Page 1
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`
`11
`
`12
`13
`
`14
`
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
` __________________
` LIBERTY MUTUAL INSURANCE CO.
` Petitioner
` v.
` PROGRESSIVE CASUALTY INSURANCE CO.
` Patent Owner
` __________________
` Cases CBM2012-00002; CBM2012-00004(JL)
` Patent 6,064,970
`
` Case CBM2013-00004(JL)
` Patent 8,090,598
` Cases CBM2012-00003; CBM2013-00009(JL)
` Patent 8,140,358
`
` __________________
`
` DEPOSITION OF MARY LOU O'NEIL
` Washington, D.C.
` Friday, September 13, 2013
`
`Reported by: John L. Harmonson, RPR
`Job 65806
`
`TSG Reporting - Worldwide 877-702-9580
`
`Liberty Mutual Exhibit 1037
`Liberty Mutual v. Progressive
`CBM2012-00002
`Page 00001
`
`

`

`
`
` _'.__|_._......_......_.
`tomaaoxmewNI—i
`mflamdhmMI—‘OkomqmmuE-WMH
`
`
`
`Page 7 _
`M. O'NEIL
`
`M. O'NEIL
`
`introduced by Mr. Miller.
`Q. Were you familiar with this document
`before it was offered by Progressive in its
`reaponse to the Liberty Mutual petition?
`A. Yes. I've been familiar with this
`
`the -- you know, I guess there are 50 standards
`of practice and several statements of principle,
`daily.
`I mean, it's on an as-needed basis. You
`go back to the document whenever you might want
`to use it.
`
`Q. And in fact, you've used this document
`in your actuarial work in the past?
`A. Yes, I have used it in the past.
`Q. And in fact, in the past has some of
`your actuarial work been conducted in a way
`that's consistent with the statement of
`
`principles set forth in this document?
`MR. MYERS: Objection. 402, 403.
`I believe all of my actuarial work has
`A.
`always been consistent with all of the standards
`of practice, which includes the statements of
`principles.
`,
`Q. And by that, you mean this statement
`of principles that is Progressive's Exhibit 2012‘?
`A. Well, there are several statements of
`
`
`
`I
`document since its initial publication.
`believe it was back in the 19803, maybe 1982 or
`Somewhere in that range.
`Q.
`So it's a dowment you're well
`acquainted with; is that right?
`A.
`It's not something --
`MR. MYERS: ObjectiOn. 402, 403.
`I'm familiar with the document as any
`A.
`other professional actuary might be.
`Q.
`It's not a document that would be
`obscured to actuaries practicing in their field;
`is that right?
`MR. MYERS: Objection. Federal Rule
`P9
`of Evidence 402, 403.
`I'm going to give the
`20
`rule number as we go forward, but I'm
`21
`principles attached to the standards of practice.
`referring to the Federal Rules of Evidence.
`:22
`This is one of them. So I included everything.
`A.
`I‘m not Sure what you meant by
`P3
`
`
`"obscure." We don‘t -- at least I as a
`.4
`Q. Okay. Let me now hand you another
`
`
`
`
`exhibit, Liberty Mutual Exhibit 1022 in case
`
`
`
`Page 9
`
`
`M. O‘NEIL
`
`A. This paragraph mentions something I
`]
`
`
`did as a consultant to the New Jersey Market
`1
`j
`Transition Facility.
`Q. And what you did was help to develop
`f
`an actuarial class system using driving record
`1
`'
`points; is that right?
`A. Correct.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`M. O'NEIL
`
`CBM2012-2.
`(Liberty Mutual Exhibit 1022, having
`been marked for identification, is attached
`hereto.)
`Q. Can you identify that document,
`Ms. O'Neil?
`
`Is
`
`A. That is correct. That is what the
`
`A. Do you wish for me to read the entire
`description on the cover?
`Q.
`Ifthat‘s how you would like to
`identify it.
`Let me ask a different question.
`this your rebuttal declaration in the
`CBM2012-2 case?
`identifying caption says.
`Q. And you recognize it as such, correct?
`A. Yes.
`Q. Okay. Now, I w0uld ask you to turn to
`Paragraph 46. Are you there?
`A. Yes.
`
`LIIwaHommqmmmewommummewNI—I
`
`E
`[l
`E
`El.
`3.
`it
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Q. And when you did that, was your work
`consistent with the risk classification statement
`
`Q.
`
`
`
`
`of principles that's Progressive's Exhibit 2012?
`MR. MYERS: Objection. 402, 403.
`A. Yes. As I said, my work has always
`been consistent with the statement of principles.
`Q.
`So in your experience, yOu have always
`adhered to this statement of principles whenever
`you have provided your professional services as
`an actuary?
`MR. MYERS: Objection.
`Is that right?
`MR. MYERS: Objection. 402, 403.
`I wouldn't use that terminology, that
`A.
`Q. Okay. In this paragraph you're
`terminology of "adhered.“ As I’ve explained
`providing testimony about helping to develop an
`other places in my declaration, these statements
`actuarial class system in New Jersey, correct?
`of rinci les and the standards of aractice are
`MR. MYERS: Ob'ection. 402 403.
`
`
`
`
`
`
`
`TSG Reporting — Worldwide
`
`877—?02—9580
`
`3
`
`(Pages 6 to 9)
`
`Page 00002
`
`Page 00002
`
`

