U S Identity Fraud Rates by Geography Tom

U. S. Identity Fraud Rates by Geography Tom Oscherwitz Vice President of Government Affairs & CPO ID Analytics, Inc. Economic Crime at Home and Abroad 2008 ECI Conference October 22 -24 ID Analytics Confidential © 2008

Outline • Basic Facts About Identity Fraud • About ID Analytics, Inc. • Methodology of Geographic Studies • Original Study: Key Findings • New Findings and Insights - Emerging Hot Spots - Case Studies ID Analytics Confidential © 2008

Difference Between Lost/Stolen Credit Card and ID Fraud Lost/Stolen Credit Card • Misuse of an existing account • Found by examining transactions • Single institution, single account • Financial loss limited • Straightforward to close account, issue new cards • Easily reconciled; not much consumer harm Identity Fraud • Misuse of a complete identity: stolen or fabricated ID • Found by examining credit initiations • Typically many new accounts of many types opened at many different institutions • Cross-industry misuse: financial, insurance, health care, legal… • Very hard to identify and reconcile • Substantial consumer harm: financial, psychological and lost time FRAUD TYPES ID Analytics Confidential © 2008

Identity Fraud – Criminal Attraction • Financial fraud - $6000 avg. new account fraud loss. Doubled from 2005 -2006 - Auto loans, mortgage, tax returns and utilities - Wireless: subscription fraud, equipment loss; account take over • Insurance fraud - Auto, workers comp, life, health and disability • Healthcare benefits - Estimated 250 k Americans had health records stolen and misused • Immigration fraud - Employment related fraud accounts for 14% of stolen identity usage • Identity manipulation - Altering identity to hide negative events • Avoiding legal sanctions - Warrants, child support and alimony • Identity fraud costs the U. S. about $48 -53 billion/year 1 Gartner, Survey, 2006 Privacy Forum, Businessweek 3 Federal Trade Commission, Report, 2006 2 World ID Analytics Confidential © 2008

Cost to Obtain an Identity • Aspects of an identity (bank account, credit card, date of birth and government-issued identification number) can be purchased for $14 to $18 • Quality documents can be purchased in chat rooms: - $150: Driver's license - $150: Birth certificate - $100: Social Security card Symantec Corp. , Internet Security Threat Report, March 2007 Business. Week, “Coming to Your PC's Back Door: Trojans”, January 2006 USA Today, “Cybercrime flourishes in online hacker forums”, October 2006 ID Analytics Confidential © 2008

Two Types of Identity Fraud ID Theft (True-Name ID Fraud) Synthetic ID Fraud • Fraudster steals and uses a real person’s identity information • Fraudster invents a fictitious, synthetic person • There is a clear victim who can report theft • Frequently combines pieces of real IDs (names, addresses, SSNs…) • Minority of identity fraud • Usually makes slight modifications (first/last name changes or modify last 4 digits of SSN) • There is no clear victim to report theft • These are not seen through victim surveys • Majority of identity fraud ID Analytics Confidential © 2008

ID Analytics Overview • Leader in on-demand identity intelligence - Identity Risk Management - Compliance and Authentication - Customer Management - Credit Analytics • Pioneered identity scoring technology • Provides unprecedented real-time visibility into the risk of individuals, protecting both organizations and consumers • Solutions drive new revenue opportunities, reduce financial losses, and facilitate compliance with federal regulations • World-class customer base - 4 of top 5 communications companies - 6 of top 10 financial services companies - Major retailers, government agencies and healthcare insurers • Key strategic partners such as Trans. Union and VISA ID Analytics Confidential © 2008 8

ID Network® ID Network • First national, cross-industry compilation of identity information • 360 billion total aggregated attributes • 750 million unique identity elements • Average daily flow = 45 million • 2 million reported frauds • 1 billion consumer transactions • Contains information about: Credit applications Card transactions Payments Change of name/address Demographics Data is never sold or distributed ID Analytics Confidential © 2008 9

