The Underwriting Cycle: The Rule of Six

Alice RosenblattEVP, Integration, Information Management & Chief Actuary

* Views and opinions expressed by WellPoint, Inc. are not necessarily shared by The Ronald Reagan Presidential Library and Foundation.

2

• The statements made during this presentation that are not historical facts are forward-looking statements within the meaning for the Federal securities laws, and may involve a number of risks and uncertainties.

• Factors that could cause actual results to differ materially from expectations include, but are not limited to, the risks discussed in the Company’s most recent filings with the SEC, including its Annual Report on Form 10-K for the year ended December 31, 2005, and its Quarterly Reports on Form 10-Q for the reporting periods of 2006.

Safe Harbor Statement Under The Private Securities Litigation Reform Act of 1995

3

Agenda• Six-Year Underwriting Cycle Pattern

• Six Contributors to the Underwriting Cycle

• Why the Cycle is Six Years

• Action Steps

• Conclusion

4

Source: Blue Cross Blue Shield Association, 2005.

* Underwriting gain (loss) is calculated as revenue minus claims expense and administrative expense. Beginning in 2004, revenues and claims expense are reported gross of administrative services only amounts, however prior periods have not been reclassified to conform to this reporting change.

Six Year Underwriting Cycle PatternUnderwriting gain (loss) as a percent of revenue*

Data represents all Blues plans 1960 - 2005

-8%

-4%

0%

4%

8%

1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005*

5

Six Contributors to the Underwriting Cycle

• Claims payment cycle time

• Renewal dates and process

• Growth versus profit objectives

• Role of the actuary

• Rate regulations

• Reimbursement methods

6

Claims Payment Cycle Time

• Historical cycle times: 90-120 days

• Current cycle times: 40-60 days

7

Days in Claims Payable

Days In Claims Payable*

010203040506070

12/31/01 12/31/02 12/31/03 12/31/04 12/31/05 9/30/06

Day

s

* Includes the operations of the former WellChoice, Inc. and the former WellPoint Health Networks Inc. in all periods. Refer to “GAAP Reconciliation” section in back of book for further details regarding comparable basis information.

65.959.8 57.6

48.9 47.0 45.6

8

Renewal Dates and Process

Experience Period

16 Months of Trend

9/1/053/1/06 Midpoint

8/31/06Rating Period

1/1/077/1/07 Midpoint

12/31/07

9

Renewal Dates and Process

Trend Assumption

7%8%9%10%

Factor

1.0941.1081.1221.136

Difference

1.4%2.8%4.2%

10

Growth versus Profit Objectives

Sales

Growth Profitability

Underwritersand Actuaries

11

1970 1995 2006

WellPointBCBS Plans

Role of the Actuary

15

245

Total: 429134

295

Number of BCBS Plans: 140 65 37

Number of Credentialed Actuaries in BCBS Plans

Source: Blue Cross Blue Shield Association, R.J. Myers, “The Blues Sing Praises for Actuaries,” Actuary (May 1996): 8, and Company estimates.

12

Rate Regulation

Rate IncreaseApproved

SurplusPosition

13

Reimbursement Methods

DRGPer Diem

FFSDiscount from billed charges

LESS CONTROL

MORE CONTROL

14

Why the Cycle is Six Years

Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Prof

it (L

oss)

15

Action Steps –Claims Payment Cycle Time

• EDI

• Data warehouses

• Auto-adjudication

16

Action Steps –Renewal Dates and Process

• Small Group and Individual

• Government Programs

• Self-funded business

• Focal renewals

• Small Group and Individual

17

Action Steps –Growth Versus Profit Objectives

• And not or

• Profitable membership growth

• Pricing discipline

• Value not cost

GrowthProfit

18

Action Steps –Role of the Actuary

• Actuaries part of teams, including executive team

• Adequate actuarial staffing

19

Action Steps –Rate Regulation

• Lower the number of uninsured

• High risk rating pools

• Relationship building

20

Action Steps –Reimbursement Methods

• Actuarial models

• Use of data

• Pay for performance

21

-8%

-4%

0%

4%

8%

1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005*

Conclusion

From:

22

Conclusion

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

% o

f NSR

To:

Perspect ive

The Underwriting Cycle: The Rule Of SixSix factors have historically influenced a six-year health insuranceunderwriting cycle.

by Alice Rosenblatt

ABSTRACT: The underwriting cycle is a thing of the past for most health insurance compa-nies. There were six primary factors that caused the six-year pattern of the underwriting cy-cle for 1965–1991. These factors were claims payment cycle time, renewal dates and proc-ess, growth versus profit objectives, role of the actuary, rate regulation, and reimbursementmethods. Most companies have made major changes to influence these factors, which willprevent a recurrence of the underwriting cycles of the past.

The phenomenon of the underwrit-ing cycle has been well documented.The health insurance underwriting cy-

cle demonstrated amazing regularity during1965–1991—a six-year cycle of three years ofgains followed by three years of losses. Theyears since 1991 have shown a different pat-tern, however, which has led some observersto question whether the underwriting cyclewill continue into the future.

Based on my experience in the industry, theunderwriting cycle will continue into the fu-ture for some insurers but not for most. This isattributable to the actions taken by insurers,both profit and not-for-profit, that will helpthem avoid the swings of the underwriting cy-cle in the future. Even for those companies thatdo experience underwriting cycles, the swingswill not be as severe as they have been in thepast.

There are six causes of past underwritingcycles that explain the cycle itself as well asthe six-year pattern. These factors work inconcert with each other. In this brief commentI outline these factors and what lies ahead forthe health insurance industry.

Six Contributors To TheUnderwriting Cycle

� Claims payment cycle time. Before therecent improvements in electronic claims sub-mission, auto-adjudication, and other techno-logical advances, cycle time as measured fromservice date to payment date of a claim was of-ten 90–120 days. This implied that at any pointin time, the three or four most recent monthsof experience used in financial or pricing cal-culations were based primarily on projectionsof past claims data, generally per member permonth claims projected forward with a trendassumption. This straight-line type of projec-tion for the most recent months often led to adelay in recognizing that trend was turningupward or downward, causing pricing errorsand earnings swings.

� Renewal dates and process. The ma-jority of renewals for group health insuranceoccur on January 1; in the past, these rates wereguaranteed for twelve months. For the cus-tomer, this ties in nicely with fiscal periodsand with calendar-year deductibles. Renewalson any date are subject to a gap in time be-tween the prior experience and the future pro-

P e r s p e c t i v e : U n d e r w r i t i n g C y c l e

H E A LT H A F F A I R S ~ V o l u m e 2 3 , N u m b e r 6 1 0 3

DOI 10.1377/hlthaff.23.6.103 ©2004 Project HOPE–The People-to-People Health Foundation, Inc.

Alice Rosenblatt (Alice.Rosenblatt@wellpoint.com) is executive vice president, integration planning andimplementation, and chief actuary at WellPoint in Thousand Oaks, California.

jection period. For example, for a January 1 re-newal, the underwriter needs to release therenewal to the salesperson by October 1, andthe salesperson needs to release the renewal tothe account by November 1. For larger ac-counts, this might occur several months ear-lier. Therefore, the underwriter would be us-ing paid claim experience through the end ofAugust and projecting this experience for-ward. Health care actuaries often refer to the“midpoint to midpoint” rule for trend, whichmeans that a past twelve-month period (in this case,the twelve-month periodfrom September 1 in year xthrough August 31 in yearx+1) is being projected withsixteen months of trend tothe renewal twelve-monthperiod (in this case, January 1in year x+2 to December 31 inyear x+2). If one now combines this with theclaims projection process mentioned above,the projection period increases from sixteenmonths to nineteen months, since the lastthree months of incurred claims are themselvesbased on a projection. That is a long period ofprojection for health care costs. If a 12 percenttrend rate was used, and the trend rate shouldhave been 15 percent, the difference on a com-pounded basis for nineteen months is worth5.1 percent. That is a huge difference for an in-dustry that usually runs pretax net gains of3–7 percent. If most of the business renewedon January 1, then that 5 percent miss wouldapply to the majority of the business for a fulltwelve months.

� Growth versus profit objectives. In theearly years of for-profit managed care, inves-tors rewarded growth over profitability. At thepoint of greatest gain in the cycle, managementwas faced with the daunting question ofgrowth versus profitability. If enrollment hadslowed down and if management was com-fortable with surplus levels, there was an at-tempt to lower trend rates or to take otherpricing actions to improve the competitive po-sition. This started the cycle in the oppositedirection.

� Role of the actuary. The underwritingcycle has been studied primarily for Blue Crossand Blue Shield (BCBS) plans, since the datahave been available through the Blue Cross andBlue Shield Association. Many Blues plans didnot have actuaries on staff until fairly recently.An article published in the May 1996 issue ofthe Actuary reported that the number of actu-aries at BCBS plans increased from 15 in 1970to 245 in 1995.1 When health maintenance or-ganizations (HMOs) relied on capitation,

many companies did notthink that actuaries wereneeded. As a member of theactuarial profession, I caneasily point to the lack of ac-tuaries as one of the causalagents of swings in the un-derwriting cycle. Actuariesare trained in using data toanalyze trends and in build-

ing financial models. However, the most diffi-cult part of a health care actuary’s job is pre-dicting when trend rates are changingdirection—either up or down; a delay in rec-ognizing that inflection point is often the keydriver of the underwriting cycle.

� Rate regulation. In states where ratesfor individuals, small groups, or seniors re-quired approval from regulators, it was easierto get approval for higher rates when the plan’ssurplus position was low than when it washigh. It was also difficult to convince regula-tors that trend was accelerating when thiscould not be statistically demonstrated fromprior experience. Depending on the size ofthese blocs of business, this might compoundthe effect, so that when the surplus position isat its highest and the plan starts trying to raiseenrollment in response to competitive pres-sures, it is also not getting sufficient rate in-crease approvals from regulators; this drivesearnings toward the downward part of the cy-cle. The rate approval process also introducesan even larger time gap between the experi-ence period and the rate projection period, be-cause of the time needed for regulatory review.

� Reimbursement methods. BCBS planshave historically used network models, even

1 0 4 N o v e m b e r / D e c e m b e r 2 0 0 4

P e r s p e c t i v e

“Since 1991 thecycle has evolvedfrom a pattern of

gains and losses to apattern of greaterand lesser gains.”

before the rise of managed care. However, if thereimbursement mechanism was a discountfrom billed charges, the projection of futuretrend rates was just as uncertain as a projec-tion of fee-for-service trend rates. For com-mercial carriers, payments were generallymade on a “usual and customary” fee basis, soonce again they were inherently uncertain.This resulted in misses on setting trend ratesin pricing, which caused the earnings swings.

How The Six Factors WorkTogether

If one has a majority of renewals on January1, and the trend factor used for pricing is toohigh or too low, pricing itself will be too highor too low, causing the underwriting cycle.The six years can be characterized as follows:(1) Year 1: Trend starts to increase above theprojection. Because of the nineteen-month lagfor renewal rating, this inflection point ismissed, and actuaries hold the trend constant.There is a minor loss, or earnings are close tobreak-even for the year.

(2) Year 2: The company has missed thetrend for January renewals on the low side.The majority of cases have been underpricedfor a full twelve months, leading to heavylosses. The error would normally be correctedby the middle of the year.

(3) Year 3: Trend is correct for the Januaryrenewals, but there are still some inadequaciesin the early months, because of underpriced re-newals from the months of year 2 prior to thecorrection (such as underpriced results forJanuary, since the correction cannot be madeuntil the February renewal). Financial resultsare closer to break-even.

(4) Year 4: Trend and pricing are correct forall renewal months, and the profitability pic-ture improves, and is better than break-even.

(5) Year 5: Rates are correct for all renewalmonths, and trend may have started to de-crease. The gains are highest. Market pressurestarts to build, and discounts are built intorates.

(6) Year 6: Rates have been discounted toomuch, so the gains are lower than they were inyear 5, but there are still gains.

In Exhibit 1 in the paper by Joy Grossmanand Paul Ginsburg, the pattern describedabove is not entirely consistent.2 This is be-cause the six-year timing may affect companiesat different times, so that when one looks atthe entire system of Blues plans, the pattern isnot exactly as described above. However, itdoes follow three years of underwriting mar-gin losses followed by three years of gains.

Companies within the BCBS system as wellas HMOs and commercial carriers will con-tinue to experience the underwriting cycle.However, since many companies have takensteps to avoid the downward part of the cycle,since 1991 the pattern has been different thanthe earlier pattern; the cycle has evolved from apattern of gains and losses to a pattern ofgreater and lesser gains.

Steps Insurers Have Taken ToAvoid The Cycle

The steps that most health insurers havetaken to avoid the down side of the underwrit-ing cycle fall into the six categories describedabove. (1) Claims payment cycle time: Throughtechnology enhancements, more claims arenow received electronically (some carriershave even given computers to providers to as-sist in this goal). Many companies have de-creased cycle times to fifty to sixty days on adollar-weighted basis. They also have investedin data warehouses that allow actuaries to re-trieve claims data shortly after month’s end, sothat actuaries are always working with cur-rent data.

(2) Renewal dates and process: Companieshave diversified the customer segments theyserve. Large-group business is heavily January-dominated; small-group business is less so; in-dividual business is not at all. Governmentprograms generally have July 1 or October 1 re-newal dates. Large-group business can bewritten on a self-funded basis. Ancillary linesof business not subject to medical price trendscan help moderate the financial impact of therenewal distribution. Some companies havechanged their contracts to al low off-anniversary rate actions and will take this stepif there has been a big miss in pricing. Also,

P e r s p e c t i v e : U n d e r w r i t i n g C y c l e

H E A LT H A F F A I R S ~ V o l u m e 2 3 , N u m b e r 6 1 0 5

some companies may have blocs of business,such as individual or small-group, for whichfocal renewals are done—meaning that all ofthe business in a bloc is given a rate increase onthe same date, instead of having each piece ofbusiness renewed on an anniversary date.

(3) Growth versus profit objectives: Somecompanies view this as an “and,” not an “or,”and have set goals that use words such as“profitable membership growth.” These com-panies have also developed pricing disciplineand will not discount adequate rates to thepoint of inadequacy, in spite of market pres-sures. Companies have also learned the impor-tance of focusing on value instead of cost. Thevalue comes from network quality, informa-tion, pricing stability, financial stability, prod-uct innovation, disease management and othermedical management programs, and customerservice. Many companies realize that this isnot an industry in which discounting works,with pretax margins of 3–7 percent, not a 30percent margin business in which a companywould be satisfied with a 20 percent margin.

(4) Role of the actuary: In companies thathave taken steps to avoid the cycle, not onlyare there an adequate number of professionalactuaries, but the actuaries hold positions atthe executive team level. Actuaries are inte-grated into the entire operation of the com-pany—not just valuation and pricing, but un-derwriting, network contracting, medicalmanagement, and product design.

(5) Rate regulation: Some companies canobtain adequate rates in spite of rate regula-tion. They have developed good working rela-tionships with their regulators and have as-sisted in efforts to lower the number ofuninsured people in key markets, participatedin high-risk rating pools, and worked to keepclaims costs and administrative expenses aslow as possible.

(6) Reimbursement methods: Many com-panies have developed provider reimburse-ment mechanisms that do not make them de-pendent on changes to bil led charges.Actuaries at these companies have developedmodels that allow them to predict contractualrates for future periods fairly accurately. These

companies also review utilization patternsand look for ways to improve quality by elimi-nating inappropriate use of health care ser-vices.

In my opin ion, the underwriting cyclepattern seen from 1965 to 1991 is a thing ofthe past. However, it will probably con-

tinue for some companies and still swingearnings up and down slightly for others. Ascompanies continue to make the changes de-scribed previously, the swings will continueto moderate. It is in consumers’ best interestfor the underwriting cycle pattern to miti-gate, since the result will be more stable pre-miums.

The author thanks Dave Colby, WellPoint chieffinancial officer, and the senior actuaries at WellPointfor their review of this paper and helpful comments.

NOTES1. R.J. Myers, “The Blues Sing Praises for Actu-

aries,” Actuary (May 1996): 8. The article also re-ported that the number of Society of Actuariesmembers increased from four in 1970 to 206 in1995.

2. J.M. Grossman and P.B. Ginsburg, “As the HealthInsurance Underwriting Cycle Turns: WhatNext?” Health Affairs 23, no. 6 (2004): 91–102.

1 0 6 N o v e m b e r / D e c e m b e r 2 0 0 4

P e r s p e c t i v e

1

WellPoint, Inc. GAAP RECONCILIATIONS Investor Conference Dinner

December 11, 2006 WellPoint, Inc. (the “Company”) has referenced certain non-GAAP financial measures throughout this presentation book. Included below and on the following page are descriptions of those measures, including reconciliations to the most directly comparable measures calculated in accordance with GAAP. Comparable Basis Days in Claims Payable Statistics On November 30, 2004, Anthem, Inc. acquired WellPoint Health Networks Inc. and Anthem, Inc. changed its name to WellPoint, Inc. On December 28, 2005, WellPoint, Inc. acquired WellChoice, Inc. For accounting purposes, the WellChoice transaction was assumed to have closed on December 31, 2005. Accordingly, operations of the former WellPoint Health Networks Inc. and the former WellChoice, Inc. are only included from the respective transaction closing dates forward. In order to provide more meaningful comparisons between and among periods, the Company has included “Comparable Basis” calculations of the Days in Claims Payable statistic for periods prior to 2006. These Comparable Basis statistics have not been calculated in accordance with GAAP. The methodologies for calculating the Comparable Basis statistics are described in the footnotes to the tables where such information is reconciled to WellPoint, Inc.’s historical GAAP financial results.

2

WLP WLP($ In Millions) As Reported (1) WHN (2) WC (3) Comparable Basis (4)

December 31, 2005Medical claims payable $4,853.4 – – $4,853.4Benefit expense $8,200.3 – $1,305.9 $9,506.2Days in period 92 – – 92Days in claims payable 54.5 – – 47.0

December 31, 2004Medical claims payable $4,202.0 – $678.8 $4,880.8Benefit expense $5,051.7 $2,961.1 $1,164.8 $9,177.6Days in period 92 – – 92Days in claims payable 76.5 – – 48.9

December 31, 2003Medical claims payable $1,841.7 $2,747.1 $609.5 $5,198.3Benefit expense $3,083.0 $4,149.5 $1,072.9 $8,305.4Days in period 92 – – 92Days in claims payable 55.0 – – 57.6

December 31, 2002Medical claims payable $1,826.0 $2,422.3 $559.9 $4,808.2Benefit expense $2,901.6 $3,526.2 $974.4 $7,402.2Days in period 92 – – 92Days in claims payable 57.9 – – 59.8

December 31, 2001Medical claims payable $1,360.3 $1,934.6 $634.1 $3,929.0Benefit expense $1,967.1 $2,611.4 $903.6 $5,482.1Days in period 92 – – 92Days in claims payable 63.6 – – 65.9

(1)

(2)

(3)

(4)

Represents the historical, GAAP information of WellPoint, Inc. as of the date indicated and for the respective three month periods then ended.

The "Comparable Basis" information was calculated by adding the historical, GAAP information of the former WellPoint Health Networks Inc. and the former WellChoice, Inc. to WellPoint, Inc.'s historical, GAAP information for the respective periods.

Represents the historical, GAAP information of the former WellPoint Health Networks Inc. as of the date indicated and for the respective three month periods then ended.

Represents the historical, GAAP information of the former WellChoice, Inc. as of the date indicated and for the respective three month periods then ended.

WellPoint, Inc.Days in Claims Payable

(Unaudited)