`

`
`
`Page 11 -
`
`Page 10
`
`l
`
`M. O'NEIL
`
`l
`
`M. O'NEIL
`
`
`
`
`
`
`
`
`
`
`something, I would be aware of all ofthose
`things. Now, if it turns out for some reason I
`may not have appropriate data or the problem, if
`it doesn't fit with something here, I was
`certainly aware of this, and to the extent
`possible would certainly consider it. But if
`something that I might need to do differs from
`this, there could be a situation where there is a
`law or regulation, something like this, I will do
`what I believe is the best job and then document
`what I have done.
`Q. That was a pretty long answer.
`trying to ask a really simple questiOn.
`Can you tell me whether you've ever
`departed or deviated from the statement of
`principles that's set forth in Exhibit 2012‘?
`MR. MYERS: Objection. 402, 403.
`
`I‘m
`
`2
`3
`4
`5
`6
`7'
`8
`9
`0
`1
`2
`ttB
`14
`it 5
`P6
`7'
`E
`8
`
`
`
`90 1
`
`2 3 4 5
`
`guidelines for actuaries in their professional
`practice. To the extent that someone might
`deviate from them for some reason or other, that
`is to be documented.
`However, I do believe I also quoted
`somewhere that it even says that the professional
`judgment of the actuary in the end is the most
`important factor. But nonetheless, these are all
`to be treated as considerations and guidelines,
`not rules that must be strictly followed. So the
`word "adhere" sort of brings that to mind.
`Q. Have you ever intentionally deviated
`from any ofthe principles set forth in the
`statement of principles that's Exhibit 2012?
`MR. MYERS: Objection. 402, 403.
`I guess I wasn't clear perhaps in my
`last answer. These are guidelines. So all of
`
`A.
`
`I guess I wouldn't call it deviated.
`A.
`the standards of practices are guidelines. In
`I would call it I've documented my work and
`addition to these guidelines, there's a body of
`considered these as required in all instances.
`literature to guide actuaries, professional
`
`Q. You're aware, Ms. O'Neil, that this
`experience. There is literature that's published
`
`
`statement of principles set forth certain
`outside of the actuarial profession. There are
`
`
`
`statistical considerations to be considered,
`many sources ofinformation.
`including homo eneity, credibility, and
`
`
`M. O'NEIL
`1
`M. O'NEIL
`
`
`MR. MYERS: Objection. 402, 403.
`2
`predictive stability?
`
`A. The data which I used were the onIy
`3
`MR. MYERS: Objection.
`data available. And so I'm not sure how to
`4
`Is that correct?
`5
`MR. MYERS: Objection. 402, 403.
`6
`I believe you're referring to Section
`A.
`7
`4wtsuMMnsL2,mfl3,whflehhmsmom
`8
`items you just mentioned, homogeneity,
`9
`credibility and predictive stability.
`I O
`Q. And based on your past work, you're
`1 l
`familiar with those considerations; am I right?
`
`I 2
`A. That is correct, I am familiar with
`
`I 3
`these.
`
`I 4
`
`I 5
`
`I 6
`
`| 7'
`
`i 8
`
`9
`
`' 0
`
`' 1
`
`' 2
`
`’ 3
`4
`' 5
`
`2
`3
`4
`5
`6
`7
`8
`9
`I O
`I 1
`I 2
`
`I 3
`
`I 4
`I 5
`I 6
`
`I 7’
`I 8
`
`
`I 9
`' 0
`' 1
`
`' 2
`
`' 3
`
`
`
`
`
`
`
`
`
`
`
`Q.
`
`I'll ask you to go back to your
`Q.
`statements in Paragraph 46 of your rebuttal
`declaration. Did you consider those statistical
`considerations in the work you did to help
`develop an actuarial class system using driving
`record points in New Jersey?
`A. Of course I considered them. And then
`
`once I considered those, then I looked at what
`there was available to work with, and those are
`
`the data that were used to put forth this plan.
`Q.
`In your opinion, was the data you used
`homo eneous?
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(II-l:LDMHOKOODHJONLNJBUJNl—‘OKDCDHJGXMJBUJMH
`
`answer the question homogeneous in this case. A
`lot has been made of the term homogeneous, and I
`guess I'm not sure what further to say about that
`regarding this particular situation. The goal of
`that particular analysis was to generate revenue
`based on driving record points which related to
`tickets or accidents. Then these were the points
`that would be on someone's DMV record.
`
`The state would operate the plan, the
`state would collect the money, and the state
`would remit the money over to the New Jersey
`Market Transition Facility, which was the
`residual market mechanism in New Jersey. The
`rates being charged by the New Jersey MTF which
`I‘ll call it were inadequate for the losses that
`that organization or entity was incurring.
`So to supplement that organization,
`someone, I'm not sure who, decided this was one
`way to get revenue under the political theory
`that let the bad drivers pay, which means anybody
`who, in that definition, was someone who had
`
`
`
`T86 Reporting — Worldwide
`
`877—702e9580
`
`4
`
`(Pages 10 to 13)
`
`Page00003
`
`Page 00003
`
`

