Corporation should regularly check the probability of insider wins, using accurate data, and carry out an investigative review of security procedures when warranted. He suggested that this could be done or supervised by the Corporation’s in-house statistician.
125 While, predictably, the experts do not all agree, it seems logical that the more accurate information the Corporation has at its disposal regarding the incidence of retailer lottery wins, and for that matter any insider wins, the better it will be equipped to assess whether there are security issues relating to its various lottery products. By failing to keep track of this information, the Corporation was rendered vulnerable – and so in turn were its customers.
126 In addition to the time Corporate staff spent gathering information and generating statistics on insider wins, the Corporation ended up paying $5,000 for the survey of retailer spending and $44,250 to obtain the views of external statistical experts. The OLG should not be caught off guard again when it comes to gathering and analysing statistics. The Chief Executive Officer told us it would keep track of insider wins in future. However, it is not clear what form this will take.
127 As it became clearer to the Corporation that it could not categorically contest the statistical allegations made by the fifth estate, it attempted to divert the critical public focus with a campaign to restore trust in the lottery system by initiating a number of improvements.
128 Without actually acknowledging that it had not been vigilant enough in protecting the public from retailer fraud and theft, the Corporation began to take steps to win back the “goodwill” it had lost as a result of Luck of the Draw. It retained the forensic audit firm KPMG to conduct the review called for by the Minister. KPMG began its work on Oct. 29, 2006 and provided the Corporation with its Phase I Report on Nov. 3, 2006, in which it made the following five recommendations:
1. OLG should initiate a comprehensive educational campaign to make customers aware of the techniques they should employ to protect purchased tickets. In particular, the campaign should emphasize the need to complete and sign the back of the ticket at the time of purchase.
2. OLG should establish a policy that retailers will only validate tickets if the customer has completed and signed the back of the ticket. Tickets should not be validated and processed unless signed. OLG will need to provide training to retailers on this policy and will need to consider an appropriate approach to motivating and enforcing compliance.
3. OLG should consider a more comprehensive rollout of customer-accessible ticket checkers with an appropriate awareness program.
4. OLG should consider a review of the “insider” definition and policy with a view of making the policy more practical and controllable. As part of the changes to the “insider” policy we would suggest that the threshold of $50,000 that initiates more specific procedures and checks be reduced. Consideration should be given to returning to the 1995 level of $10,000.
5. OLG should consider the implementation of an automatic call logging facility at the Contact Centre. In conjunction with such an implementation, appropriate monitoring and follow-up processes should be developed and rolled out as well.
129 While KPMG was conducting its review, the Corporation was engaged in its own self-diagnosis. Armed with KPMG’s first report, and based on suggestions from its own executive-level brainstorming session, on Nov. 9, 2006, the Corporation implemented a Seven-Point Trust and Security Action Plan, which it touted as a comprehensive package of new and expanded lottery security measures “to maintain and reinforce the trust of every lottery player in the province.” The Chief Executive Officer stated at the unveiling:
If even one player feels that our systems have failed them in any way, that’s one too many… To anyone who has felt the need to question their trust in the OLG, to anyone who feels that we could have done more to protect the integrity of our games, to anyone who feels their concerns were not taken seriously enough, we apologize and we will do better.
130 The Corporation committed to the following seven steps, which mirrored KPMG’s recommendations in a number of respects:
1. Installing at every lottery terminal location, a device enabling consumers to electronically check their own tickets and to see the value of their prize instantly;
2. Lottery retailers will be permitted to check tickets only for those customers who have been asked to sign the back of their tickets. There will be escalating penalties for retailers who fail to follow correct validation procedures, up to and including the removal of their lottery terminals;
3. Customer-facing video screens at lottery terminal locations will tell players in larger characters if they’ve won and if they are a big winner when they have retailers check their tickets;
4. OLG will conduct a public education campaign to assist consumers to protect themselves;
5. OLG’s special toll-free telephone line will allow consumers to provide the OLG with ideas on additional security measures;
6. OLG will use an expanded team of professional investigators to probe every win of $10,000 or more by lottery retailers, compared to the current $50,000;
7. OLG will implement an enhanced complaints process and investigation procedure to improve customer service and to ensure that all accusations are handled effectively and properly.
131 KPMG issued a Phase II report on Nov. 23, 2006, containing 18 additional recommendations relating to specific controls for purchasing and authenticating tickets, retailer management, insiders, the Prize Office procedures, the interview process, handling of tickets, escalation procedures, tracking of incidents, counter ticket redemptions, call centre technology, the investigative process and practices from other jurisdictions. KPMG then moved to consider the issue of instant tickets, issuing a Phase III report entitled Instant Ticket Gaming Review on Feb. 29, 2007. This report contains 17 recommendations relating to instant ticket integrity, instant ticket processes and overall controls. The Corporation has expended more than $644,000 on KPMG’s review alone. That review has identified a number of areas where improvement of the Corporation’s practices and policies is warranted.