Personal Topology™ Personal Topology • A representation of individual’s particular identity characteristics • Including connectedness to other individuals and their unique identity characteristics ID Analytics Confidential © 2008 10

Advanced Analytics. SM Advanced Analytics • The use of machine learning algorithms, complex rules, and optimization processes • Applied to terabyte-sized information systems • Real time delivery capabilities ID Analytics Confidential © 2008 11

On-demand Identity Intelligence Through on-demand identity intelligence, businesses and government organizations can immediately reveal the risk level and revenue opportunity of any individual at critical points of interaction and realize more potential from more customers. ID Analytics Confidential © 2008 12

Geographic Fraud Study: Methodology • This research is based on frauds confirmed by businesses rather than on consumer victim reports • First study: 2003 through early 2006 • New hotspots study: all of 2006 • Frauds examined at different levels: state, three-digit and five-digit ZIP Code • Fraud rates are naturally normalized to a per capita basis: Large city Fraud rate = ID Analytics Confidential © 2008 # frauds # applications 1% = Small town 1, 000 = 10 100, 000 1, 000

Some Caveats… • Victim or perpetrator? For this study we use the address on the application. This address is associated with fraud. This uncertainty becomes less important at larger area groupings (e. g. ZIP Code). • Not all markets are created equally - 3 and 5 digit ZIP Codes can be rural or urban - We care about fraud rate and % increase vs. absolute # of frauds • We can conjecture, but we don’t definitely know why fraud rates increase ID Analytics Confidential © 2008

U. S. Identity Fraud Rates by Geography Original Findings Data from 2003 - Early

ID Fraud Rates by States Very low Low Average High Very high ID Analytics

Risky States CONSISTENTLY RISKY STATES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. New York Nevada California Michigan Arizona Illinois Oregon Washington Texas Georgia DC Florida CONSISTENTLY NON-RISKY STATES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. New Hampshire Vermont Montana Ohio Idaho Wyoming Iowa Maine South Dakota West Virginia Kentucky South Carolina Maryland New Jersey Arkansas Data from 2003 through early 2006 ID Analytics Confidential © 2008

ID Fraud Rate by 3 -Digit ZIP Code Very low Low Average High Very

Riskiest 3 -Digit ZIP Codes ZIP 3 Region 100 112 482 110 101 118 104 948 900 303 114 916 914 106 903 902 113 906 752 924 891 972 927 116 907 850 111 New York, NY Brooklyn, NY Detroit, MI Nassau County, NY New York, NY Nassau County, NY Bronx, NY Contra Costa, CA Los Angeles, CA Atlanta, GA Queens, NY North Hollywood, CA (LA) Van Nuys, CA (LA) White Plains, NY Ingelwood, CA (LA) Bell, CA (LA) Flushing, NY Norwalk, CA (LA) Dallas, TX San Bernardino, CA Las Vegas, NV Portland, OR Tustin, CA Far Rockaway, NY (Queens) Harbor City, CA (LA) Phoenix, AZ Long Island City, NY (Queens) Data from 2003 through early 2006 ID Analytics Confidential © 2008 Fraud Rate Over Background 4. 0 3. 2 2. 9 2. 8 2. 5 2. 3 2. 2 2. 0 1. 9 1. 8 1. 7 1. 7 ZIP 3 Region 102 908 103 946 911 606 770 913 904 331 919 200 821 918 912 722 986 917 203 387 905 333 947 984 910 115 941 New York, NY Long Beach, CA (LA) Staten Island, NY Oakland, CA Pasadena, CA (LA) Chicago, IL Houston, TX Northridge, CA (LA) Santa Monica, CA (LA) Miami, FL San Diego, CA Washington, DC Yellowstone National Park, WY Alhambra, CA (LA) Glendale, CA (LA) Little Rock, AR Vancouver, WA San Bernadino, CA Washington, DC Greenville, MS Torrance, CA (LA) Fort Lauderdale, FL Berkeley, CA Tacoma, WA Altadena, CA (LA) Garden City, NY San Francisco, CA Fraud Rate Over Background 1. 7 1. 6 1. 5 1. 5