`

`M. O‘NEIL
`
`M. O'NEIL
`
`Page 19
`
`
`
`L0
`
`to
`
`121
`22
`23
`24
`
`kOCDflO'tU'Idh-UJNl—l
`
`I guess I don't understand that
`A.
`question, because ifthey're in conflict with it,
`they can't be part of the generally accepted
`principles and practices.
`I would think the body
`of it is fairly cohesive.
`Q.
`I would think so too.
`So all the other sources of generally
`accepted actuarial principles and practices, to
`the extent they relate to this same subject
`matter as Exhibit 2012, would be consistent with
`it; is that correct?
`MR. MYERS: Objection. 402, 403.
`1 don't know if] would put it exactly
`A.
`that way. Part ofthe thing about actuarial
`science is that it's not an exact science.
`
`Everything that has been written in various
`papers, including this, has been written by
`people. And it's not like mathematics where two
`times two is four. This is in the bestjudgment
`of the pe0p1e who wrote it at the time.
`The same thing with various papers in
`the literature.
`I would assume that a
`
`statistician or even another actuary may not
`
`necessarily, without this in front of them, write
`Page 20
`
`M. UNI-311..
`
`of classifications and so on which might expand
`upon the concepts as they relate to
`classification. This is more simplistic than a
`lot of things one might read on the subject.
`Q. Are you aware of any other guidelines
`that conflict with the Exhibit 2012 statement of
`
`principles?
`MR. MYERS: Objection. 402, 403.
`A. These are the only guidelines per se
`in terms of something that's been set forth as
`part of the standards of practice. The rest of
`the body of generally accepted principles and
`practices would come from professional writings,
`publications, the body of actuarial literature.
`That‘s what I'm referring to that would expand
`upon this and actually be more technical than
`this. You would find that in other writings,
`which would become part of generally accepted
`actuarial principles and practices as opposed to
`something that is actually part of the standards
`of practice.
`Q. Are there other standards of practice
`that are in conflict with the statement of
`unrinciles in Exhibit 2012?
`
`{DODmJQLflJE-LAJNF‘
`acumen-Eton)
`
`I—lO
`
`“3...:
`
`PHnnwrmrp-p
`
`25
`01.x:wmpommummhwwi—aoquoxmuzme
`
`It doesn't
`the same definition for homogeneity.
`make it wrong; it makes it different. So I would
`think that if we did some research, we would find
`other criteria that people have put forward
`related to classifications, not making any one of
`them more wrong or right, it becomes part of the
`total body of knowledge of an actuary.
`So I really can't — I can't really
`agree to what you're saying, that every document
`we look at is going to have the same thing in it.
`1 don't think so.
`
`In fact, hasn't this statement of
`Q.
`principles in Exhibit 21 12 —- or excuse me, 2012
`been widely accepted by actuaries practicing in
`the field?
`
`MR. MYERS: Objection. 402, 403.
`I really wouldn't know what "widely
`A.
`accepted" means.
`I know that it's part of the
`Standard of practice 12., which is part of the
`entire set of standards of practice.
`I'm not
`sure what the total number is at this time.
`
`Nevertheless, there are other papers and so on
`that have been written related to
`
`classifications, related to_statistical_ analyses
`Page 21
`
`M. O'NEIL
`
`MR. MYERS: Objection. 402, 403.
`A. This is the only standard of practice
`that deals with classifications.
`
`Q. Okay. In your opinion, has this
`statement of principles been relied upon by
`actuaries on matters that it's germane to?
`MR. MYERS: Objection. 402,403.
`A. This sounds like the same question you
`asked me before, only instead of "adhered" you've
`now said "relied upon." My answer is the same.
`It's been considered as appropriate along with
`the rest of the body of actuarial literature.
`Q. Have you ever publicly criticized any
`of these statement of principles?
`I may
`A.
`I do not recall having done so.
`have objected to someone perhaps stating that
`something should be adhered to or something is
`set in stone type of rule. But I don't recall
`necessarily criticizing.
`I don‘t recall doing
`that. Anything is possible, I suppose.
`Q. Let me refer you to Paragraph 7 of
`your rebuttal declaration, Exhibit 1022. Feel
`free to refresh your memory by looking at that
`uara- ra h, or the surroundin_
`'
`
`TSG Reporting ~ Worldwide
`
`6
`877—702—9580
`
`(Pages 18 to 21)
`
`Page00004
`
`Page 00004
`
`