132 While the Corporation has expressed a willingness to act, and has taken some initial steps, it still has a long way to go to restore public confidence in the lottery system. At this point it is worthwhile examining what the Corporation has done so far in implementing KPMG’s recommendations and its own Seven-Point Plan.
133 A sea change in the insider win review process has occurred since October 2006. Instead of looking for ways to evade responsibility for retailer conduct, the Corporation has done an about-face and beefed up its procedures. It has reduced the prize threshold for consideration of insider wins to $10,000 and required that all insider win claims be investigated at the first instance by the Investigations Department.
134 The Corporation’s Vice-President, Corporate Security and Surveillance is a seconded Ontario Provincial Police chief superintendent. He believes that the new requirement that investigators probe insider wins from the outset will bring more credibility to the process. It is clear that the techniques formerly used in Prize Office interviews, even when a Security Assistant was present, were lacking in rigour. While the investigators have taken some police training courses, when it comes to lottery investigations, they essentially learn on the job. KPMG’s Phase II report recommended that the Corporation document its investigative processes. It recently issued new investigative guidelines.
135 Until 2004, the Investigations Department had reviewed all insider wins. After the change in practice in August 2004, the number of investigations decreased substantially. In the 2005/2006 fiscal year, the Corporation’s Prize Office processed 36,155 tickets and paid out over $400 million in prizes. There were 763 major winners and 31 insider wins (10 retail owners, 16 retail employees and five Corporation employees). However, the Investigations Department investigated only two insider wins that year. During our investigation, we were told that under the new post-10/25 practice, the Investigations Department has conducted 14 insider win investigations.
136 Remarkably, KPMG’s review revealed that the majority of retail insiders identified during prize claim interviews were unfamiliar with the Insider Win Policy. Our investigation confirmed this. The term “insider” is not precisely defined, and not consistently applied by the Corporation. We were told, for instance, that the policy does not apply to immediate family members of retailers or their employees’ families unless they are also employed in the lottery retail business. However, one official we spoke to suggested that the policy did apply to family members of retailers – and a document we found describing the “Big Win Terminal Freeze” process refers to spouses of retail owners as being affiliated with the Corporation. The Insider Win Policy as currently drafted captures all retail employees, even those in large supermarkets, whether or not they have any contact with lottery tickets. KPMG has recommended that the insider definition be reviewed and the policy further refined. Corporate officials indicate this is in the works.
137 At present, the Toronto Prize Office asks winners if they are “affiliated” with the Corporation. The word “affiliated” has become a term of art for the Corporation, and its meaning is not necessarily self-evident to the general population. Yet it is used by the Corporation on a daily basis to attempt to weed out those who may be subject to its Insider Win Policy. The Vice-President, Sales and Service noted that the issue of who is affiliated with the Corporation “is probably the biggest area of confusion.” The Vice-President, Corporate Security and Surveillance noted that most people would say “no” when asked if they were “affiliated with the OLG.” The Corporation should establish a series of questions that focus on whether the individual actually claiming a prize has any direct contact with the sale or validation of lottery tickets, for instance, as a retailer or employee, or indirectly – for example, through a family relationship. Once again, this is an area in which KPMG has recommended change, suggesting that the Major Prize Win Form be redesigned to ask more specific questions. The Corporation does intend to enhance the interview process for major winners. However, determining the right questions will depend on the definition of “insider,” which is still very much up in the air.
138 There is also potential for smaller-scale criminal activity relating to lotteries which will not not necessarily be captured by this revamped process. Consumer complaints are a fertile resource for the Corporation to use to identify retailers who may be taking advantage of the public through discreditable practices such as taking free tickets, switching tickets, or shortchanging consumers on winnings. However, the Corporation has traditionally not paid significant attention to consumer complaints. The staff who receive these complaints are the same individuals responsible for supporting the Corporation’s retailer partners, and for promoting lottery products. Handling complaints and finding fault with the retailers the OLG depends on does not fit well with these workers’ promotional role.
139 While not as headline-grabbing as theft and fraud, this is another area where my investigation found the Corporation is betraying the public’s trust – and the right of all Ontarians to be treated fairly by a public institution.
140 One of the disturbing features of Mr. Edmonds’ story was the short shrift he received when he initially called the Corporation to complain. While phone records confirm Mr. Edmonds called the Corporation on Aug. 1, 2001, it has no record of any contact with him. Mr. Edmonds is adamant that he spoke to a staff member at the Corporation’s Customer Excellence Centre, the frontline contact for consumers. He claims the person he spoke to was dismissive of his allegations and told him, “That will teach you for leaving a ticket in the store.”