ID Fraud Rate by 5 -Digit ZIP Code Very low Low Average High Very

ID Fraud by ZIP Code: California, Nevada Data from 2003 through mid 2006 ID

ID Fraud by ZIP Code: New York City ID Analytics Confidential © 2008

ID Fraud by ZIP Code: Los Angeles ID Analytics Confidential © 2008

Riskiest 5 -Digit ZIP Code ZIP 11005 57438 11804 10022 11030 60654 10026 49839 10021 97532 10017 15821 10016 58466 58541 78879 11228 94972 50433 55801 10007 38132 10918 25843 08526 Region Floral Park, NY (Queens) Faulkton, SD Old Bethpage, NY New York, NY Manhasset, NY Chicago, IL New York, NY Grand Marais, MI New York, NY Merlin, OR New York, NY Benezett, PA New York, NY Marion, ND Golden Valley, ND Rio Frio, TX Brooklyn, NY Valley Ford, CA Dougherty, IA Duluth, MN New York, NY Memphis, TN Chester, NY Ghent, WV Imlaystown, NJ Data from 2003 through early 2006 ID Analytics Confidential © 2008 Fraud Rate Over Background 63. 3 12. 3 10. 9 9. 7 8. 6 8. 4 7. 8 7. 5 7. 2 7. 1 6. 8 6. 7 6. 6 6. 5 6. 4 6. 2 6. 1 ZIP 90024 10014 62833 11226 48235 10003 11530 65501 69167 10309 61460 10024 99557 95232 10950 10028 50558 11362 11203 10011 10020 65258 70084 10025 94074 Region Los Angeles, CA New York, NY Ellery, IL Brooklyn, NY Detroit, MI New York, NY Garden City, NY Jadwin, MO Tryon, NE Staten Island, NY Media, IL New York, NY Aniak, AK Glencoe, CA Monroe, NY New York, NY Livermore, IA Little Neck, NY (Queens) Brooklyn, NY New York, NY Holliday, MO Reserve, LA New York, NY San Mateo, CA Fraud Rate Over Background 6. 1 6. 0 5. 9 5. 8 5. 7 5. 6 5. 5 5. 4 5. 2 5. 1 5. 0

ID Fraud by Population Density FRAUD RATE ABOVE AVERAGE Urban identity fraud rates much

ID Fraud Risk Shows an Income “Smile” INCOME • High AND low incomes are more risky 130% 120% • Middle incomes are less risky 110% • Effect is strong and evident in a number of measures 100% 90% 00 -35 K 35 -50 K 50 -65 K 65 -80 K Income Range ID Analytics Confidential © 2008 80 -95 K 95+K • The shape can be called an “Income Smile” for Identity Risk

U. S. Identity Fraud Hot Spots Updated Findings Data During 2006 ID Analytics Confidential

What’s Happening Recently with ID Fraud? Compare ID Fraud Rates 2003– 2005 to 2006 2003 - 2005 Same • • Southern CA, NV, AZ TX/Mexican border Seattle, Portland NY, Chicago, Springfield, Miami, Atlanta, Detroit, Memphis, Denver… ID Analytics Confidential © 2008 2006 Changing Very Different • Northern CA, NV, UT, OR, TX, FL, LA, MS. . . • MT, ND, SD, MN, WI, ME, ID • Generally, the South is getting better • The upper Midwest is getting worse

Emerging ID Fraud Hotspots in 2006 *These regions were bad in the first half

Emerging Identity Fraud Hot Spots % INCREASE IN FRAUD RATE DURING 2006 FRAUD RATE TIMES AVERAGE Divide, Mc. Kenzie and Williams, ND 10025% 2. 6 586 Adams, Billings, Bowman, Dunn, Golden Valley, Hettinger, Mc. Kenzie, Morton, Slope and Stark, ND 3584% 9. 4 598 Granite, Lake, Missoula, Mineral, Powell, Ravalli and Sanders, MT 2534% 12. 9 584 Barnes, Burleigh, Dickey, Foster, Griggs, Kidder, Lamoure, Logan, Mc. Intosh, Sheridan, Stutsman and Wells, ND 2145% 10. 5 599 Flathead, Lake and Lincoln, MT 1663% 12. 9 585 Burleigh, Emmons, Grant, Logan, Oliver, Mc. Lean, Mercer, Mc. Intosh, Morton and Sioux, ND 1531% 13. 1 582 Cavalier, Grand Forks, Nelson, Pembina, Steele, Traill and Walsh, ND 1228% 4. 4 583 Benson, Bottineau, Cavalier, Eddy, Nelson, Pierce, Ramsey, Rolette, Towner and Wells, ND 1006% 3. 7 587 Bottineau, Burke, Divide, Mc. Henry, Mc. Kenzie, Mc. Lean, Mountrail, Renville, Sheridan, Ward and Williams, ND 791% 2. 9 597 Beaverhead, Deer Lodge, Gallatin, Jefferson, Madison, Powell and Silver Bow, MT 350% 6. 5 573 Aurora, Beadle, Bon Homme, Brule, Buffalo, Charles Mix, Clark, Davison, Douglas, Gregory, Hand, Hanson, Hutchinson, Hyde, Jerauld, Kingsbury, Lyman, Mc. Cook, Miner and Sanborn, SD 344% 1. 4 625 -627 Cass, Christian, Cook, Logan, Macoupin, Menard, Montgomery, Morgan, Peoria, Sangamon, Schuyler, Scott and Shelby, IL 284% 3. 7 570 -571 Bon Homme, Brookings, Clay, Hutchinson, Kingsbury, Lake, Lincoln, Mc. Cook, Minnehaha, Moody, Todd, Turner, Union and Yankton, SD 283% 1. 6 999 Ketchikan Gateway, Prince Wales Ketchikan and Wrangell Petersburg, AK 184% 2. 2 884 Guadalupe, Quay, San Miguel and Union, NM 145% 1. 6 3 -DIGIT ZIP CODE COUNTY NAMES 588 ID Analytics Confidential © 2008

Cities with Increasing ID Fraud Problems Zip City 62707 Springfield, IL 59715 Bozeman, MT 59804 Missoula, MT 59803 Missoula, MT 59937 Whitefish, MT 59847 Lolo, MT 58504 Bismarck, ND 59840 Hamilton, MT 59911 Bigfork, MT 59718 Bozeman, MT 58201 Grand Forks, ND 59808 Missoula, MT 58104 Fargo, ND 59802 Missoula, MT 59901 Kalispell, MT 58501 Bismarck, ND DET 48215 Detroit, MI 59860 Polson, MT 58701 Minot, ND 58301 Devils Lake, ND 57108 Sioux Falls, SD 58601 Dickinson, ND 58401 Jamestown, ND 58554 Mandan, ND 59870 Stevensville, MT DET 48226 Detroit, MI NY 11210 Brooklyn, NY CHI 60093 Winnetka, IL 59912 Columbia Falls, MT 11225 Brooklyn, IDNY Analytics Confidential © 2008 NY % Increase in Fraud Rate During 2006 Fraud Rate Times Average 5515% 2495% 2480% 2297% 1785% 1782% 1764% 1707% 1704% 1456% 1454% 1443% 1395% 1383% 1319% 1288% 1155% 1153% 1143% 1063% 1021% 988% 979% 967% 923% 761% 732% 693% 54. 52 11. 46 11. 49 10. 62 10. 02 7. 90 7. 70 8. 88 7. 57 8. 05 7. 04 6. 29 6. 83 6. 63 6. 71 7. 02 8. 75 6. 06 5. 35 6. 00 5. 61 4. 57 4. 56 4. 82 3. 95 5. 15 8. 64 4. 58 4. 27 6. 22 Zip City 59714 57105 92663 00627 38614 06830 59801 48202 57103 58703 58203 77024 48219 46619 77005 48210 11429 33162 30344 91776 33068 48340 91915 95348 19126 30260 90505 46201 77004 06114 Belgrade, MT Sioux Falls, SD Newport Beach, CA Camuy, PR Clarksdale, MS Greenwich, CT Missoula, MT Detroit, MI Sioux Falls, SD Minot, ND Grand Forks, ND Houston, TX Detroit, MI South Bend, IN Houston, TX Detroit, MI Queens Village, NY Miami, FL Atlanta, GA San Gabriel, CA Pompano Beach, FL Pontiac, MI Chula Vista, CA Merced, CA Philadelphia, PA Morrow, GA Torrance, CA Indianapolis, IN Houston, TX Hartford, CT % Increase in Fraud Rate During 2006 Fraud Rate Times Average 652% 651% 544% 536% 472% 450% 445% 379% 378% 375% 316% 298% 285% 168% 167% 201% 186% 168% 167% 158% 157% 156% 152% 146% 140% 126% 122% 121% 116% 115% 3. 48 3. 52 4. 34 5. 60 2. 93 2. 91 2. 34 4. 51 2. 30 2. 31 2. 23 2. 14 3. 64 2. 41 2. 54 2. 80 2. 49 2. 05 1. 84 2. 80 1. 62 1. 84 1. 97 1. 85 2. 02 1. 76 1. 59 1. 74 1. 60 1. 59 DET HOU DET NY MIA ATL LA MIA DET ATL LA HOU

California ID Fraud Hotspots High fraud rates in 2006 ZIP 90024 93616 92055 95258 92663 90007 93615 90293 94801 90010 91011 92254 93258 94572 90036 92843 90020 90077 90211 92173 90033 93219 93647 90061 91602 92704 92683 93458 93221 90701 ID Analytics Confidential © 2008 City Los Angeles Del Rey Camp Pendleton Woodbridge Newport Beach Los Angeles Cutler Playa Del Rey Richmond Los Angeles La Canada Flintridge Mecca Porterville Rodeo Los Angeles Garden Grove Los Angeles Beverly Hills San Ysidro Los Angeles Earlimart Orosi Los Angeles North Hollywood Santa Ana Westminster Santa Maria Exeter Artesia 2006 Fraud Rate Times Average 8. 44 6. 94 5. 23 4. 87 4. 34 4. 30 4. 23 4. 22 3. 75 3. 71 3. 66 3. 63 3. 62 3. 50 3. 36 3. 28 3. 27 3. 20 3. 06 2. 89 2. 78 2. 77 2. 76 2. 74 2. 72 2. 70 2. 69 2. 66 2. 65 ZIP 94804 90003 93662 92706 90058 93250 95030 90240 95334 90006 90005 90044 95330 92840 90040 92701 91403 90065 90004 90057 90069 93215 90602 90045 90042 91108 94535 93204 92325 93223 City Richmond Los Angeles Selma Santa Ana Los Angeles Mc Farland Los Gatos Downey Livingston Los Angeles Lathrop Garden Grove Los Angeles Santa Ana Sherman Oaks Los Angeles West Hollywood Delano Whittier Los Angeles San Marino Travis AFB Avenal Crestline Farmersville 2006 Fraud Rate Times Average 2. 64 2. 63 2. 60 2. 57 2. 56 2. 55 2. 54 2. 53 2. 50 2. 49 2. 45 2. 44 2. 42 2. 41 2. 40 2. 39 2. 38

Emerging ID Fraud Hotspots in CA Both high AND substantially rising fraud rates in 2006 % Increase in 2006 Fraud Rate in ‘ 06 Times Average CA ZIP Name 93926 Gonzales 766% 95030 Los Gatos 92663 % Increase in 2006 Fraud Rate in ‘ 06 Times Average CA ZIP Name 1. 87 90670 Santa Fe Springs 90% 1. 78 274% 2. 55 92833 Fullerton 89% 1. 63 Newport Beach 268% 4. 34 91978 Spring Valley 86% 1. 55 92832 Fullerton 202% 1. 53 90056 Los Angeles 82% 1. 65 90248 Gardena 201% 2. 37 91763 Montclair 82% 1. 73 95694 Winters 169% 1. 96 95386 Waterford 79% 1. 70 95388 Winton 155% 1. 68 91301 Agoura Hills 78% 1. 51 91377 Oak Park 151% 1. 79 90230 Culver City 78% 1. 75 92868 Orange 133% 1. 65 95678 Roseville 77% 1. 51 94612 Oakland 128% 1. 51 95357 Modesto 69% 2. 14 95348 Merced 120% 1. 85 90620 Buena Park 68% 1. 51 91776 San Gabriel 119% 2. 01 91321 Newhall 68% 1. 63 91915 Chula Vista 116% 1. 97 96150 South Lake Tahoe 66% 1. 73 90505 Torrance 115% 1. 59 94506 Danville 65% 2. 09 94065 Redwood City 113% 1. 51 94502 Alameda 63% 2. 12 92603 Irvine 105% 1. 51 92866 Orange 58% 1. 66 91108 San Marino 102% 2. 40 93647 Orosi 55% 2. 76 94535 Travis AFB 99% 2. 39 95815 Sacramento 55% 1. 67 92602 Irvine 96% 1. 56 90701 Artesia 54% 2. 65 91104 Pasadena 91% 1. 64 92254 Mecca 52% 3. 63 ID Analytics Confidential © 2008

SPRINGFIELD, ILLINOIS Springfield Synthetic Fraud Ring • 936 fraudulent wireless applications originating from Springfield, IL over the course of four years - 70% of the applications occurred between July and November of 2006 - 921 unique SSNs used - 641 unique addresses used • Fraudsters targeting cell phone hardware - All events were post-book verified frauds • Plausible stories - Fraud perpetrated through post office? - The addresses make “loops” through the communities, implying fraudsters stealing mail from physical addresses? ID Analytics Confidential © 2008

SPRINGFIELD, ILLINOIS “Loop” “Loops” ID Analytics Confidential © 2007 ID Analytics Confidential © 2008

SPRINGFIELD, ILLINOIS ID 1 is listed under the home phone at the correct address.

SPRINGFIELD, ILLINOIS Steady fraud activity in 2003 & 2004 Break in activity in early 2005 Significant activity begins late 2005 Predominately wireless application fraud ID Analytics Confidential © 2008 Extremely high velocity in 2006

SPRINGFIELD, ILLINOIS Many Synthetic Frauds in Springfield, IL Synthetic Identities – SSN Overlap LEGEND Name SSN Address Synthetic ID overlap with valid ID Phone (Cell) Address Phone Date of Birth Synthetic Element Synthetic SSN overlap ID Analytics Confidential © 2008 38

MONTANA & NORTH DAKOTA MT & ND Fraud Population Demonstrate Identity Theft Ring Behavior • 99% of the applications are credit cards through online channel (95% targeted a single issuer) • Fraud population contains what appears to be valid identity information, but altered DOBs - Real people’s identities stolen and one or two additional applications are fraudulently added - Fraudsters do not have DOBs but are able to manufacture a valid one using SSN • Low application velocity per identity - Majority of identities within the fraud population have only a few applications each - However, when reviewed as a group, extremely high velocity ID Analytics Confidential © 2008

MONTANA & NORTH DAKOTA Example Application Velocity Per Identity DB BB HB Flurry of fraudulent applications LB Two different zip codes CC KK SA MK JB SE JG MD 02/06 04/06 Bank Card Application ID Analytics Confidential © 2008 06/06 08/06 Wireless Application 10/06 12/06 Confirmed Fraud within the ID Network

Zip Code 3 Zip Code 2 Zip Code 1 MONTANA & NORTH DAKOTA DB BB HB LB CC KK AT BB JD GB LT PR RN RF BF CR MN SC TS JB VG TH DH BH LD AA ST AH CS KR BC PP PD DS GT RR LO EZ Broader View of Velocity • Fraud population demonstrates a strong degree of “flurrying” • Flurries correlate with ZIP Code change • Fraudsters tend to target bank cards due to higher immediate value Bank Card Application Wireless Application Confirmed Fraud within the ID Network TRUE NAME ID THEFT RING 01/06 ID Analytics Confidential © 2008 06/06 12/06

MONTANA & NORTH DAKOTA Fraud Ring Migration in MT & ND • Fraud ring focuses on one area, then relocates to a new ZIP Code to avoid detection • Ring neighborhood size suggests intercepting mail • Started in July 2006 • August 2006: ring activity shifts between states, moving from MT to ND ID Analytics Confidential © 2008

MONTANA & NORTH DAKOTA ID Fraud Ring Movement Throughout 2006 July 30 Canada Sept

MONTANA & NORTH DAKOTA Activity shifts from MT to ND ID Analytics Confidential ©

MONTANA & NORTH DAKOTA Comments for MT & ND • Fraud ring likely responsible for fraud increase in MT & ND in 2006 • Fraudsters targeting one credit issuer via online channel • Plausible methods of acquiring cards: - Stealing cards from mailboxes - Cards intercepted before being sent out • Lingering questions - Why target one particular credit issuer? - How did the ring gain access to identity information for 1, 700 individuals? ID Analytics Confidential © 2008

Executive Summary • Identity fraud is accessible, prevalent and damaging • Two types of identity fraud: true-name and synthetic • We have identified some consistently risky: - States (NY, NV, CA, MI, AZ, IL, TX, …) Metropolitan areas (New York, Detroit, Los Angeles, …) ZIP Codes (Queens, Manhattan, Chicago, Los Angeles, …) Other areas: Southern CA, the Mexican border of Texas and in cities like Seattle and Portland • Identity fraud risk increases strongly with population density • In general, identity fraud rates are - Increasing in the upper Midwest, Northern CA, UT, NV and ME - Decreasing in the Southern United States • Have identified emerging hotspots: Western MT, ND, Eastern SD, Springfield, IL, Detroit, Brooklyn, Houston, Miami, Atlanta, Philadelphia, … • Lots of synthetic identity fraud in Springfield, IL • Apparent identity theft ring hotspot in MT and ND ID Analytics Confidential © 2008

For a copy of the “U. S. Identity Fraud Rates by Geography” or “U. S. Identity Fraud Rates Hot Spots” white papers, visit our website at www. idanalytics. com or email Marketing. Info@idanalytics. com ID Analytics Confidential © 2008

Appendix: Several Other Case Studies ID Analytics Confidential © 2008

LEGEND Applicant Name Case Study 1 (671) SSN Name (non-applicant) Address Phone Fraud Application Asserted Info on App Invalid SSN Info on post-event apps ID Analytics Confidential © 2007 ID Analytics Confidential © 2008 Risky behaviors related to the application are demonstrated by multiple and shared Social Security

LEGEND Applicant Name Case Study 2 (726) SSN Name (non-applicant) Address Phone Fraud Application Asserted Info on App 200+ Invalid SSN Info on post-event apps ID Analytics Confidential © 2007 ID Analytics Confidential © 2008 A phone number connected to the applicant’s address is linked to a fraudulent application and over 200

LEGEND Case Study 3 (826) Applicant Name SSN Name (non-applicant) Address Phone Fraud Application Asserted Info on App Invalid SSN Info on post-event apps ID Analytics Confidential © 2007 ID Analytics Confidential © 2008 Multiple unexpected behavior patterns are shown by the many addresses linked

LEGEND Case Study 4 (893) Applicant Name SSN Name (non-applicant) Address Phone Fraud Application Asserted Info on App Invalid SSN Info on post-event apps ID Analytics Confidential © 2007 ID Analytics Confidential © 2008 This applicant’s address has three fraudulent applications associated with