`

`M. O'N EIL
`
`CD-aJO‘nU'IIbUJMH
`
`could be off. but my recollection is I might have
`used it in an insolvency case I was working on.
`In such situations, once again, it is necessary
`to look and see what was in place at the time
`something was happening. So it was necessary to
`look back and see what opinions or standards were
`in place when actuaries completed a reserve
`analysis. And 1 was the one reviewing what other
`actuaries had done and if indeed they had
`followed the standards.
`
`So in your prior use ofthe opinions
`Q.
`in Exhibit 1023, you didn't rely on them as a
`basis for departing or deviating from the
`actuarial statement of principles?
`MR. MYERS: Objection.
`Q. Am I right?
`MR. MYERS: Objection. 402, 403.
`A. My recollection is that is also
`possible. But my recollection of the specific
`use relates to using it to see if someone else
`had done so.
`
`\Dmflmifi-file—i
`'l—‘I.'.““'“H01.01wa
`
`
`
`1—!O
`
`(Du-J
`{L9
`
`E:
`3
`£4
`
`
`
`
`
`
`l—tI"aommqmmbwmwommqmmewmw
`
`Page 26:
`
`Page 27
`
`M. O'NEIL
`the declaration of Mr. Miller so that it will
`
`make sense to you.
`So I'll hand you what's been marked as
`Progressive Exhibit 2010 in CBM2012—2.
`(Progressive Exhibit 2010, having been
`marked for identification, is attached
`hereto.)
`Q. Do you recognize that as the
`declaration of Michael Miller to which you're
`referring in your rebuttal declaration from time
`to time?
`A.
`
`I believe so.
`
`Q. Okay.
`A.
`I see that this cite also refers to
`
`Standard Number 12. Would you happen to have
`that as well or do we need that?
`
`Q. We'll get to it, I think.
`A. All right.
`Q.
`I mean, if you want it, I can supply
`it. You‘ll just tell me when I ask the question.
`How's that?
`
`A. Okay.
`So let me ask you to turn back to your
`Q.
`rebuttal declaration that's Exhibit 1022 and look
`Q. My question relates to really the
`
`25
`at Paragraph 8. And_1_think_1 b_e_tter show you
`Page 28
`
`Page 29
`
`M. O'NEIL
`
`M. O'NEIL
`
`the rebuttal declaration where you refer to --
`Well, let me just quote:
`Indeed, one such standard of practice,
`Number 12, attached to Mr. Miller's supplemental
`declaration, Exhibit 2020, further belies
`Mr. Miller‘s strict adherence to the statement of
`
`principles and repeated insistence on the use of
`actual claims data to generate actuarial classes.
`Do you see that?
`A. Yes.
`
`Q. With reference to the phrase "repeated
`insistence on the use of actual claims data,"
`that you attribute to Mr. Miller, that's not
`really true, is it?
`I mean, if you
`A.
`I wouldn't say that.
`wish to go through all the instances to -- I
`mean, it's up to you.
`Q. Well, do you want to look at
`Paragraph 16, which is the first paragraph that
`you cite there? Paragraph 16 of Mr. Miller's
`declaration is what I was referring to.
`Is there something in there where you
`find Mr. Miller insisting on the use of actual
`claims data?
`
`
`
`,_.,l._+.1._.‘
`
`A. Yes. He says here in Paragraph 16:
`An actuarial class or risk class is a
`
`grouping of risks with similar risk
`characteristics and expected insurance claims
`loss. Expected insurance claims loss is actual
`claims there.
`He comes down to --
`
`I'm sorry, where are you?
`Q.
`Paragraph 16.
`A.
`16.
`Q.
`A. Let's see. He comes down to:
`Actuarial claims data is used to
`
`determine expected insurance claims less.
`So it's mentioned twice there.
`
`So where do you find the words "actual
`Q.
`claims data" there?
`
`It doesn't use the word "actual," but
`A.
`that's what it means.
`
`Q. That's your gloss on it; is that
`right?
`A. Excuse me?
`
`A.
`
`MR. MYERS: Objection. 402, 403.
`I'm not sure what that meant.
`Well Mr. Miller doesn't sa actual
`
`TSG Reporting — Worldwide
`
`arr—702w9580
`
`8
`
`(Pages 26 to 29)
`
`Page 00005
`
`Page 00005
`
`