141 While we could not confirm the details of Mr. Edmonds’ contact with the Corporation, our investigation did show that it would not have been unusual for Mr. Edmonds to have received this type of treatment from the Customer Excellence Centre staff. Our review of hundreds of complaint records demonstrated that the Corporation typically treats consumers who complain about retailer fraud as the authors of their own misfortune. In the past, those issues that were pursued further were often simply referred to sales representatives for follow-up during their monthly retailer visits. Very few complaints resulted in a full investigation.
142 The Customer Excellence Centre deals with a very large volume of calls on a daily basis. The Centre is part of the Corporation’s Sales and Service Section, and it is responsible for providing support services to retailers and selling lottery products, as well as answering consumer inquiries. From 2001 to October 2006, the Corporation received 21,995 complaints, the majority of which were against retailers, and 722 of which related to allegations of retailer fraud. The Centre recorded a large number of complaints as “generic,” meaning the callers didn’t give their names. Of 156 calls in 2005-2006 logged as retailer fraud/validation discrepancy complaints, 42 were logged as “generic” with no further follow-up, and closed.
143 A May 27, 2002 Corporate Procedure document instructs Customer Excellence Centre staff on how to address complaints against retailers. It says, in part:
If a caller says they didn’t receive what they paid for, inform the caller that a player is responsible for ensuring they receive what they paid for at the time of purchase. However, we will investigate concerns and follow up with the retailer….
144 An October 2005 document indicates that the Customer Excellence Centre typically would only refer complaints to investigations in the case of a “repeat offender.” Many individuals who complained to my Office about the Corporation stated that staff failed to ask them specific questions, seemed disinterested, did not provide information on its process for examining complaints and placed the onus on them to pursue their complaint against the retailer or advised them that it was their responsibility to pre-check or sign the back of their ticket. We also learned that the Corporation did not have a way to track the number of complaints about individual retailers or to cross-reference information between consumer and retailer files.
145 We reviewed records relating to 722 consumer complaints received by the Corporation about retailer fraud and validation discrepancies. Our review revealed that the level of detail captured by Customer Excellence Centre staff varied significantly from one employee to another. The level of initiative also varied. Some staff would attempt to verify retailer transactions, consult the database for retailer complaint history, or contact the retailer for information, while others would take no action in response to similar complaints.
146 In one case, a consumer complained that a retailer denied his ticket was a winner even though a scanner confirmed it was. The Centre’s database showed this was the third such complaint against the retailer. The complaint was forwarded to a sales representative, and there is no record of what happened after that. In another incident, recorded March 1, 2006, a consumer was told “once he leaves the store, he accepts the transaction and it is now between him and the retailer.” In another incident on Sept. 20, 2006, a consumer received only $10 from a retailer even through the validation slip showed two free tickets had also been won. She was advised by Centre staff that the problem was between her and the retailer.
147 The Customer Excellence Centre deals with two categories of people – the public and retailers. Given that retailers are the lifeblood of the organization, it is not surprising that we discovered that Centre staff paid particular attention to documenting retailer contacts. But when they followed up on complaints about retailers, their inquiries were not particularly robust. In one case, a customer alleged that a retailer was charging commission for prize redemptions. The Corporation’s records indicate that a Centre staff member called the store, and the retailer flatly denied the allegation. The data entry notes, “I told him that I was glad to hear it … because charging for redeeming of lottery prizes is a violation of our contract.” This is where the inquiry ended. In one case we reviewed, a senior gave a store clerk a ticket for validation in July 2005, but the clerk refused to give it back. When the customer threatened to check her numbers elsewhere, the clerk relented. The ticket was a $20,000 winner. The store clerk was eventually terminated.
148 The Customer Excellence Centre experienced a wakeup call when the fifth estate aired Mr. Edmonds’ story. From Oct. 25 to Dec. 10, 2006, the Centre received 454 complaints from consumers. The OLG hastily moved to add three new choices to its interactive voice response telephone system: One relating to lost or stolen tickets, one relating to claiming a prize and the third relating to suggestions for improving the Corporation’s security measures. In November 2006, it adopted the practice of assigning all validation discrepancies immediately to the Investigations Department. However, not all instances relating to potential retailer fraud are being referred directly to Investigations. A Dec. 19, 2006 information document instructs Customer Excellence Centre staff that complaints involving free tickets or exchange tickets that have not been received by the customer are to be assigned to the Prize Office and “will be monitored, reviewed and escalated by management team as required.” This is the very area that has created the most controversy relating to suspicious and disputed claims.
149 KPMG made a number of recommendations for improving the Customer Excellence Centre’s systems, including an automatic call logging facility, together with appropriate monitoring and follow-up processing, defining “key calls,” and updating the technology so that calls are recorded and supervisors can listen to calls anonymously for quality assurance and training purposes. The Corporation is investigating the installation of technology that will allow it to record calls. We were advised by a senior official that this had been considered three or four years ago, but a decision was made not to proceed because of the cost.
150 The Corporation is also considering the creation of an “expert” group within the Customer Excellence Centre. Its members would receive additional training, and the telephone system would route incoming calls to the appropriate subject matter expert. This approach was encouraged by KPMG, however, it recommended a separate complaint line that routes appropriate calls to trained staff who can fully diagnose the call and gather preliminary information, and that these calls go directly to Corporate Security and Surveillance.
151 It makes eminent sense that the sales staff in the Customer Excellence Centre be removed as the first line of contact for complaints. Given the need for them to foster their relationship with retailers, requiring them to deal with consumer complaints about those same retailers creates a potential for bias to emerge. The current situation in which sales staff may be asked to follow up on compliance issues is also inherently inconsistent with their primary role. It is inappropriate to place them in such a position.
152 As a result of the deluge of complaints after Oct. 25, 2006, the OLG’s Investigations Department (which consisted of three investigators and a manager; a fourth investigator was hired Jan. 8, 2007) had to enlist the aid of a private firm, which has supplied 20 former police officers to assist with investigations at an estimated cost of $20,000 a week. The Investigations Department told us it had received 591 retailer fraud complaints from the Customer Excellence Centre since Oct. 25, 2006, compared to 64 retailer integrity and improper prize complaints received during the entire last fiscal year from all sources. However, it found only two cases where there was a suggestion of retailer fraud, both involving small amounts of money. The Vice-President, Corporate Security and Surveillance indicated that once the Corporation is able to manage its complaint volume, it will reassess its investigative staffing needs. The Corporation also plans to hire three senior managers in the Corporate Security and Surveillance Division, one of those for Investigations.
153 Another failing highlighted by the Edmonds case is the Corporation’s investigative record-keeping. The Investigations Department uses a format in recording case information in its database known as “Occurrence Details.” Information is entered chronologically and in this way, it is similar to a log or a police officer’s notebook. Investigators are expected to enter information from their notes into the investigations database as soon as possible. At the conclusion of an investigation, investigators are required to prepare a summary or a closing report. In the course of the OPP’s investigation of Mr. Edmonds’ allegations, the Corporation forwarded the Occurrence Details to the investigating officer in January 2002. It was later discovered that additional entries had been made, which were not part of the record provided to the OPP. According to the Corporation, the hard copy of the Occurrence Details in its file had not been reconciled with the information kept by its Investigation Department at another office location. The missing information was eventually forwarded to the OPP in another format. What the incident revealed, though, is troubling – the OLG’s system does not prevent someone from altering the chronology after the fact.
154 While OLG investigators do receive some police training, they have not been held to the standard of record-keeping typically used by police officers and many regulatory agencies. In such organizations, for instance, notes must be taken contemporaneously with events, any amendments must be recorded under a new entry date along with a reason for the addition, and no original notes are ever discarded. KPMG’s Phase II report recommended that the Corporation adopt specific investigative documentation standards and guidelines. The Corporation has indicated that it is in the process of documenting procedures for its Investigations Department. The integrity of the Corporation’s investigations would be bolstered by a more disciplined approach to investigative record-keeping.
155 The OLG’s Seven-Point Plan calls for “escalating penalties for retailers who fail to follow correct validation procedures.” KPMG has recommended that any retailer “not following specific directives relating to the validation and redemption of tickets may be terminated” – that is, the retailer would lose the right to sell OLG products. However, the Corporation has been tentative about instituting a new compliance regime for retailers. In the past, it has relied significantly on its sales force to assist with identifying and following up with issues relating to non-compliance. Since the fifth estate broadcast, sales representatives have been asked to openly observe how retailers validate tickets – including whether customers are being asked to sign tickets. However, this practice would appear to be of limited value, given that retailers can be expected to be on their best behaviour during these visits, which are usually made with advance notice.
156 During our investigation, senior corporate officials emphasized that termination of a retail relationship is a very serious step that has been relatively rare and generally only used as a last resort. We were advised that most terminations have occurred as a result of retailers failing to meet their financial commitments to the Corporation.
157 KPMG noted “it may be beneficial to have periodic compliance reviews of retailers by individuals who are independent of the sales force.” With more at stake now, as their retail customers face the very real prospect of termination, the Corporation’s sales staff would be placed in an even more untenable position if they were to continue to have a significant compliance role. From a practical point of view, the skill set required to promote products and cultivate relationships with retailers conflicts with the talents needed to ensure proper enforcement of the rules.