`

`Page 30
`
`Page 31
`
`U'IJhLAJMI—ioiommetflanNl—‘DkomhlmmvwaI—l
`
`
`
`1-1-1“-
`
`M. O'NEIL
`estimated as we]! as taken from actual historical
`
`results. Am I right?
`A.
`Indeed, l‘m not disagreeing with you.
`l'm saying that expected claims loss still would
`need to be based on the company's own data based
`on what Mr. Miller has said here, is what l'm
`saying.
`l'm saying in my thing that Mr. Miller
`has not allowed for the fact that data could be
`from some other outside source.
`
`Q. Nonetheless, he does not say "actual
`claims data" in that paragraph?
`A. He does not use that exact word. That
`is -- That is what I read whenever I read his
`
`Okflm‘JOhLfi-LNWNI—‘OQCDQGWLb-UJNH
`
`p...-
`
`Page 33
`
`M. O'NEIL
`
`So you don't read his entire
`Q.
`declaration as being consistent with his opinions
`here?
`
`A. This statement is in a particular
`section of the declaration. The other -- I
`
`assume that this definition that we just read in
`16 was meant to stand alone.
`
`Q. Oh, I see.
`Isn't it cOrrect that the statement of
`
`principles that Mr. Miller cites in his
`declaration allows for using data other than
`actual claims loss data?
`A. We should check. Do we have a
`
`specific spot in here?
`I wouldn't want
`Q. You're the actuary.
`to point you to the wrong section, Ms. O'Neil.
`But l‘m referring to Exhibit 2012.
`A.
`l'm not finding it right now, but I
`suspect that -- l'm not sure that it actually
`addresses the exact source of the data. Because,
`
`once again, these are guidelines. They're not
`meant to be recipe books.
`And Ijust note one more thing
`re ardin Mr. Miller's sentence that on ointcd
`
`testimony. He said expected insurance claims
`loss. That implies or basically is a statement
`that you would use the company's own expected
`insurance claims loss. Expected is estimated,
`basically.
`So when I say "actual" here, it refers
`to within the company as opposed to from some
`other source, which is what this paragraph talks
`2
`about. is that you can use data from other
`Q 3
`sources, you could use industry data, you could
`24
`
`?5
`use data from another company it‘it were
`Page 32:
`
`‘:-.
`wmwmrr-mn":
`
`
`
`mammawwp
`
`OOMJChU'lJb-WNI-J
`
`M. O'NEIL
`
`In fact, he says actuarial class
`claims data.
`claims data, correct? Which could include other
`
`types of claims data besides actual claims data,
`correct?
`
`A. No. He says expected insurance claims
`loss, which is actual claims data.
`It doesn't
`
`say actuarial.
`Q.
`So all expected claims loss data has
`to be actual? It can't be estimated?
`
`A. Even ifit is estimated, it's still --
`
`you're mistaking the word "actual" meaning of a
`company as opposed to from some other source.
`Q. What did you mean by "actual"?
`A.
`I meant ofa company as opposed to
`some other source.
`
`Q. You meant their own actual historical
`experience; is that right?
`A. Correct.
`I meant the company's own
`experienCe as opposed to some other data from
`some other source.
`
`Q. But you don't find that in
`Paragraph 16, do you?
`A.
`I believe I do.
`
`Q. Expected claimfls loss could be
`
`M. O'NEIL
`available. You could use data from -- there are
`
`other collection places that you can get data
`from. And that's what's not mentioned here.
`
`Q. Mr. Miller doesn't say that you have
`to use data from a particular company, does be?
`A. The way this is written, a POSITA
`would understand that he's saying that it‘s used
`from the company that is setting up the class.
`0.
`Let me direct your attention to
`Paragraph 29 --
`A. Okay.
`Q.
`-- of Exhibit 2010. Do you see the
`first Sentence of that paragraph, Ms. O'Neil?
`A. Yes, 1 see that.
`Q. Okay. And there Mr. Miller indicates
`that other considerations can be taken into
`
`account, including the experience of other rate
`filers, business judgment, and all other relevant
`information and data within and outside the
`state.
`
`Do you see that?
`A. Yes, I see that he wrote that there.
`Nonetheless, he didn't allow for that in these
`.other niaces.
`
`T86 Reporting — Worldwide
`
`8T7—7D2—9580
`
`9
`
`(Pages 30 to 33)
`
`Page 00006
`
`Page 00006
`
`