158 The Corporation’s past approach to monitoring retailers has very much been a conciliatory one. From time to time, the Sales and Service section has engaged in “mystery shopper” campaigns in which retailers were rewarded when “caught” doing the right thing. A similar strategy, minus the rewards, was carried out recently through an external audit to test retailer compliance with ticket validation procedures. However, so far, the OLG has not shown much vigour when it comes to investigations relating to the lotteries.
159 It has been said that “quiet persuasion will usually produce one result – quiet inaction.” It is time for the kid gloves to come off. There should be zero tolerance for retailer dishonesty. This should be clearly set out in the OLG’s retailer agreements and rigidly enforced. Once dishonesty is established, termination should be immediate, and in most cases permanent.
160 The OLG’s greatest stumbling block in this regard, however, is its own culture. It spends so much time promoting its relationships with retailers, it is difficult for it to see itself as the enforcer. We were advised that the Investigations Department has occasionally conducted surveillance of retailers, for instance, to address specific allegations of sales to minors. However, the Corporation has never implemented any systematic program to test retailers. It is now proceeding with caution – and some reluctance – into the enforcement waters. The Senior Vice-President, Lottery suggested to us that it may implement different incentive-based programs and at some point introduce a “penalty equation” to the process. He explained “this is all about helping retailers understand it’s important, why it’s important … learn the behaviour, use positive reinforcement before we start to say ‘you just lost your right to sell lottery tickets.’”
161 With respect, his comments sorely miss the mark. Enforcement should be designed to ensure compliance and protect the public from the risk of potential theft and fraud or other practices that give retailers an unfair advantage. The Corporation’s friendly, educational approach is too touchy-feely when a get-tough approach is what is required.
162 In a second fifth estate program on Nov. 22, 2006, the Corporation suffered renewed criticism relating to its instant-win games. Mohan Srivastava, a geo-statistician, told the CBC that in 2003, he had discovered how the Corporation’s “Tic Tac Toe” game could be decoded – based on numbers visible on the unscratched face of the ticket – to predict winning tickets with 95% accuracy. He said he had alerted the Corporation, and it had removed the tickets from the market. However, in typical fashion, it had never gone public with this information.
163 The Corporation’s records reveal that since its discussions with Mr. Srivastava, it has asked an external auditor to review a number of instant-win games. We found two letters, dated July 7, 2003 and Sept. 22, 2003, and 12 audit reports dating from May 3, 2004 to March 20, 2006 addressing this issue. While these documents confirmed that the games examined complied with OLG specifications, all but one contain the following disclaimer to the effect that the review
… cannot provide complete assurance that a statistically based method or algorithm for increasing the probability of selecting winning or non-winning tickets, based solely on visual analysis of tickets, does not exist.
164 Despite this, the Corporation has shown no particular interest in pursuing the issue of possible exploitation of instant ticket games further. It saw the Tic Tac Toe flaw as a fluke printing error, and chose not to obtain an extensive report that Mr. Srivastava had prepared. On Feb. 16, 2007, at the CBC’s request, he conducted a demonstration using 20 SuperBingo scratch-and-win tickets. Employing his analysis of statistical patterns and specific ticket flaws, he was able to predict winners and losers with a 70% success rate.
165 Mr. Srivastava told our investigators that the methodology employed by the Corporation’s external auditors would not catch the problem he has identified and warned that his statistical examination could be reproduced on a large scale to defraud the lottery system. The Corporation should examine such claims closely. Retailers and other insiders with access to large numbers of tickets are in the ideal situation to take advantage of ticket design defects. This reinforces the need for the Corporation to gather more specific information on the winning patterns of retailers generally, as well as to track individual retailer wins, in order to identify suspicious behaviour.
166 KPMG’s Phase III report refers to the ticket flaw revealed by Mr. Srivastava in 2003, and notes that since the incident a number of new preventive procedures have been put in place. These include expanded risk management procedures in the printer’s programming department and independent audits to ensure adherence to specific game standards. KPMG recommended the Corporation implement similar risk management procedures for all new game types, security features and/or printing techniques. While not directly addressing the potential for instant-win tickets to be deciphered and manipulated, KPMG suggested that the Corporation consider the periodic need for advanced statistical expertise as part of the ticket design process. KPMG also noted that at present, those responsible for ticket security report through Marketing to the Senior Vice-President, Lottery, raising the potential for security decisions to be influenced by marketing considerations. KPMG recommended that the reporting relationship be redirected away from the Lottery Group.