`

`M. O'NEIL
`
`Page 34
`
`ODHJQU'It-D-LAJNl—l
`
`me to. This particular sentence, if you look at
`it carefully, says:
`Demonstration that an insurer's rates
`
`comply with the statutory rate standards
`typically involves consideration of past and
`prospective loss experience of the insured, the
`experience of other rate filers' business
`judgments, and all other relevant information and
`data within and outside the state.
`
`That statement, that sentence, refers
`to the overall rate level. Because if you read
`the next sentences, it speaks about
`classifications. Mr. Miller is not permitting
`these types of data -- at this time, at least,
`suggesting these types of data would apply to
`classifications.
`The next sentences refer to
`classifications.
`
`Q. That's how you read it, at least?
`A. Well, that's a standard way one would
`state. Because he says insurer's rates. That
`refers to a rate filing, has an overall rate
`level within it, and that's what you would use.
`l__ The data that he's listed here are what are put
`Page 36
`
`M. O'NEIL
`
`rate standards," are you excluding the standards
`that relate to rates not being unfairly
`discriminatory?
`A. No, l'm not. As a regulator,
`obviously you want the rates to be not
`inadequate, not excessive and not unfairly
`discriminatory. The question as a regulator is:
`How often must I look at the rates to determine
`
`compliance?
`Most companies revise their base rates
`frequently as it relates to economic trends and
`so on. So they may want to make a rate filing
`with an insurance regulator every six months. At
`the same time, everyone recognizes, the regulator
`and the company, that classification
`differentials, or whatever you might want to call
`it, classifications don't change all that
`frequently. And so that may not be reviewed by
`the company or by the regulator more than once a
`year or once every several years. So that's why
`it's oftentimes a separate filing.
`And so it's assumed that as long as
`that filing is in place, there has been a filing
`to show that the rates are not unfair]
`
`
`
`M. O'NEIL
`forth for the actual base rate. Then the
`
`classification plan follows from that, which is
`what he has got in the next sentence.
`Q.
`In fact, isn't the insurer required to
`justify that its rates are fair and not unfairly
`discriminatory?
`A. All right, that is one ofthe
`In order to
`ratemaking standards in the statute.
`satisfy that demonstration, first the overall
`rate level is in the filing. Then the regulator
`would look at classifications.
`
`Many times, however, classification
`filing is a separate filing. It's not part of
`the overall rate level filing because of its
`additional complexity. And generally the
`classification filing is more complex and
`therefore takes more time and it's not done as
`
`frequently as the update of the overall rate
`level.
`
`So yes, they would have to demonstrate
`the classifications as well and not unfairly
`discriminatory, but not as frequently.
`Q. When you read the words "demonstration
`that an insurer's rates comply with the statutory
`
`Page 3?
`
`M. O'NEIL
`
`discriminatory, you don‘t have to revise it as
`often as you do the rates that are based on
`trends.
`
`Q. My question is: Doesn't demonstration
`that an insurer's rates comply with the statutory
`rate standards refer not just to the base rates
`but also to the classifications that the insurer
`
`is proposing to use?
`MR. MYERS: Objection. 402, 403.
`I believe I said that compliance is
`A.
`required with all aspects of the rating law.
`However, the actual implementation on the part of
`the regulator of measuring that compliance can be
`done in a distinct fashion with separate rate
`filings for different aSpects of that compliance.
`Q. Do you think Mr. Miller here in
`Paragraph 29 is insisting that the insurer must
`rely on actual claims data in justifying
`actuarial classes?
`
`MR. MYERS: Objection. 402, 403.
`A. That's what he says. If you look at
`the second sentence, he says:
`Any claims data gathered by the
`insurer via its risk classification Ilan, when
`
`wlbUJNl—‘OKDCD«Jmmb‘hUJMl—‘OKOGDHJONU'ILEUJMH
`LflubbJNl—‘QkomflmtflbWNl—‘Ommqmmtt‘awNI—l
`
`TSG Reporting — Worldwide
`
`10 (Pages 34 to 3?)
`877—?02—9580
`
`Page00007
`
`Page 00007
`
`

`

`M. O'NEIL
`
`Page 43
`
`| l
`
`Page 42
`
`M. O'N El 1.
`
`MR. MYERS: We disagree. Please
`proceed.
`Q.
`If one were to create or evaluate a
`potential actuarial class relating to number of
`accidents, expected claims loss data would be
`used in doing that. Am I right?
`MR. MYERS: Objection. 402, 403.
`I'm not sure what you mean by
`A.
`"evaluate."
`
`Q. To assess whether an actuarial class
`can be created.
`
`I believe
`I don‘t agree with that.
`A.
`you can create classes, zero, one, two, three,
`four accidents. And regardless, as a hypothesis,
`one would create those classes.
`
`Q. But in order to actually use them as
`classes in setting insurance premiums, am I
`correct that expected claims loss data would be
`used to determine whether they were actually
`useable in that fashion?
`
`MR. MYERS: Objection. 402, 403.
`It is not that simple ofan analysis
`A.
`to determine what the appropriate -- Well, let me
`go back.
`
`l l
`
`l l
`
`EL
`
`refresh your memory and look at that material.
`A.
`I've taken a quick look at that.
`0. Okay. Let me direct your attention to
`the first sentence of the paragraph, Ms. O‘Neil.
`Would you agree with me that automobile accident
`statistics such as the number of at-fault
`accidents can be considered as a risk
`
`characteristic for which an actuarial class might
`be created?
`
`MR. MYERS: Objection. 402, 403.
`I believe there are existing actuarial
`A.
`classes based on number of at-fault accidents,
`zero, one, two, three, four.
`MR. WAMSLEY: My issue is, Jim, that
`rather than answer yes or no and explain,
`Ms. O'Neil provides a narrative answer which
`does not frequently respond to my question.
`So I really would, in the interest of
`getting this done expeditiously, like to
`suggest that the witness, when I present a
`simple question that asks for a yes or no,
`give a yes or no. If she feels the need to
`explain or qualify, she's obviously free to
`do so.
`
`mumm-bUJNI—l
`
`01.1thl-JOKOCDHJO’XLfi-waI—‘Otpm‘dmmflb-UJNH
`
`aJmU‘LD-UJNH
`
`
`
`Page 44
`
`Page 45
`
`M. O'NEIL.
`
`M. O'NEIL
`
`
`
`You're setting up the class, is what I
`understand.
`Is that correct?
`
`Q. You're considering setting up that
`class, that's right. That's my hypothetical.
`A.
`So we don't already have it. Well, I
`think what would be done, if one did not want to
`collect any information ahead oftime, is we
`would hypothecate that drivers with more
`accidents might be drivers who have more claims.
`So we set up the classes and we collect data, and
`then we look at the data to see if indeed it is
`true that those drivers that have more accidents
`have more claims.
`
`So the data you collect is claims
`Q.
`data. Am I right?
`A. Well, you collect more than that. You
`would collect premium data, claims data,
`obviously the number ofexposures. You would
`collect a lot of data.
`
`So when you were doing your work in
`Q.
`New Jersey that you testified about previously
`this morning, what data did you refer to in your
`work there?
`
`A.
`
`For determinin what nortion of that
`
`mbmmHOmaummhwmHowm
`
`analysis are you refening to?
`Q.
`It's the activity that's mentioned in
`Paragraph 46 of your rebuttal declaration,
`Ms. O'Neil.
`A. Yes.
`
`Q. That's what I‘m referring to.
`A. Yes. But what portion of that are you
`referring to here now?
`Q.
`I'm referring to the part where ~—
`your activity in support of the development of an
`actuarial class system using driver record
`points.
`A. As I explained before, that system was
`deveIOped, as ljust mentioned, using the initial
`hypothesis that drivers with more driving record
`points would result in more claims. So given
`that hypothesis and the absence of any prior
`historical data of any kind, and the idea was to
`generate revenue, I believe we looked at what had
`been charged in -- by some companies that
`actually charged for violations and began with
`that, and then created scenarios that if these
`prices were utilized, what would be the revenue
`enerated, what would be a reasonable char e over
`
`TSG Reporting - Worldwide
`
`877-702—9580
`
`12
`
`{Pages 42 to 45)
`
`Page00008
`
`Page 00008
`
`