167 Recently concern has also focused on the Corporation’s practice of keeping instant game tickets on the market long after all of the top prizes have already been won. Some jurisdictions remove tickets in these circumstances, and others such as British Columbia and some U.S. states include cautionary disclaimers on their tickets. Unlike retailers, who have ready access to information about prize claims for instant wins, unsuspecting consumers continue to purchase tickets dreaming of big wins, when only smaller prizes remain. Now that the OLG is more attuned to consumer concerns, it has acted to address this issue. Since Dec. 18, 2006, it has changed its policy so that games with “life-changing” prizes of $500,000 and up are removed from the market when the top prize has been won. The Corporation is also contemplating a way for information on unclaimed instant prizes to be viewed on its lottery terminals and has added a disclaimer on the back of new tickets that were ordered in December 2006. Tickets coming onto the market in April or May of this year will carry this message:
At the time of purchase, some prizes, including top prizes, may have already been claimed. Visit www.OLG.ca for prize details.
168 The new tickets will also refer consumers to the Corporation’s website for more information on the prizes available. As well, new validation procedures will soon require that a validation slip be produced for each instant ticket.
169 So, preliminary changes have been implemented to the OLG’s procedures for insider wins, investigations and complaints. The Corporation has also assembled a task force to implement technological improvements. These include the installation of self-serve ticket checkers, customer display modifications, terminal freeze changes and enhanced interactive capabilities. However, many of the initiatives are in the early stages or are still under review. The cost of implementation of the Seven-Point Plan alone is an estimated $11,230,000. The bulk of this expense – $10 million – is attributable to the ticket checkers to be installed at all online retail locations.
170 Self-serve ticket checkers allow consumers to check their own online lottery tickets. They are used in a number of provinces and territories in Canada and in at least five U.S. states. The Corporation had launched a pilot project to install these checkers in 250 locations in the summer of 2005, but some retailers expressed concerns that they reduced their interaction with customers and resulted in missed sales opportunities. We were also advised that this initiative did not move more quickly because of cost. But Oct. 25, 2006 changed all that, and it is anticipated that installation of 8,900 ticket checkers will be complete by the end of March 2007. The next step will be for the Corporation to look at ways to adapt the machines to allow consumers to check instant tickets as well. KPMG’s Phase III report highlighted concerns with the current technology for validating instant tickets. It recommended that the Corporation move away from the current instant ticket validating technology, which involves dialing up on a phone line. KPMG’s review revealed that some of these instant ticket devices are located in back rooms away from customers’ view, there is no sound or other signal that a ticket is a winner (except for the receipt printout), and there is no equivalent of the “terminal freeze” for major wins.
171 The Seven-Point Plan requires retailers to ask customers to sign their tickets before validating them. KPMG’s Phase I report had actually gone further, suggesting that it be mandatory that tickets be signed before they were validated. At first, the Corporation had notified retailers through its November 2006 Retailer Bulletin that tickets had to be signed, but this message was modified in December 2006. Retailers were then told that they must ask customers to sign their tickets, but that they could still validate tickets if the consumer refused. In November, the messaging on the front of online tickets was changed to remind consumers to always sign the back of their tickets. In November and December 2006, the Corporation’s sales staff visited retail outlets to provide training on the new validation process.
172 The Corporation engaged a marketing company to measure retailer compliance with the new procedures. The company visited 1,949 retail outlets in London, Ottawa, Hamilton/Burlington, and Toronto between Dec. 4, 2006 and Jan. 5, 2007. The audit found that 77% of retailers did not check the back of the ticket when it was submitted for validation and that 78% did not request that the ticket be signed prior to validation. The Corporation is considering having the company conduct a second audit. The Vice-President, Sales and Service suggested that making signatures mandatory would create operational problems for store owners because of the additional time it would take. One of the Directors of Lottery Marketing told us lottery tickets are a “low consideration purchase” and suggested that any attempt to impose additional requirements on consumers could have an adverse affect on sales.
173 The officials we interviewed differed on their understanding of the OLG’s past practice with respect to signing tickets as a means of consumer protection. One official suggested that the Corporation had always encouraged this. Another commented that it had not asked consumers to sign their tickets for 30 years, and expressed concern about a backlash. One director observed, “Did I ever sign the back of my ticket before the fifth estate? No, but do I do it now? Yes, I do.”
174 In its Phase II report, KPMG recommended repositioning all retail terminals so that it would be easy for ticket holders to observe their tickets being checked. Between Nov. 13 and Dec. 15, 2006, the Corporation’s sales force inspected Customer Display Units at all retail outlets and found only a few required readjustment. We were told that in the process of installing ticket checkers, the Corporation will assess both terminal and Customer Display Unit positioning. The “freeze” now occurs for online wins of $10,000, consistent with the insider win threshold.
175 The Corporation is planning several additional enhancements to its Customer Display Unit screens. On Dec. 5, 2006, the Corporation’s Executive approved display of all prize amounts of $1,000 or less. The Big Winner video, jingle and text message have also all been expanded, providing more signals to customers that they have won. The font size of screen messages is also being increased for all validation messages. All of these measures were to be in place by March 2007.
176 Public education activities launched by the OLG since November 2006 include:
· “sign your ticket” decals, which appear on Corporation merchandise inside stores;
· “check your lottery tickets/always sign ticket back” messaging on online tickets;
· new messaging on Customer Display Units;
· “sign your ticket/check your ticket” advertisements at play stands (i.e., where selection slips are filled out);
· “silent seller” strips (signs) which are placed inside the unit used to display instant tickets;
· the introduction of a 15-second “sign your ticket/check your ticket” video clip; and
· tear pads (flyers), which identify sources where consumers can check their numbers.
177 The Corporation has also placed ads encouraging consumers to sign their tickets in 40 Ontario daily newspapers, which ran on Dec. 18, 2006, and Jan. 18, 2007, with the message “Play with confidence.” More advertisements are being considered for newspapers, radio and television. The Corporation has retained an advertising agency and public relations firm to assist it with its media messaging. However, it is already planning to scale back on its consumer messaging. Beginning Jan. 15, 2007, the OLG began to limit the appearance of the “sign your ticket/check your numbers” messages on online tickets to once a week. Next, it plans to place ads in newspapers and other media relating to self-serve ticket checkers – how they work, and where they are available. The Marketing division has ordered 1-2 million ticket wallets (vinyl envelopes) that are used by some players to retain their selection slips and/or tickets, which will carry the slogan “Sign Your Ticket” and which retailers will distribute. In the long term, the Corporation is also considering placing a “Sign Your Ticket” logo permanently on its online ticket stock and instant tickets.
178 The Corporation will measure the effectiveness of its public education initiatives. It already uses a research firm to assess consumer behaviour, attitudes, and support for lotteries. It is considering having this company conduct a follow-up survey to gauge public awareness of the changes. Another idea being considered is to sample retailers and have them audit consumer behaviour.
179 The Corporation has undertaken a number of initiatives to facilitate public suggestions for improvement. In addition to contacting the Customer Excellence Centre between 8:30 a.m. and 8:00 p.m., seven days a week, customers can now reach the OLG through e-mail addresses created specifically for suggestions, questions, concerns and reporting lost or stolen tickets. The Corporation’s goal is to respond to 99% of its inquiries within seven days.
180 With all the frenzied changes taking place at the Corporation, the question remains, has it learned its lesson? What will happen if another retailer shows up to report a win and can’t answer basic questions about the ticket? How will the Corporation respond to another Bob Edmonds?
181 One of the Corporate officials we spoke to during our investigation indicated that when he heard about the Edmonds case years ago, his first reaction was that the Corporation should pay Mr. Edmonds “sooner rather than later,” if he was indeed the genuine winner. But business and legal decisions were made to the contrary, and the rest is history. The Vice-President, Corporate Security and Surveillance, who was not with the Corporation at the time Mr. Edmonds initially complained, remarked that “hindsight’s always fantastic.” His advice now would be, “Give him the $250,000 and go after [the retailers].”
182 The Chief Executive Officer also acknowledged that the Corporation could have done better and could have been “more aggressive in trying to ferret out what had transpired.” When asked, “If Mr. Edmonds came to the organization today, would his experience be different?” the Chief Executive Officer offered:
To say that we’ve all had sensitivity training would be an understatement… What we have learned is we need to be more sensitive and put more time and effort into it.”
183 The Corporation did recently settle a dispute over a Pro-Line sports lottery that might otherwise have resulted in costly and prolonged litigation. In that case, the OLG had inadvertently transposed the odds for a soccer game for a period of nine hours. The complainant had bet on the game during that time. When he went to collect his winnings, he was told his ticket was void. At a pre-trial meeting, the presiding judge told the Corporation that its case was vulnerable, as the game rules were not readily available, so the average person would not be aware of the rules. The Corporation settled for $65,000, and as a result of this case, it now publishes all Pro-Line sports rules on its website.
184 However, the Corporation still seems to treat such complaints as one-offs. It requires complainants to sign confidentiality statements when they settle their disputes, and does not post information about errors on its website. At least one other Canadian counterpart, the Western Canada Lottery Corporation, ensures that error information is posted on its site. This would appear to be a simple, transparent and reasonable practice.
185 While the OLG deserves some credit for finally taking some decisive action to address the fallout from the fifth estate, there is a very real danger that some of its initiatives will result in mere window dressing. It continues to exhibit a reluctance to get tough when it comes to retailer compliance issues – as its own research shows, most retailers still (as of January 2007) seem to resist even asking people to sign their tickets. The OLG still appears not to recognize that controls need to be designed and enforced to protect the public and the integrity of the system. Compliance cannot be treated as an option merely to be negotiated with retailers.