`

`
`mdeWLbUJNH
`
`M. O'NEIL
`
`what period of time, and so on.
`And so it was a revenue-based program
`based on the hypothesis of more points is more
`risky and relates to more losses.
`Q.
`So when you were doing that work, did
`you estimate expected claims losses?
`A. No.
`
`Q. What information did you use to set up
`those classes in the first instance?
`
`A. Well, first ofall, class definitions
`
`1
`
`2
`3
`4
`5
`6
`7'
`8
`
`9
`0
`
`1
`
`Page 47
`
`M. O’NEIL
`
`politically feasible to penalize someone with one
`ticket, so there was the forgiveness idea of,
`well, let‘s not charge people with two points.
`But by the same token there was a need to
`generate revenue, and so what source would you
`look at to start -— for starting price?
`As I said, we would look at premiums
`that might have been charged by insurers. We
`would have looked at fines that DMV was already
`imposing. You know, on the bottom line, in the
`end, the numbers related to total revenue
`
`
`
`were straightforward enough in that we had
`driving record points of zero to -- some people
`actually had more than 16 points and so on. So
`you had a range of actual points. You had the
`number of drivers at each point count. And based
`on that, there were clusters, of course, because
`
`2
`'I 3
`I 4
`I 5
`I 6
`I 7
`
`
`
`
`
`
`
`| 8
`points are based on different types of
`
`
`I 9
`violations. You might get two, four, you don't
`
`
`’ O
`get too many with three points, and so on. So
`
`
`' 1
`anyway, there were break points, obvious break
`
`
`' 2
`points within that particular frequency
`
`
`' 3
`distribution. Obviously, most of the people fell
`
`
`’ 4
`in the lower points areas, like two.
`
`
`But by the same token, it didn't sound
`-' 5
`
`
`Page 49
`
`
`Page 48_
`M. O'NEIL
`M. O'NEIL
`
`generated from the program.
`Q. Have you ever sought to set up an
`actuarial class in which you rely on accident
`statistics rather than expected claims losses?
`MR. MYERS: Objection. 402, 403.
`1 think that's what we were just
`A.
`talking about, is a frequency-based
`classification system.
`Is that correct?
`Q.
`I'm not necessarily confining my
`question to your experience in New Jersey now.
`My question is: In your career as an actuary,
`have you ever looked at and based an actuarial
`classification on accident statistics alone?
`
`But then again, that doesn't preclude the
`possibility that someone has done that.
`Q.
`It'sjust not something you've ever
`had any exposure to; is that right?
`A.
`l've not observed that, that's
`correct.
`
`Is this a good time for
`MR. WAMSLEY:
`a break, Jim? It's a good time for me to
`take a break.
`MR. MYERS: Then let's break.
`THE VIDEOGRAPHER: This is the end of
`DVD 1. Otfthe record at 10:45.
`
`(Recess taken.)
`THE VIDEOGRAPHER: This is the
`
`beginning of DVD 2 in the deposition of
`Ms. O'Neil. On the record at 10:58.
`BY MR. WAMSLEY:
`
`Q. Ms. O'Neil, let me hand you another
`exhibit previously marked by Liberty Mutual as
`Exhibit 1009 in CBM2012-2.
`
`(Liberty Mutual Exhibit 1009, having
`been marked for identification, is attached
`hereto.)
`Q. Can ou identif that as the ori,inal
`
`
`
`
`
`
`
`
`
`A. The basic classification plan utilized
`by my company and by other companies that I have
`worked with as a consultant was already in place.
`I did not need to set up classification plans for
`them.
`
`Q.
`
`A.
`
`So the answer is no?
`MR. MYERS: Objection. 402,403.
`1 think I said that I didn't need to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`set up any new classification plans for any
`companies.
`Q. Did you believe that the classes you
`set up in the New Jersey situation you‘ve been
`testifying about were f