186 The Ontario Lottery and Gaming Corporation is very good at making promises – after all, it is well versed in convincing Ontarians to follow the dream of the big win. As the Vice-President, Sales and Service commented, “my job is about celebrating the win.” So when the Corporation boasts that its security is the most stringent in North America, such statements should be taken with a grain of salt. Unfortunately, prior to Oct. 25, 2006, the Corporation had fallen into the trap of buying into its own hype, instead of looking inward with a critical eye. It has finally awakened to the reality that it is expected by the public – and by the courts – to take an active role in safeguarding the lottery system from fraud and theft.
187 One OLG official commented to our investigators:
We will be better because of the fifth estate, we will be better because of KPMG and we will be better because of the Ombudsman’s review…
The Corporation is clearly trying to say all the right things in order to convince the public that its commitment to improving the security of the lottery system is a reality, not just a sales pitch. Whether this commitment will ultimately be fulfilled remains to be seen.
188 In order to ensure that the focus on implementing proper controls is not simply a passing fancy, the responsibility for review of security issues relating to insiders needs to be assigned to a senior level of the organization. The Corporate Security and Surveillance Division seems the most logical fit for this. The momentum for change must not be lost in a series of task forces or committees. In its Phase III report, KPMG recommended that the Corporation consider creating a position focused on managing the comprehensive risks of lottery activities, first focusing on instant-win tickets and then expanding over time. It is clear that there must be leadership and direct accountability to ensure that the lessons learned from the Edmonds case are not lost. The Chief Executive Officer has already recognized this necessity, noting, “we need to embed within the Lottery Division a resource or resources whose sole purpose is to essentially continue to review the status quo.”
189 I am not convinced, however, that the public can rely on the Corporation alone to ensure that real reform takes place. The danger is too great that the OLG will continue to fall back into its old habits of coddling retailers and dismissing consumers’ legitimate complaints.
190 One option the Corporation has not considered to counteract retailer theft and fraud is to ban retailers from participating in lotteries altogether. It remains confident that its security measures, which include internal and external audits of its processes and games, are sufficient for the most part to ward against insider fraud. Our review of other jurisdictions confirms that the overwhelming majority do not have any special restrictions on retailer play. A number of U.S. jurisdictions prohibit lottery employees from playing, but not retailers. We only found one jurisdiction, Argentina, which banned retailers outright from participating in lotteries, and the sales structure in that country appears to be significantly different from Ontario’s. In British Columbia, retailers are restricted from playing lottery games while on duty, and their claims are also subject to additional review procedures.
191 Banning retailers from buying lottery tickets altogether might appear to be an obvious solution, since it would remove any possibility of them claiming prizes. However, OLG officials have questioned the practicality of this approach, suggesting that a ban would only serve to force retailer lottery playing underground. They argue that British Columbia’s approach would be virtually impossible to police, as retailers could simply ask others to present their tickets for them. While the Corporation could try to identify relatives of retailers as insiders, the list of friends and acquaintances who could potentially be asked to redeem winning tickets on their behalf is endless. The retail employee population is also large, transient and difficult to quantify with accuracy. There is also the question of fairness in punishing thousands of retailers for the crimes of a few.
192 The idea of an outright ban on retailer lottery playing should not be dismissed out of hand, but I am satisfied that other measures for ensuring the honesty and integrity of the lottery system should be explored first.
193 The Corporation does not pre-screen those people who apply to sell OLG lottery products, or concern itself with questions of retailers’ past conduct or personal suitability. As the Vice-President, Sales and Service put it to us:
We don’t believe that’s in our responsibility area… there’s human rights issues and all sorts of things…[there] might be very large issues for which we have just no responsibility.
194 The Terminal Selection Guidelines in use at the OLG focus on profit potential. To qualify for consideration as an on-line retailer, a store must generally have sold instant-win products for a minimum of 26 weeks and must have high sales. There are exceptions for retailers with high profit potential, those in remote areas, or with seasonal businesses. A letter of credit or security deposit is also required. Once appointed to sell OLG products, retailers are entitled to commissions on sales and cash redemptions.
195 While the Corporation does not carry out any form of credit or criminal reference check when an application is made by a retailer, retailers convicted of offences under the Retail Sales Tax Act can have their right to sell OLG products suspended. Similarly, retailers who sell to minors in contravention of the Ontario Lottery and Gaming Corporation Act, 1999 may face penalties. Failure to meet financial obligations is also grounds for suspension. The Corporation has generally used a three-strike approach: The first compliance issue results in a one-week suspension; the second in a one-month suspension and the third in a review of the retailer’s right to sell lottery products.
196 The Corporation’s standard-form contract with retailers states that: