SCHIFFS - Schiff's Insurance Observer€ states that “situations which could result in conflicts of interest or the appearance of a conflict of interest should be avoided whenever

In February, we republished three ar-ticles from 1996 about AIG’s rela-tionship with Coral Re, a nebulousBarbados reinsurer to which AIG

ceded about $1.6 billion in reserves.Although Coral Re looked and smelledlike an AIG affiliate, AIG denied that itwas. On the following pages we are re-publishing nine articles about AIG thatwere written between 1998 and 2004. (Forall of our articles about AIG, please referto the index in the March 15 issue.)

Although AIG has long been extremelyprofitable, it has also been a “black box”that appeared to have a predilection for in-novative bookkeeping. If you mentionedthat AIG may have engaged in some sortof legerdemain, it tended to elicit a strongreaction from 70 Pine Street, the com-pany’s headquarters. “AIG has always pro-vided complete and accurate financial in-formation,” was a standard response.(Yelling, threatening, and bullying werealso standard responses.)

On March 30, AIG issued a press re-lease in which it admitted the ugly truth:it did not provide complete and accurate fi-nancial information. The press release

stated that AIG entered into transactionsthat “appear to have to have been struc-tured for the sole or primary purpose ofaccomplishing a desired accounting re-sult.” Translated into English, that meansthat AIG screwed around with its num-bers, thereby misleading everyone whorelied on them.

What follows are some examples ofAIG’s mischief.

AIG entered into $500 million of“reinsurance” transactions with GeneralRe. Because no risk was transferred, thetransactions weren’t really reinsurance.This phony “reinsurance” made AIG’sloss reserves appear greater than theywould have been otherwise, giving themisleading impression that the company’sreserving practices were more conserva-tive than they really were.

Between 1991 and 2004, AIG ceded alot of reinsurance business to a littleBarbados company called Union ExcessReinsurance. These “reinsurance” trans-actions accounted for $1.1 billion of AIG’snet income. Upon closer inspection, AIGdiscovered that these transactions weresomething of a sham. AIG doesn’t knowwhat the hell happened, but it says that it“now believes” that Union Excess’sshareholders have a financial arrangementwith Starr International Company(SICO)—a private holding company thatowns twelve percent of AIG and is con-trolled by Hank Greenberg and other AIGhonchos. The bottom line: it appears thatthere was no economic substance to the$1.1 billion of net income that AIG re-ported from these transactions.

The Union Excess transactions also

SCHIFF’S

SCHIFF’S INSURANCE OBSERVER • 300 CENTRAL PARK WEST, NEW YORK, NY 10024 • (212) 724-2000 • DAV I D@IN S U R A N C EOB S E RV E R.C O M

April 4, 2005Volume 17 • Number 9 I N S U R A N C E O B S E R V E R

The world’s most dangerous insurance publicationSM

Say it Ain’t So, Hank

TA B L E O F C O N T E N T SOctober 1998: Darkness on the Edge of Town—AIGand SunAmerica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

September 2000: An Extreme Price for Growth . . . .7

April 5, 2001: Hostile Takeover Attempt for AmericanGeneral—Is AIG’s Stock Too High? . . . . . . . . . . . . .9

May 2, 2002: The Greatest Risk is Taking Too MuchRisk—AIG’s Audit-Committee Report . . . . . . . . . .12

July 25, 2002: Audit Committees, Legends, and P/ERatios, Part 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

August 16, 2002: AIG to Change Audit CommitteeReport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

April 7, 2003: Inside AIG’s Proxy Statement . . . . . .23

June 23, 2003: AIG’s Secret Connection with Director . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

February 26, 2004: The Art of Manipulation? . . . . .30

AIG and the Art of Financial Prestidigitation

SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 APRIL 4, 2005 2

raise questions that AIG has not ad-dressed. How could AIG have failed todisclose such large “related-party” trans-actions to its shareholders? Did AIG’sboard of directors know about these trans-actions? If not, why didn’t AIG’s senior ex-ecutives—who are officers, directors,shareholders, or beneficiaries of SICO—tell the board about the transactions?AIG’s “Code of Business Conduct andEthics” states that “situations whichcould result in conflicts of interest or theappearance of a conflict of interest shouldbe avoided whenever possible.” If an AIGofficer or director is aware of anything thatcould reasonably be expected to create aconflict of interest, he’s supposed to dis-cuss it with the company’s general coun-sel. Ernest Patrikis has been AIG’s gen-eral counsel since 1998. His predecessor,Florence Davis, now runs the StarrFoundation, which owns 2.05% of AIG.What do they know? Or, what don’t theyknow?

AIG ceded a significant amount of“reinsurance” to Richmond InsuranceCompany in Bermuda. Or did it just shiftassets from one of its pockets to anotherand call it reinsurance? AIG’s recent “re-view of operations” turned up “previouslyundisclosed evidence” that AIG controlsRichmond. As a result, there was no transfer of risk, which means that AIG’sfinancials didn’t accurately portray thereal income statement or balance sheet—or maybe both.

Between 2000 and 2003, AIG engagedin some nifty transactions with CapcoReinsurance Company, located in lovelyBarbados. Somewhere between NewYork and the Caribbean these transac-tions magically turned $200 million ofAIG’s underwriting losses into $200 mil-lion of capital losses (losses from invest-ments). That means that AIG’s operatingincome appeared $200 million greaterthan it really was. (Operating income isof great importance, because, as HankGreenberg pointed out in AIG’s 2000 let-ter to shareholders, it’s “the way we andthe investment community look at our re-sults.” Considering that AIG’s stock hasoften traded at thirty times earnings, itseems reasonable to say that the Capcodeals inflated AIG’s market cap by about$1.5 billion.

AIG had other ways to make its oper-ating income appear larger than it reallywas. Between 2001 and 2003, it sold call

REGISTER NOW

David Schiff, editor of Schiff’s Insurance Observer, will tell you whathe’s riled up about these days. Throughout the conference he will, as always,interrogate the speakers and force them to answer brazen questions.

In June 1994, Schiff ’s wrote an admiring profile of Christopher Davis, portfoliomanager of the Davis Funds, which had $300 million under management. (Chrisis the only money manager we’ve ever profiled.) We picked a winner. The DavisFunds now manage $40 billion, and the firm’s primary fund has outperformed theS&P 500 during every meaningful period since its inception in 1969. Chris will tellus about the Davis’s sixty-year history of investing in the insurance business, andshare his thoughts on the mutual-fund industry, shareholder activism, and more.

Two years after receiving his Ph.D. in economics from Harvard, 27-year-oldJames Stone became the youngest insurance commissioner ( Massachusetts)in history. Four years later, in 1979, Jimmy Carter appointed him as chairmanof the Commodity Futures Trading Commission. When his term ended in 1983,he moved back to Boston and founded The Plymouth Rock Company, aprivately-held insurance holding company that now writes well over $1 billionin premiums—quite profitably. Jim will share his perspective on auto insurance,regulation, public policy, and being an entrepreneur in the insurance business.

William Koenig is Senior Vice President and Chief Actuary of “the quietcompany,” Northwestern Mutual. Bill will give us his perspective aboutreserving—especially when it involves universal-life products with secondaryguarantees. His comments, which will not be quiet, should leave some membersof the insurance industry feeling worried.

Lunch: Decent food, fine conversation.

Andrew Kaufman, a founding partner of Kaufman Borgeest & RyanLLP, is one of the leading attorneys specializing in the defense of health-careproviders and hospitals. He’s tried more than sixty cases to verdict, and is thepast president of the New York State Medical Malpractice Defense Bar and pastvice chairman of the American Bar Association Section on Law and Medicine.Andy will give you a view from the battlefield, tell you his thoughts on tortreform, and explain why he’s not for caps on pain and suffering.

continued on next page

9:00 a.m.

9:30 a.m.

10:40 a.m.

11:20 a.m.

Noon

1:00 p.m.

‘A Vast Wasteland’PRESENTING THE ANNUAL

SCHIFF’SINSURANCE CONFERENCE

Tuesday, April 12, 20058:30 am - 5:30 pm

New York City

Registration fee: $695 for subscribers

$745 for non-subscribersCall (434) 977-5877 for more

information, or reserve a place [email protected]

http://www.snlcenter.com/schiff/spring2005/intro.asp

mailto:[email protected]

http://www.snlcenter.com/schiff/spring2005/intro.asp

options on some of its bonds that had un-realized gains. It then entered into a seriesof forward transactions and swaps that,somehow, transformed $300 million ofcapital gains into $300 million of “invest-ment income.” (Since investment incomeis a component of “operating income” andcapitals gains aren’t, this had the effect ofoverstating AIG’s earnings power.)

Finally, it turns out that AIG misclassi-fied “certain items,” and, as a result, its re-ported net investment income was over-stated by four percent between 2000 to2004. That doesn’t sound so bad, does it?After all, what’s four percent in the grandscheme of things? Well, it turns out to bea lot—$3 billion. In 2003, for example,this “misclassification” caused AIG’s op-

erating income (the key figure everyonelooks at), to be overstated by about fourpercent—$660 million.

It’s likely that there will be more rev-elations about the unsavory inner work-ings of AIG. Perhaps that’s why AIG putout another press release last night. It wasa letter from CEO Martin Sullivan that,we suppose, was meant to reassure share-holders that AIG wasn’t a house of cardsrun by a gang of con men overseen by di-rectors who are deaf, dumb, and blind.“We are committed to improving trans-parency and corporate governance,” wroteSullivan.

We’re certain that AIG’s governanceand transparency will improve. The ques-tion, however, is this: “Exactly how badare they right now?”

Sullivan also said that it was “unfortu-nate that current circumstances have ob-

scured the reality that AIG’s unique globalfranchise is sound.” Alas, he got it back-wards. It’s unfortunate that AIG’s uniqueglobal franchise obscured the reality of thecompany’s financial condition.

Please continue to the following pages toread articles about AIG from 1998 to 2004.


Property insurers’ combined ratios are five to eight points higher than theyshould be, says Robert Dowdell, CEO of Marshall & Swift/Boeckh(M&S/B), which is doing something to remedy that. M&S/B, long known as abuilding-cost provider in claims and underwriting, has become a corporateSherlock Holmes that uses logic and statistical analysis on the massive amountsof data it processes to improve carriers’ underwriting results. “The data has animportant story to tell,” says Bob, who will tell us an important story about riskdifferentiation, pricing, database analytics, and much more.

Warren Buffett talked to just one securities analyst: Alice Schroeder ofMorgan Stanley. In 2003, Alice, then Institutional Investor’s top-ranked P/Canalyst, made an unusual career move—she left the day-to-day world of WallStreet to write a book about Buffett’s life and philosophies. Alice, who is toBuffett what Boswell was to Johnson, won’t be finished with her tome (which wepredict will be a best seller) for a couple of years. In the meantime, she’ll tell youwhat’s on her mind.

David Schiff will have his say on the great insurance issues of the day, and discusswhere he sees value and solvency (or the lack thereof).

Attendees will socialize with their fellow insurance mavens and observers, dis-cussing the day’s events and making deals over cocktails while taking in the viewfrom the top of the New York Athletic Club.

There will be an additional reception and dinner for those who want more of agood thing. The venue is the Coffee House, a convivial, somewhat worn-at-the-edges private club devoted to “agreeable, civilized conversation.” Attendance islimited to 36 people.

1:45 p.m.

2:45 p.m.

3:45 p.m.

4:45 p.m.

6:00 p.m.

‘A Vast Wasteland’PRESENTING THE ANNUAL


Tuesday, April 12, 20058:30 am - 5:30 pm

New York City

Editor and Writer . . . . . . . . David SchiffProduction Editor . . . . . . . . . Bill LauckForeign Correspondent . . Isaac SchwartzEditorial Associate. . . Yonathan DessalegnCopy Editor . . . . . . . . . . . . John CaumanPublisher . . . . . . . . . . . Alan ZimmermanSubscription Manager . . . . . . Pat LaBua

Editorial OfficeSchiff’s Insurance Observer300 Central Park West, Suite 4HNew York, NY 10024Phone: (212) 724-2000Fax: (434) 244-4615E-mail: [email protected] Website: InsuranceObserver.com

Publishing HeadquartersSchiff’s Insurance ObserverSNL c/o Insurance Communications Co.One SNL Plaza, P.O. Box 2056Charlottesville, VA 22902Phone: (434) 977-5877Fax: (434) 984-8020E-mail: [email protected]

Annual subscriptions are $189. For questions regarding subscriptions pleasecall (434) 977-5877.

© 2005, Insurance Communications Co., LLC.All rights reserved.

Reprints and additional issues are avail-able from our publishing headquarters.

Copyright Notice and WarningIt is a violation of federal copyright law toreproduce all or part of this publication. You arenot allowed to e-mail, photocopy, fax, scan, dis-tribute, or duplicate by any other means thecontents of this publication. Violations of copy-right law can lead to damages of up to $150,000per infringement.

Insurance Communications Co. (ICC) is controlled bySchiff Publishing. SNL Financial LC is a research and pub-lishing company that focuses on banks, thrifts, real estateinvestment companies, insurance companies, energy andspecialized financial-services companies. SNL is a nonvot-ing stockholder in ICC and provides publishing services to it.

SCHIFF’SThe world’s most dangerous insurance publicationSM

I N S U R A N C E O B S E R V E R

It is said that markets are efficient.We won’t bother to debate that. Buteven if they’re efficient, that doesn’tmean they’re always rational.

Markets are made by people, who aregiven to feelings such as optimism, fear, exuberance, and depression. Theirbehavior will now and then drive pricesto extreme highs or lows. (Even Schiff’sInsurance Observer’s subscribers aren’talways rational; several hundred havefailed to sign up for our Evening TelegraphEdition, which is delivered by e-mail orfax and included with subscriptions at noadditional charge.)

When markets become too irrational—when pricing, supply, or demand gets wayout of whack—something usually hap-pens. If, for example, gold were selling for$275 in London and $273 in New York,arbitrageurs would short London gold andbuy New York gold. These actions wouldeventually result in a convergence of theLondon and New York prices.

Insurance can work in a similar fashion.If writing non-standard auto insurance inNorth Carolina is unusually profitable, thesmell of money will cause numerous insur-ers to flock to that market. The increased

competition will then drive profit marginsdown, or eliminate them entirely. Theabsence of profits will cause some insurersto exit the market, which will, in turn,reduce competition and, eventually, createan environment in which profits can bemade—at least for a while.

In a brief examination, we shall turnour attention to AIG, an example of agreat company whose stock trades at anextreme, optimistic, exuberant valuationthat leaves little margin for safety.

There is, of course, a certain logicbehind AIG’s rich valuation. It has a$200-billion market cap and its stock isextremely liquid (which means that insti-tutions can easily buy and sell in size).More importantly, AIG has a long historyof steady growth. (Because AIG hasnever disappointed in the past, many

take it on faith that it will never disap-point in the future.) AIG is a core holdingof institutions and mutual funds, and,according to Zacks, is rated a “buy” by 21of the 24 securities analysts that follow it.

Because of its virtues, AIG’s shareschange hands at 37.4 times earnings and5.8 times book value—levels that arestratospheric, at least as measured byboth basic math and financial history. (Ata 37.4 p/e multiple, investors are earninga 2.7% yield on their investment in AIG.)To justify its current valuation, AIG mustcompound its earnings at a breathtakingrate for a long period—a feat thatbecomes increasingly difficult with size.

The definitive study of AIG, AmericanInternational Group: Cultivating GlobalGrowth, was published in May, when AIG’sstock was at 74. (The report’s authors,

SCHIFF’S


September 2000Volume 12 • Number 1 I N S U R A N C E O B S E R V E R


“You suffer from the delusion that your casualty reserves are adequate.”

Walk Softly and Carry a Big Multiple

TA B L E O F C O N T E N T S

An Extreme Price for Growth: There’s no roomfor error in AIG’s stock price . . . . . . . . . . . . . . . . .1

The Catcher in the Rye Reliance: Saul Steinberg’sstory, as you’ve never read it before . . . . . . . . . . . .3

Comparing the Rating Agencies: Best, Moody’s,and S&P on Reliance . . . . . . . . . . . . . . . . . . . . . .4

Steal This Insurance Company: MONY,MetLife, and John Hancock • Demutualization andIts Discontents . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

Insurance Companies’ Secret ‘Public’ Data:What’s the Insurance Industry Afraid Of? • TraipsingThrough the Regulatory Morass . . . . . . . . . . . . . .9

An Extreme Price for Growth

Alice Schroeder, Gregory Lapin, and ChrisWinans, were then at PaineWebber; theyare now at Morgan Stanley Dean Witter.)Their 286-page tome projects that AIG’searnings will grow at a 16% rate through2005. If AIG does indeed do that, its cur-rent stock price, 86, will be 16 times thatyear’s earnings. Viewed another way, if allgoes as planned, in 2005 an investor will“earn” a 6.25% return, based upon AIG’scurrent stock price.

Rather than dispute the detailedanalysis of Schroeder, et al. (Schroeder,after all, is a friend, subscriber, and fea-tured speaker at our spring conference),we’ll stick to the risks of investing inAIG at its current, extreme valuation.

For starters, AIG will not be a big ben-eficiary of any upturn in the domesticproperty-casualty market, since its domes-tic property-casualty income is only 25% ofits operating earnings. (Domestic property-casualty is projected to grow at an 8% clip.)

Although long viewed as a property-

casualty company, AIG has changed itsstripes, and a disproportionate amount ofits future growth is expected to comefrom life insurance, financial services, andasset management. Domestic and foreignlife insurance are projected to grow atmore than a 17% pace, and earnings fromasset management are projected toquadruple by 2005, and comprise 10% ofAIG’s earnings then (up from 5% now).

Way back in our October 1998 issue,when AIG’s stock was 49, we observedthat it was selling at a lofty 24 times earn-ings. The stock is now 86—an even lofti-er 37.4 times earnings.

Although AIG’s earnings haveexpanded at a 17% annual pace over thelast two years, its price-earnings ratio hasexpanded 60%. If AIG’s p/e multiple hadremained constant, its stock price wouldbe 66 rather than 86. (If its multiple hadshrunk to 20 times earnings, its stockwould still be at 49.)

The bottom line: 54% of the gain inAIG’s share price over the last two yearshas been due to the expansion of its p/emultiple, rather than to earnings growth.

Price-earnings multiples cannotexpand indefinitely. Indeed, they havebeen known to contract. This happenedto AIG (and many others) during the1970s (see chart). During the 1990s,however, AIG’s p/e multiple regained itslost ground, and then some, as AIGbecame the insurance stock.

According to Value Line, AIG’s earn-ings have grown at a 13.5% rate over thepast ten years. During that same periodits p/e multiple has expanded from 10.9times earnings to 37.4 times earnings.

An advantage of having a high p/e mul-tiple is that a company’s stock becomes afine acquisition currency. (BerkshireHathaway’s acquisition of General Re forstock is a case in point.) Interestingly, AIGhas not benefited much from its high mul-tiple. Although it acquired SunAmerica forstock, SunAmerica had an even higher p/ethan AIG; thus the acquisition was notimmediately accretive to earnings.

AIG should earn about $5.8 billion in2000. If it is to grow at the projected16% next year, it must come up with anadditional $900 million in earnings. (Byway of comparison, Chubb’s total earn-ings for next year are projected to beabout $825 million.) One way AIG cangrow is by using its high-p/e currency tobuy earnings. AIG is acquiring HSB (for-

merly Hartford Steam Boiler) for $1.2billion in stock—a price equal to 20times earnings. Because AIG’s p/e ratiois almost double that of HSB, the acqui-sition will be accretive to AIG’s earningper share, and, in fact, should representabout 3% of AIG’s earnings-per-sharegrowth next year.

But AIG is so large that it’s difficultfor it to make acquisitions that, by them-selves, materially alter its growth rate. Atthe margin, however, if it can use itsstock to buy lower-multiple companies,then it can eke out incremental growthvia an arbitrage of earnings’ multiples.

Absent an expansion in its multiple, ifAIG grows at a 14% rate forever, aninvestor could expect no more than a14% annual return. If AIG’s growth ratefails to achieve this difficult hurdle, it’sslower growth would likely lead to amuch lower multiple. (A much lowermultiple could also occur if investors’exuberance subsides.)

From our vantage point, the risk ofbuying AIG outweighs the reward. �

2 SEPTEMBER 2000 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

Editor and Writer . . . . . . . David SchiffCopy Editor . . . . . . . . . . . . . John CaumanEditorial Associate . . . . . . Isaac SchwartzProduction Editor . . . . . . . . . . Bill Lauck

Publisher. . . . . . . . . . . . . . . . . . Reid NagleSubscription Manager . . . . . . . Pat LaBuaAdvertising Manager . . . . . Mark Outlaw

Editorial OfficeSchiff’s Insurance Observer300 Central Park West, Suite 4HNew York, NY 10024Phone: (212) 724-2000Fax: (212) 712-1999E-mail: [email protected]

Publishing HeadquartersSchiff’s Insurance ObserverSNL c/o Insurance Communications Co.321 East Main Street, P.O. Box 2056Charlottesville, VA 22902Phone: (804) 977-5877Fax: (804) 984-8020E-mail: [email protected]

Subscriptions are $149 for one year, whichincludes a subscription to our EveningTelegraph Edition. For questions regardingsubscriptions please call (804) 977-5877.

© 2000, Insurance Communications Co., LLCAll rights reserved.

Copyright Notice and WarningIt is a violation of federal copyright law toreproduce all or part of this publication. Youare not allowed to photocopy, fax, scan, orduplicate by any other means the contentsof this publication. Violations of copyrightlaw can lead to damages of up to $100,000per infringement.



0

100

200

300

400

500

600

700

-10

-5

0

5

10

15

20

25

30

35

40

AIG: Does Anyone Remember 1972?

P/E ratio

American International Group

Price to Book Value

In 1972, AIG’s shares traded at 518% of bookvalue and 32.6 times earnings. Between 1972and 1974, AIG’s stock fell 66%, as these inflat-ed multiples shrank, even though AIG’s earn-ings grew. During the last two years, AIG’sprice-to-book-value ratio and p/e ratio haveentered uncharted territory.

Source: Value Line

Price to Book Value P/E ratio

0

100%

200%

300%

400%

500%

600%

700%

’72 ’76 ’80 ’84 ’88 ’92 ’96 ’00

40

35

30

25

20

15

10

5

On Tuesday, after the markethad closed, AIG announced anunsolicited $46-per share offerto acquire American General

in a $23-billion stock transaction—a 25%premium over the closing price.

One wouldn’t go so far as to call HankGreenberg a corporate raider, but the factremains: American General had previ-ously agreed to be acquired by Britain’sPrudential Plc. The value of that transac-tion—$26 billion when announced—hadfallen due to the decline in Prudential’sshares. Hank Greenberg, seizing themoment, made a big move at a timewhen his actions were likely to be metwith acceptance from Wall Street, andscant resistance from American General.(Since the company was already in play,management could not easily rebuff asignificantly higher offer.)

That AIG would attempt to use itsrichly valued stock to make acquisitionsisn’t surprising. On February 16 wewrote that we expected insurance com-panies to issue equity to take advantageof the favorable market for insurancestocks. We didn’t expect AIG, “whichsells for 548% of book value, to issuestock in a secondary offering. AIG is solarge ($200-billion market cap) that itcouldn’t do an offering large enough tobe meaningful. Hank Greenberg hassaid, however, that AIG is looking atacquisitions, and given his company’sstupendous price-earnings (p/e) multi-ple, we’d be surprised if his currency ofchoice was not AIG stock.”

Last September, in an article dis-cussing the optimistic valuation accorded

AIG’s shares, we noted that becauseAIG’s stock had an extremely high p/eratio (37.4), it made a fine acquisitioncurrency. We also noted that AIG hadn’tbeen able to put that currency to gooduse. (Given AIG’s multiple, almost anyacquisition would be accretive to earn-ings the first year—although not neces-sarily in later years.)

In order for AIG to maintain its sky-high p/e ratio, at the very least it mustcontinue to achieve the rapid and steadygrowth in earnings per share for which itis known and loved. Given AIG’s sizeand its cyclical businesses—includingproperty-casualty insurance, life insur-ance, investment, finance, financial ser-vices, and aircraft leasing—we’ve beenskeptical (for several years) of AIG’s abil-ity to accomplish that. Consequently,we’ve felt that the risk in owning AIG’sstock was greater than the reward.

Price-earnings ratios and cyclicityaside, acquisitions are one way for AIG togoose its earnings, at least for a while. But,as we observed, “AIG is so large that it’sdifficult for it to make acquisitions that, bythemselves, materially alter its growthrate. At the margin, however, if it can useits stock to buy lower-multiple companies,then it can eke out incremental growth viaan arbitrage of earnings multiples.”

AIG’s proposed takeover of AmericanGeneral would be an example of such anarbitrage.

Hostile?Although AIG’s offer for American

General was unsolicited, there’s somequestion as to whether it’s “hostile,” andwhether AIG engages in hostiletakeovers. The New York Times reported

that Greenberg said his offer was “nothostile.” The Wall Street Journal statedthat “AIG has never pursued a hostiletakeover.”

One could get into a long discussionof what “hostile” means, which we aren’tinclined to do. However, a deal is gener-ally considered hostile if the CEO of thetarget doesn’t want to be taken over—regardless of whether the deal is good forshareholders. We don’t care if a deal ishostile or not, and neither do sharehold-

SCHIFF’S




AIG’s ‘Hostile’ TakeoverAttempt for American GeneralIs AIG’s Stock Too High?

Average Year annual p/e ratio1972 32.61973 28.21974 17.51975 161976 12.41977 9.21978 8.91979 8.81980 8.71981 9.61982 9.71983 11.61984 15.51985 17.11986 15.61987 13.71988 9.21989 11.21990 10.91991 12.11992 12.61993 14.41994 13.21995 14.51996 161997 19.81998 26.71999 28.88/18/00 38.44/5/01 31.9

The rise and fall and rise of AIG’s p/e ratio.

AIG’s Price-Earnings Ratio

2 APRIL 5, 2001 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

ers. They generally care about whichdeal gives them the best value.

Whether AIG engages in “hos-tile” deals, however, is a subjectworth a few paragraphs. For exam-ple, AIG has been gradually increas-ing its ownership in 21st CenturyInsurance, and now has 63% of thecompany. Shareholders might right-ly view AIG’s accumulation as a“creeping takeover”—one in whichit gains control without paying thecontrol premium that a tender offerfor the entire company would neces-sitate.

AIG has also struck fear in thehearts of insurance companies in thepast. In 1974, American Reinsurancefiled suit to prevent AIG from buyingmore than 9.9% of American Re’sstock. In 1979, Mission InsuranceGroup rejected an unsolicited mergeroffer from AIG (which then owned4.5% of Mission), stating that thedeal was “not in the best interest ofMission and its stockholders.”(Mission was wrong.)

In 1981, AIG disclosed that it hadacquired 8.53% of USLife, promptingthat company’s chairman, GordonCrosby, to state that USLife’s board wasopposed to any attempt to take over thecompany, and that it was in USLife’s bestinterest to remain independent. In 1982,AIG sold its USLife shares back toUSLife. (In 1997, USLife was acquiredby American General in an all-stocktransaction.)

In 1983, AIG bought 8% ofProgressive and was planning to pur-chase a 12.3% stake held by AmericanFinancial. This threat prompted a groupof Progressive shareholders who held a39% interest in the company to form abloc opposing AIG’s accumulation ofProgressive shares. As a result, AIG can-celled its agreement to buy AmericanFinancial’s 12.3% stake, and AmericanFinancial subsequently sold these sharesback to Progressive.

The American Re, Mission, USLife,and Progressive situations differed fromthat of American General in at least onerespect: none of those companies wasalready “in play,” and AIG would havehad considerable difficulty accomplish-ing a takeover that was unwanted bythose companies. (In order to acquire aninsurance company—especially one with

licenses in many states—the approval ofeach state’s regulator is generallyrequired. A hostile insurance takeover istime consuming, and the regulatory road-blocks can make a deal impossible.Allegheny, for example, was unable totake over St. Paul.)

A final thought: Greenberg hadbreakfast with American General’s CEO,Robert Devlin, six months ago and,according to Greenberg, there was sup-posed to be some follow-up, but it neveroccurred. One presumes that if Devlinhad wanted AIG to acquire AmericanGeneral, then he’d have picked up thephone and asked Greenberg to make abid.

Anyway, Hank Greenberg is a genius,and if he says that his unsolicited offer tobuy American General isn’t “hostile,”then who are we to disagree?

Thoughts on SpeculationBefore discussing this deal further, we

want to step back and examine the cur-rent stock-market environment, specula-tion, and p/e multiples, because theseaffect AIG’s ability to complete a deal,and because they’re driving forces in theindustry.

We conducted a Dow Jones NewsRetrieval search to see how many timesthe words “stock,” “market,” and “bub-

ble” appeared in articles duringMarch. The number—1,710—wassizable, apparently demonstratingthat reporters are good at identifyinga stock-market bubble after it hasburst. (In March 1999, for instance,these words appeared one-third asoften as they did this past March.)

Although we labeled “Internet-stock mania” a “speculative bubble”in our March 1999 issue, we didn’tprofess to know when it would end,even though we had thoughts abouthow it would end. As we wrote,“Whether one chooses to call thecurrent U.S. economic environmenta boom, bubble, bull market, or newera, it will, in all likelihood, be fol-lowed by what will be known as abust, bear market, recession, ordepression.”

While our call was accurate, itwasn’t necessarily something onecould profit from. Indeed, the priceof Internet and tech stocks contin-ued to rise sharply for the next 12

months. In December 1999 we noted that

Yahoo’s market cap—then $93 billion—was equal to those of Marsh & McLennan,Allstate, Cigna, Hartford, Chubb, St. Paul,and Progressive combined.

Things have changed. Yahoo is nowvalued at $7 billion, while the insurancecompanies are worth $25 billion, $30 bil-lion, $16 billion, $14 billion, $12 billion,$9 billion, and $7 billion, respectively, ora total of $113 billion.

How could Yahoo, which had $1.1 bil-lion in revenues in 2000, ever carry a $93billion valuation? (Indeed, one must makevery optimistic assumptions to justify thecompany’s current valuation.) The answeris that Yahoo’s valuation was wildly specu-lative, and represented investors’ frenziedand unwarranted optimism about thecompany’s long-term prospects. Yahoo waspriced for permanent perfection, andwhen that didn’t materialize, its absurd p/eratio gave the company a long way to fallbefore it would be priced rationally. AsJames Grant, editor of the marvelousGrant’s Interest Rate Observer recentlywrote, “Booms don’t last forever: they arecut short by their own excesses…However, busts, too, generate excessesthat tend to hasten cyclical reversals, or atleast to exaggerate their magnitude oncethey start.” continued

Internet 12/10/99 01/14/00 10/13/00 04/04/01America Online $205 $141 $123 $155*Yahoo 93 93 33 7Amazon 36 22 10 3CMGI 23 30 5 0.65

eBay 21 17 15 8.2E*Trade 9 7 4 1.8InsWeb 1 0.7 0.06 0.04Quotesmith 0.2 0.2 0.03 0.006

Insurance 12/10/99 01/14/00 10/13/00 04/04/01AIG $172 $177 $214 $179Marsh & McLennan 24 28 32 25Allstate 22 19 24 30Cigna 15 15 18 16

Hartford 10 10 16 14Chubb 9 10 13 12Progressive 6 5 5 7W. R. Berkley 0.6 0.5 0.8 1.3

*Valuation is after stock merger with Time Warner

Market caps of various companies, in billions of dollars.

E-Madness: Internet vs. Insurance—An Update


We bring up Yahoo not just becausewe’ve written about it in the past, butbecause it’s a good example of anextreme. Financial history is filled withcompanies that sported wildly high valu-ations during periods of mass euphoria,and depressed valuations (or no valua-tion) during the ensuing busts.

The boom-and-bust cycle isn’t limit-ed to technology stocks—over the yearsit has embraced virtually every industry,from automobiles, oil & gas, telephones,utilities, and conglomerates, to electron-ics, specialty retail, entertainment,restaurants, finance, and, yes, insurance.

All of which brings us back to AIG.We first expressed concern about thecompany’s p/e and price-to-book ratioback in October 1998 when its stockprice was $49—$27 lower than it is now.We revisited the subject in ourSeptember 2000 issue, when AIG’s stockwas $86. Although AIG’s excellentrecord of earnings growth is one of thefactors in its stock’s superior returns, itisn’t the only factor. AIG’s p/e ratio hasbeen in a long-term bull market of itsown; more than quadrupling since itsbottom in 1979.

A recent Merrill Lynch studyshowed that since 1980, AIG’s averagep/e ratio based on forward consensus esti-mates has been 13.8. The lowest p/eratio—6.8—was recorded in June 1982,and the highest—35.1—occurred inDecember 2000. Perhaps coincidental-ly, AIG’s average p/e, according to theMerrill study, is not very different fromthe S&P 500’s average p/e over the last129 years—14.5.

If one can infer anything about valua-tions from the past it is this: they fluctu-ate considerably. In 1929, for example,the Dow Jones Industrial Average(DJIA) was priced at 4.5 times book

value. Three years later, when the DJIAhit its all-time low, it was valued at one-half of book. (The p/e ratio wasn’t mean-ingful in 1932, as the companies in theDJIA lost money.)

Although the S&P 500’s p/e ratio hasaveraged 14.5 over the long term, stockshave often traded way above, or waybelow, that figure. Valuations, however,have historically reverted to the mean,and then some. Every period in whichstocks have traded in excess of a 14.5multiple has been followed by a periodin which valuations fell well below thatfigure. History, of course is just a guide,not a blueprint for the future. The pastdoes not have to repeat itself.

Thus, the history of AIG’s valuationdoesn’t foretell how AIG’s stock will bevalued in the future. Nonetheless, thepast is still worth considering. In 1972,AIG’s p/e ratio was 32.6—about what it istoday. Despite the fact that AIG’s earn-ings continued to rise steadily, AIG’sshares lost two-thirds of their value overthe next two years, and AIG’s stock pricedidn’t get back to its 1972 high until1978—even though earnings hadquadrupled and book value had tripledduring that period.

AIG is a great company, but there’sconsiderable risk in owning a finan-cial-services company selling for 32times earnings. AIG’s high valuationleaves little room for error or disap-pointment.

In order for AIG’s shares to appreci-ate, two things must happen: earningsper share must grow, and the p/e ratiomust remain the same or go higher.Steadily rising earnings per share areessential because investors, in anticipa-tion of such, have bid up AIG’s stock toan extreme p/e ratio, which, of course,facilitates AIG’s use of its stock to

acquire lower-multiple companies, thusproviding a boost to earnings per share.As AIG gets larger—and it is alreadyhuge—greater than average growthbecomes more difficult.

While it’s wise for AIG to use its high-multiple stock to make acquisitions, oneconcerned with security analysis mustask a basic question: if AIG, which tradesat 32 times earnings, buys AmericanGeneral for 18 times earnings, shouldAIG’s 32 multiple be applied toAmerican General’s supposedly lower-growth business once that businessbecomes part of AIG? According toGreenberg, the answer is yes. At yester-day’s conference call he spoke of cross-marketing and cost savings, and said,“I’m comfortable that two and two herewill make five, if not seven.”

In the 1960s, under the guise of “syn-ergy”—a 2+2=5 equation—conglomer-ates, which had staggeringly high p/eratios, acquired diverse, lower-multiplebusinesses including bakeries, foundries,machine shops, and insurance compa-nies. For a while, the market was willingto apply the conglomerates’ high p/emultiples to the earnings acquired fromthe acquisition of slower-growth busi-nesses. Eventually, however, the merry-go-round came to a halt.

In theory, AIG—or any business witha high p/e ratio—can be a perpetualgrowth machine by endlessly performingthe arbitrage of using its high p/e stock toacquire earnings that are selling at alower p/e. In practice, this is difficult todo, and, of course, is dependant upon,among other things, always having a highp/e multiple.

Investors in AIG would do well toremember that AIG, which traded at 32.6times earnings in 1972, traded at 8.7times earnings in 1980, 9.2 times earn-ings in 1988, and 13.2 times earnings in1994.

Although Yahoo traded at 100 timesrevenues last year, we doubt that AIG’sp/e ratio has much room for expansion.Absent any change in the p/e,investors’ returns will mirror AIG’sgrowth, which many analysts peg atabout 15% annually.

If that growth fails to materialize forsome reason—or if earnings actuallydecline, as they did in 1984—it’s a safebet that AIG will trade at a much lowermultiple. �

Editor and Writer . . . . . . . . . . . David SchiffCopy Editor . . . . . . . . . . . . . . . . . John CaumanEditorial Associate . . . . . . . . . . Isaac SchwartzProduction Editor . . . . . . . . . . . . . . Bill Lauck

Publisher . . . . . . . . . . . . . . . . Alan ZimmermanCustomer Service Director . . . . . . Pat LaBuaAdvertising Manager . . . . . . . . . Mark Outlaw© 2001, Insurance Communications Co., LLC.All rights reserved.


Publishing HeadquartersSchiff’s Insurance ObserverSNL c/o Insurance Communications Co.321 East Main StreetP.O. Box 2056Charlottesville, VA 22902Phone: (804) 977-5877Fax: (804) 984-8020E-mail: [email protected]



In its reports for the years ending2001 and 2000, AIG’s audit commit-tee disclaims virtually all responsi-bility for AIG’s accounting, internal

controls, and financial statements. It alsosays that it cannot assure that AIG’s inde-pendent accountants are actually “inde-pendent.” (The most recent audit-com-mittee report is on page 17 of AIG’sproxy statement.)

If AIG’s audit committee can’texpress an unqualified opinion aboutAIG’s accounting, doesn’t it makesense that the public’s faith in AIG’saccounting should be somewhat dimin-ished? And, if the public’s faith isdiminished, isn’t it reasonable toexpect AIG’s stock to trade at a lowermultiple of earnings than it would oth-erwise trade?

Before discussing these issues, we’llnote that AIG has been the greatest suc-cess story in the insurance business. It’sthe largest, most important insuranceorganization in the world. The story of itssuccess, however, is not readily available.Although Hank Greenberg is a legend, hisachievements have not received wide-spread attention. Jack: Straight from the Gutis on the best-seller list; Hank: Straightfrom 70 Pine Street, will probably not bewritten.

We have great admiration forGreenberg (given his record, it’s hard notto), and are planning to write a lot aboutAIG in the coming months. Althoughwe’d prefer to write chronologically, pub-lishing constraints make this difficult.Thus, this article focuses on currentissues rather than on AIG’s 1969 exchangeoffers or Greenberg’s letters to sharehold-ers in the 1970s, even though all of thesesubjects are of equal interest to us.

In the post-Enron Era, the minutia ofaccounting principles have becomeof greater concern to many. Investors,

having recently seen several trillion dol-lars of stock-market value melt like but-ter on a hot skillet, are more skeptical ofcompanies whose finances are complexor opaque—even those companies withfine long-term records. This wariness islogical; if you can’t understand a businessand analyze its financials, how can youplace a value on the company?

This was not a question asked oftenenough during the great bull market,when the “extrapolation method” ofanalysis was sufficient for many“investors.” (They would take recentyears’ reported earnings and project thesame growth rate for many years into

the future.) This method had its advan-tages: it was really simple and saved alot of time that would have otherwisebeen spent reading balance sheets,cash-flow statements, and footnotes.

The extrapolation method has adrawback, however—it doesn’t work.The footnotes, fine print, and SEC-man-dated disclosures are there becausethey’re important. Words really meansomething, and when a company sayssomething unusual—or doesn’t saysomething usual—one should take thatinto consideration.

AIG has a long record of growth, butthe market’s opinion of its growth hasvaried. In 1988, AIG’s stock traded at anaverage of 9.2 times earnings. ByDecember 8, 2000, when the stock hit an

SCHIFF’S


May 2, 2002Volume 14 • Number 7 I N S U R A N C E O B S E R V E R


The Greatest Risk is Taking Too Much Risk

Hank Greenberg stays ahead of his competitors.

AIG’s Audit-Committee Report

2 MAY 2, 2002 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

all-time high of $103.75 (it is now$71.51), the p/e ratio had quadrupled to42. Such a multiple is difficult to justifyin any company, much less one so largethat its future growth rate cannot possi-bly match its past.

What is the proper multiple for ahighly complex, international financial-services conglomerate whose businessesare cyclical? We don’t know—nor doesanyone else—but the lower the multiple,the more appealing we find the stock.

In 2001, AIG’s earnings did some-thing that not one of the dozens of ana-lysts following the company expected—they declined. The decline, the firstsince 1984, was a reminder that even thegreatest companies are not immune tothe vicissitudes of business. Investors,

however, don’t like being reminded thatthe earnings of “growth” companies donot always grow. (While a growthcompany’s failure to grow may be irk-some to growth-stock investors, it is notnearly so irksome as the failure of agrowth company to maintain solvency—the condition that afflicted Enron.)

According to the SEC, “Audit com-mittees play a critical role in thefinancial reporting system by over-

seeing and monitoring management’sand the independent auditors’ participa-tion in the financial reporting process.”Financial statements are prepared bymanagement and audited by independentaccountants.

PricewaterhouseCoopers, AIG’s inde-pendent accountants, says that it con-ducted its audit of AIG in accordancewith generally accepted standards, andthat the audit provides a reasonable basisfor its opinion that AIG’s financial state-ments present the company’s financialcondition fairly, in all material respects.This is standard lingo found in virtuallyevery financial statement.

AIG’s audit-committee report, how-ever, provides an opinion that’s ambigu-ous, elusive, equivocal, hedged, andoblique—qualities that aren’t particular-ly comforting to investors or creditors.(Perhaps the only outside parties thatwould like the wording in the audit-com-mittee report are the company’s D&Oinsurers.) The report does not containthe same language found in many otheraudit-committee reports. In fact, AIG’saudit committee’s disclaimers are soextensive that they render the report vir-tually meaningless.

The key paragraph in AIG’s audit-committee report follows. We’ve addeditalics for emphasis:

The members of the [Audit] Committee arenot professionally engaged in the practice ofauditing or accounting and are not experts inthe fields of accounting or auditing, including inrespect of auditor independence. Members ofthe Committee rely without independent veri-fication on the information provided to themand on the representations made by manage-ment and the independent accountants.Accordingly, the Committee’s oversight does not pro-vide an independent basis to determine that manage-ment has maintained appropriate accounting andfinancial reporting principles or appropriate inter-nal controls and procedures designed to assurecompliance with accounting standards andapplicable laws and regulations. Furthermore,

the Committee’s considerations and discussionsreferred to above do not assure that the audit of AIG’sfinancial statements has been carried out in accor-dance with generally accepted auditing standards,that the financial statements are presented in accor-dance with generally accepted accounting principlesor that AIG’s auditors are in fact “independent.”

The disclaimers in AIG’s audit-com-mittee report aren’t common. PerhapsAIG is on the cutting edge, however, andin years to come more audit committeeswill adopt similar verbiage.

Viewed by itself, AIG’s audit-com-mittee report is not such a big deal. Butviewed in the context of AIG’s inherentcomplexity and the inherent imprecisionof insurance-company “earnings,” ittakes on greater meaning and is worththinking about.

AIG’ stock has declined more than30% from its all-time high, and isnow trading at the price it was

three years ago—despite the fact that thecompany is expected to produce recordearnings this year. On many occasions,AIG has benefited from having a highp/e ratio; it has been able to use its stockto make acquisitions on attractive terms.Its current p/e ratio (about 20 times pro-jected earnings) reduces the possibilityof most stock acquisitions because theeffect of issuing stock at this level (rela-tive to what AIG would receive inreturn) would probably be dilutive toearnings rather than accretive.

None of this is lost on HankGreenberg, who seemingly knows every-thing. He is acutely aware of the impor-tance of financial strength as well as theimportance of perception. If, for example,people perceive—correctly or incorrect-ly—that AIG does not pay claims, it will, atthe margin, hurt AIG’s business. If AIG’sfinancial strength is perceived as beingweaker than it is, that can become a self-fulfilling prophecy as lenders demandslightly higher spreads, causing the compa-ny’s cost of capital to rise, thereby reducingprofitability. Finally, if AIG’s stock price istainted by Enronesque issues such as com-plexity, lack of transparency, or sheerincomprehensibility, then it stands to rea-son that the stock will trade at a lower mul-tiple of earnings than it would otherwise.

While no one knows with certaintythe reasons why a stock goes down (otherthan the obvious—that sellers were morepersistent than buyers), it appears that

Editor and Writer . . . . . . . David SchiffProduction Editor . . . . . . . . . . Bill Lauck

Publisher . . . . . . . . . . . . Alan ZimmermanSubscription Manager . . . . . . . Pat LaBuaAdvertising Manager . . . . . Mark Outlaw



For questions regarding subscriptions pleasecall (434) 977-5877.


Copyright Notice and WarningIt is a violation of federal copyright law toreproduce all or part of this publication. You arenot allowed to e-mail, photocopy, fax, scan, dis-tribute, or duplicate by any other means thecontents of this publication. Violations of copy-right law can lead to damages of up to $100,000per infringement.Reprints and additional issues are availablefrom our publishing headquarters.




SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 MAY 2, 2002 3

AIG’s stock has been under pressure forseveral reasons: 1) it had been selling atan unusually high multiple; 2) the com-pany reported a decline in earnings lastyear, 3) investors are more concernedabout accounting and complexity thanthey have been in the past; 4) AIG is dif-ficult to understand, and investors are

less willing to accord high multiples tothings they don’t understand; and 5) AIGis a diversified financial company ratherthan a pure play on property-casualty, andtherefore is not benefiting as much assome companies from the turn in cycle.

AIG’s stock price appears to be of consid-erable concern to AIG, and the company has

been attempting to respond to various crit-icisms. For example, it has been faulted forhaving too few “independent” directors. Itsresponse: Bernard Aidinoff, a director since1984, is now “senior counsel” at Sullivan &Cromwell (which represents AIG) ratherthan a “partner.” And Carla Hills, a directorsince 1993, terminated her consultingagreement with AIG in early 2002. Wedoubt that these cosmetic changes willmake Aidinoff and Hills better or worsedirectors than they were before. (Most cor-porate directors aren’t too independent,anyway. If they were, they wouldn’t be puton a board in the first place.)

AIG has now instituted quarterly con-ference calls—the first was held lastweek—and has provided additional dis-closure in its annual report and 10-K. Ithas also attempted to deal with the “suc-cession” issue by creating an Office ofthe Chairman, naming co-chief operatingofficers, and announcing several promo-tions. (The actuaries at Schiff’s think thatGreenberg is in better shape than mostinsurance-company CEOs, and won’tneed a successor for many years.)

It’s impossible to say whether any ofthe changes made by AIG will have anyeffect on the company’s stock price. AsBenjamin Graham famously wrote, in theshort term the market is a voting machine;in the long term it is a weighing machine.

Which brings us to the morning ofApril 22. AIG’s stock was down severalpoints amidst rumors that the companywould miss its second-quarter earnings(it didn’t), and that it was being investi-gated. In the early afternoon, AIG putout the following press release: “AIG’sstock is trading down significantly. Wehave observed considerable short sellingin the stock and have requested the NewYork Stock Exchange and the Securitiesand Exchange Commission to investi-gate this activity.”

Blaming shortsellers for a decline in acompany’s stock is a tactic often used byhighly promotional companies whoseshares are overvalued, and is unusual fora company of AIG’s stature, for many rea-sons. First of all, shortselling is not illegal orunethical. (At year end, AIG was short $8.3billion of securities and commodities.) Sowhy did AIG ask the authorities to inves-tigate? (“No comment,” said AIG.)

If AIG is so concerned about the trad-ing activity in its stock, why didn’t it askthe SEC and NYSE to investigate the

Do someone a favor. Send them a free gift subscription.

Give us the names of one or more people you think would enjoy reading Schiff’s Insurance Observer and we’ll send them

a free gift subscription (four issues) from you. Please send an e-mail to [email protected] and include the name, company, and e-mail address of the people you would like to give

a free gift subscription to. We’ll take care of the rest.

David Schiff looks for new subscribers.



Gift subscriptions will not be sent to current subscribers.

mailto:[email protected]

4 MAY 2, 2002 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

considerable buying (and all the broker-age “buy” recommendations) when itsshares were 50% higher and, apparently,trading under the influence of irrationalexuberance?

Also, how did AIG “observe” shortselling on April 22? (“No comment,” saidAIG.)

AIG’s request that the NYSE investi-gate carries extra weight. HankGreenberg is on the NYSE’s board, andAIG director, Frank Zarb, is the formerchairman of the NYSE’s nominating com-mittee. Section 202.03 of the NYSE’s“Listed Company Manual” provides thefollowing recommendations for dealingwith rumors or unusual market activity:

220022..0033 DDeeaalliinngg wwiitthh RRuummoorrss oorr UUnnuussuuaallMMaarrkkeett AAccttiivviittyy

If rumors or unusual market activity indi-cate that information on impending develop-ments has leaked out, a frank and explicitannouncement is clearly required. If rumors arein fact false or inaccurate, they should be promptlydenied or clarified. A statement to the effect that thecompany knows of no corporate developments toaccount for the unusual market activity can have asalutary effect…[Emphasis added.]

The Exchange recommends that its listedcompanies contact their Exchange represen-tative if they become aware of rumors circu-lating about their company...Information pro-vided concerning rumors will be promptlyinvestigated.

Why didn’t AIG use the standardNYSE comment—that it knows of nocorporate developments to account forthe unusual market activity—in its pressrelease? (“No comment,” said AIG.)

After all the “no comments” we didn’tbother asking AIG if it “observed” any ofthe alleged shortsellers reading a copy ofthe company’s audit-committee report. E

Coming soon in a future issue of Schiff’sInsurance Observer: “The Great Greenbergand the Rise of AIG.”

On May 2 we published an arti-cle about AIG’s audit-com-mittee report. Specifically,we noted that the report’s

elusive, equivocal verbiage made it lit-tle more than an extensive disclaimer—exactly the opposite of what an audit-committee report should be.

Audit-committees reports are a dullsubject. So dull, in fact, that to the bestof our knowledge, no one else in theworld had written about the disclaimersin AIG’s report. (In fairness to AIG, anumber of other large companies usedthe same evasive language.)

Our article caused a stir among insur-ance cognoscenti, and then createdsomething of a commotion when TheEconomist had the good judgment to pickup our story. Although we received posi-tive feedback from many subscribers, wewere amazed that some subscribers—including respected analysts and insur-ance-company presidents—told us thatour observations were out of line. Auditcommittees are not worthy of so muchattention, they said, and it reflectedpoorly on us to be making a big dealabout them.

It seems remarkable that less thanthree months ago learned folks stillbelieved that the numbers in companies’financial statements were sanctified justbecause CEOs and the accountants theyhired set those numbers in type.

Of course, any belief in the inviolabil-ity of corporate accounting disappearedon June 25, when WorldCom’s numericalinnovations became known. That audit-ed financial statements can be manipulat-ed so that losses become profits is noth-ing new. Nor is it new that many compa-nies are run by rapacious scoundrels.

During bull markets investors happi-ly ignore blatant warnings. In our August1999 issue, for example, we commentedon InsWeb, the Internet insurance mar-ketplace that had just gone public andcommanded a $1.5 billion market cap,even though it had virtually no revenuesand expected to “incur substantial oper-ating losses for the foreseeable future.”

Thanks to the Securities Act of 1933,there was no reason for any investor tolose a penny investing in InsWeb. TheSecurities Act—also known as the “truthin securities” law—requires issuers toprovide investors with meaningful dis-closure. InsWeb dutifully carried out itsresponsibility, and warned investorsabout the toxicity of its common stock.The “risk factors” section of its prospec-tus came in at 8,477 words, which maybe a record. (InsWeb’s stock is nowdown 99%.)

In our May 2 article, we questionedwhether the failure of AIG’s audit com-mittee to express an unqualified opinionabout the company’s accounting wouldcause AIG’s stock to trade at a lower mul-

tiple of earnings. (AIG stock was then$71.51; it is now $53.38.)

Before we delve further into AIG’saccounting and audit-committee report,the SEC, and related subjects, we wantto make sure that readers put ourthoughts in perspective. Over the yearswe’ve written about a dozen articles onAIG. We’ve commented on its success,complexity, mergers and acquisitions,and p/e ratio. In late 1994 we wrote thatAIG’s stock was cheap and that we’dbought it. (We sold it several years later.)In 1998 and 2000, we noted that AIG’sp/e ratio was so high that the stock pricehad scant margin of safety. We’ve alsowritten about companies that AIG hassubsequently acquired (SunAmerica),and about AIG’s mysterious offshorereinsurance transactions (Coral Re).

There are many reasons to writeabout AIG, not the least being that it isthe largest, most important, and greatestworldwide insurance organization. AIG,by virtue of its size, scope, “AAA” rating,and nature is a fabulous (and fabulouslycomplex) company. It is not, however,

SCHIFF’S


July 25, 2002Volume 14 • Number 10 I N S U R A N C E O B S E R V E R


AIG, Audit Committees, Legends, and P/E RatiosThe Tao of Hank, Part 1

AIG’s Audit Committee Report: Caveat Emptor

The key paragraph in AIG’s audit-committee report follows. Italics have been added for emphasis.

The members of the [Audit]Committee are not professionally engagedin the practice of auditing or accountingand are not experts in the fields ofaccounting or auditing, including inrespect of auditor independence.Members of the Committee rely withoutindependent verification on the informa-tion provided to them and on the repre-sentations made by management and theindependent accountants. Accordingly, theCommittee’s oversight does not provide an inde-pendent basis to determine that management

has maintained appropriate accounting andfinancial reporting principles or appropriateinternal controls and procedures designedto assure compliance with accounting stan-dards and applicable laws and regulations.Furthermore, the Committee’s considerationsand discussions referred to above do not assurethat the audit of AIG’s financial statements hasbeen carried out in accordance with generallyaccepted auditing standards, that the financialstatements are presented in accordance withgenerally accepted accounting principles or thatAIG’s auditors are in fact “independent.”

2 JULY 25, 2002 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

easy to understand, and cannot be fullyunderstood by an outsider. (Actually, itcannot be fully understood by an insider,either, but that’s probably true of everygiant multinational.) AIG’s history—which we’ve been researching for sometime—is a story of entrepreneurship, dar-ing, audacity, internationalism, and capi-talism. It is a remarkable feat that in 40years or so, AIG, which was a loosely-knit group of foreign underwriting agen-cies, life insurers, and second-ratedomestic insurers—managed to eclipse,by a wide margin, the titans of yesteryear:Aetna, CNA, Connecticut General,Continental, The Hartford, The Home,INA, Metropolitan, Prudential, Travelers,and USF+G. Today, AIG is worth muchmore than all these combined.

Hank Greenberg, who has led AIGfor the last 33 years, is admired,respected, and feared. Greenberg,despite his 77 years, is not mellow; he’sintense and competitive. He’s alsocharming, charismatic, funny, anddeeply concerned about every aspect ofhis business. He’s filled with energyand enthusiasm, and, despite hisinvolvement with big issues around theglobe, seems easily aggravated bydetails so small you wouldn’t expect theCEO of one of the world’s largest com-panies to pay attention to them.Greenberg’s attention to minutia doesnot seem to have hurt his company’sresults. Perhaps it has even contributedto his success.

If there’s anyone in the industry whocan be considered a living legend, it isGreenberg. This status was dramatizedat Schiff’s Insurance Conference in April, atwhich he was the first speaker. AfterGreenberg had talked for almost anhour without notes, he was asked a goodquestion: “How do you spend yourday?” He gave an answer that interestedour hardboiled, skeptical audience. (Wewon’t repeat it; you just had to bethere.) It is unimaginable that the sameaudience would exhibit much curiosityabout how other insurance CEOs spendtheir days.

Why do insurance mavens care whatGreenberg does all day? We carebecause, in an industry where it’s so easyto go awry and so hard to excel, AIG hasaccomplished what no other companyhas. Watching Greenberg’s performanceis akin to watching a sleight-of-hand

artist who makes cards appear and disap-pear. Although you know the legerde-main isn’t magic—it’s the result of prac-tice and hard work—it seems like magic.

“When the legend becomes fact,print the legend,” says the newspapereditor at the end of John Ford’s elegiacWestern, The Man Who Shot LibertyValance. But separating legends fromfacts is often impossible. “Once a news-paper touches a story, the facts are lostforever,” Norman Mailer wrote, “even tothe protagonists.” So we all read about theGreenberg of legend: the World War IIand Korean War veteran who’s tough,hard-driving, combative, and intolerantof failure. There is, of course, much moreto him.

Greenberg is a disciplined man. He islean and fit, and his posture is perfect.He is careful about what he eats andexercises regularly. He appears to havelittle interest in the trappings of extremewealth. He doesn’t have the fanciesthomes or the biggest art collection, andhis name doesn’t appear in societycolumns. He wears conservative suits,button-down shirts, and an inexpensivewatch. He loves to ski and play tennis.He can recall names and details from 50years ago.

Schiff ’s has gotten to know manyinsurance CEOs reasonably well over theyears. One could say that they all have areason to talk to us: to attempt to influ-ence us or to get on our good side (thepresumption being that we actually havea good side). Out of all those CEOs, wehave never met anyone who has been asopen as Greenberg.

And yet, there are many Wall Streetanalysts who are terrified of him becausethey believe that if he wanted to, he couldcause them to be fired. This may or may notbe true, but if it is widely believed, thenisn’t the effect the same as if it weretrue?

Many of Greenberg’s competitors—sane, successful men—are also afraid ofpissing him off because—mind you, thisis just one example—he controls the NewYork Department of Insurance and could getthem tied up in a regulatory morass.Whether he really controls the depart-ment is irrelevant to the perception thathe does. The effect on his competitorsis the same. (When we discussed thiswith him, Greenberg scoffed at thenotion that he controls the insurance

department, and grumbled somethingabout how long it takes AIG to get fil-ings through.)

The foregoing brings us back toour May 2 article about AIG’saudit-committee report, and our

musings about the effects that issues ofcomplexity and transparency have onAIG’s stock price and p/e ratio.

The gist of our article was that AIG’saudit committee, in its reports for theyears ending 2001 and 2000, used atypi-cal—and in our view, inappropriate—lan-guage: “The [audit] committee’s over-sight does not provide an independentbasis to determine that AIG’s manage-ment has maintained appropriate inter-nal controls and procedures,” statedAIG’s audit-committee report. “Thecommittee’s considerations…do notassure that the audit of AIG’s financialstatements has been carried out in accor-dance with generally accepted auditingstandards…or that AIG’s auditors are infact ‘independent.’” [Emphasis added.]

The audit-committee’s verbiageprompted us to pose two questions: 1) IfAIG’s audit committee (which, like allaudit committees, is comprised of “inde-pendent” directors), can’t express anunqualified opinion about AIG’saccounting, doesn’t it make sense thatthe public’s faith in AIG’s accountingshould be somewhat diminished, and 2)if the public’s faith is diminished, isn’t itreasonable to expect AIG’s stock to tradeat a lower multiple of earnings than itwould otherwise?

The first question is more important,because if the answer to it is “No,” thesecond question becomes moot. Since itis a fact that AIG’s audit-committeereport contains caveats that render it vir-tually meaningless, we are faced with theinevitable question: Will these caveatsdiminish the public’s faith in AIG’saccounting?

There are reasons why one couldanswer “No”: 1) Some other large com-panies use identical language, and, per-haps, hundreds use similar language; 2)The caveats are there for legal reasons; 3)The financial statements are prepared bymanagement and audited by outsideaccountants; 4) The audit committeemerely plays an “oversight” role.Assurance about the financial state-ments comes from the outside accoun-

SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 JULY 25, 2002 3

tants in the “Report of IndependentAccountants” and from AIG’s manage-ment in the “Report of Management’sResponsibilities;” 5) The audit commit-tee can’t be expected to provide assur-ance that the financial statements con-form to GAAP or that the accountants areactually independent; 6) The audit com-mittee is comprised of respectable peo-ple; and, 7) One would have to be out ofhis mind to think that anyone gives adamn about audit-committee reports.

These responses are reasonableenough, but we continue to doubt they’llsatisfy every thoughtful, intelligentinvestor. If that is correct, then it’s rea-sonable to assume that the caveats inAIG’s audit-committee report have someeffect on AIG’s p/e multiple, even if thateffect is slight. We don’t know of any way

to estimate what the effect will be or howto measure it.

As we mentioned in our previous arti-cle, in the post-Enron (and now, post-WorldCom) era, accounting minutia areof greater concern to many. By itself,AIG’s audit-committee report is not asmoking gun. However, no one viewsanything by itself. The audit-committeereport is one piece of a large puzzle. Onone hand there’s AIG’s great history andstrong businesses; on the other handthere’s the company’s inherent “black-box” complexities. Investors, for goodreasons, are now more wary of complexi-ty—and of things they don’t understand.

AIG’s caveat-filled audit-committeereport is a farce, and AIG’s board made amistake when it accepted it. Perhaps itdidn’t understand that the times werechanging and AIG’s stock, which hadtraded at an unusually high p/e multiplefor many years, was vulnerable. We wrotenumerous times over the last four yearsthat the risk of buying AIG’s shares at astratospheric p/e multiple outweighedthe reward.

AIG’s p/e multiple, which had been asingle-digit figure for much of the 1970sand 1980s, rose sharply after 1988, whichboosted the increase in AIG’s stock priceover the years. (The p/e multiple even-tually peaked at about 42—a figure thatleft virtually no margin of safety.) AIG’sstock has been declining for a year and ahalf. Viewed another way, the company’sp/e multiple has been contracting.

Beginning next month, HankGreenberg will file a sworn written state-ment with the SEC personally attestingthat AIG’s financials are materially truth-ful. (Officers at 945 large companies mustfile the same statement for their compa-nies.)

Beginning next year, we expect AIG’saudit committee to drop the caveats anddisclaimers in its report. While that won’tmake AIG easier to analyze, it will makethe audit committee more responsiblefor its work. That can only be a goodthing. E

To be continued. Part 2 of this article willprobably appear early next week.

A couple of months ago, when discussingAIG with Greenberg, we said that when valu-ing the company we put a lower multiple onearnings from GICs than on other earnings.Greenberg’s response: “I don’t think youshould value the company based on the com-

ponents of its earnings. It’s the diversificationof earnings that’s important. That’s whatmakes AIG. It’s the totality...not the pieces.AIG is a great company with an unparalleledfranchise. You couldn’t put it together today ifyou wanted to.”

We don’t use the same method to valueAIG that Hank does, but we do agree with hissentiments. Although we don’t think AIG’sstock is a bargain, in the interest of full dis-closure we must admit that we became ashareholder yesterday. We paid $49 pershare—14 times this year’s projected earn-ings. That’s higher than we like to pay, and itgives our investment a more speculative char-acteristic than we ordinarily prefer.

Unlike stockbrokers, who rate a stock a“buy” and then list a much higher targetprice, we tend to think about how much lowera stock must go before we buy more. Rightnow we’re planning to double our investmentwhen AIG hits 39. Of course, we may changeour opinion. If we do, it is highly unlikely thatwe will notify you at that moment.





For questions regarding subscriptions pleasecall (434) 977-5877.






American International Groupcan run but it can’t hide, andHank Greenberg knows that.Times have changed, and AIG

is trying to change with them. Thus, inthe new corporate spirit of opennessand transparency, AIG has held its firstquarterly conference call to discuss itsearnings, created an office of the chair-man, made two of its so-called “inde-pendent” directors more “indepen-dent,” ran an all-day meeting forinvestors, provided new disclosures inits annual report and 10-K, andannounced that it will expense stockoptions beginning next year.

Most of these changes are cosmetic—form over substance—but they’re posi-tive and make good sense for AIG which,due in part to its complexity and inher-ent impenetrability, is now viewed withconsiderably more skepticism than it hasbeen for many years.

For our money, however, the mostsignificant change that AIG will make isone that hasn’t been reported: it will alterits audit-committee report in next year’sproxy statement.

For the past two years AIG’s board ofdirectors has accepted—and fobbed off onshareholders—audit-committee reportsthat were evasive, equivocal, and not inkeeping with the spirit of last year’s SECrequirement that an audit-committeereport be included in public companies’proxy statement.

Beginning next year, AIG’s audit-committee report will, apparently, con-tain a positive opinion about AIG’s finan-cial reporting rather than a disclaimerdesigned to insulate AIG’s directors fromresponsibility. Greenberg, who is con-cerned with transparency and appear-

ances for many reasons (not the leastbeing that the perception that there’ssomething to hide affects the company’sstock price and access to capital), told ushe will “insist” upon a better audit-com-mittee report. “I don’t think anyone paidmuch attention to it,” he said, referringto the myriad qualifications in AIG’saudit-committee report. “We relied onoutside counsel. In retrospect, that was amistake.” Greenberg, who’s been in theinsurance business for 50 years, didn’tbecome The Great Greenberg by lettingmistakes go uncorrected.

For those who don’t recall the May 2and July 25 issues of Schiff’s, we’ll pro-vide a brief reminder: AIG’s audit-com-mittee report, as it is now written, is notan endorsement of the company’saccounting; rather, it’s a legal disclaimerfor the audit committee. “The [audit]

committee’s oversight does not providean independent basis to determine thatAIG’s management has maintainedappropriate internal controls and proce-dures,” states the audit-committeereport of the world’s most valuable insur-ance organization. “The committee’sconsiderations…do not assure that theaudit of AIG’s financial statements hasbeen carried out in accordance with gen-erally accepted auditing standards…orthat AIG’s auditors are in fact indepen-dent.”

If the audit-committee of the compa-ny that believes that the greatest risk isnot taking one can’t state that AIG’sfinancial statements conform withGAAP, then who needs the audit com-mittee? If the audit-committee can’tdetermine whether or not AIG’s auditorsare “independent,” then the members of

SCHIFF’S


August 16, 2002Volume 14 • Number 11 I N S U R A N C E O B S E R V E R


AIG to Change Audit-Committee ReportDon’t Look Back

“Your mother is responsible for your fear of Generally Accepted Accounting Principles.”

2 AUGUST 16, 2002 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

the audit committee should be replacedby people who can make such a determi-nation.

That AIG, which is worth approxi-mately $175 billion, will change its audit-committee report is a testament to thechanging times. If the stock-marketboom at the end of the last millenniumqualified as a “new era,” then thepresent climate of skepticism andcorporate accountability is anothernew era. (The future, of course, iscomprised of nothing but “neweras.” They come and go all thetime.)

After our May 2 article, wereceived complaints from some sub-scribers, including securities analystswho were recommending AIG’sstock. (According to First Call, 22 of23 analysts covering AIG rate it a“buy.”) Schiff’s was destroying publicconfidence, we were told, by writingabout AIG’s audit-committee report,especially during a time when themarket was so volatile. We were alsotold that the audit committee is onlyan overseer: it hires the accountantsbut doesn’t really have access to finan-cial information.

A company’s board of directorsand audit committee have broadauthority. The audit committee canmeet alone with the internal finan-cial people and the outside accountants,request whatever information it wants,conduct investigations, hire outsidecounsel, and bring in other accountantsor experts if it needs to. (AIG’s auditcommittee met seven times in 2000 andfour times in 2001.)

We were also told that the audit com-mittee shouldn’t really be held account-able because, in the end, it relies onmanagement and the auditors. But thatargument leads to a web of deniability inwhich no one is accountable: the auditcommittee relies on management andthe accountants, the accountants rely onmanagement, management relies on theaccountants, management’s financialstatements are approved by the board,the board relies on the audit committee,the audit committe relies on manage-ment and the accountants...

A company’s board of directorsshould serve as an overseer; it hires (orfires) the CEO, and approves budgets,major capital expenditures, acquisitions,

divestitures and many other corporateactions. It has been said that a directorshould be a skeptical ally of manage-ment. Directors should have knowledge,background, and skills sufficient to allowthem to perform their job well.Furthermore, they should devoteenough time to carry out their responsi-

bilities, and should have the tempera-ment to speak out and act independent-ly, regardless of the consequences.

In practice, many directors don’tmeet this standard. Boards are filled withyes-men (and token yes-women) who letCEOs do what they want. In return fordoing little, directors get paid well, makeuseful business connections, and gainstatus that generally benefits them insome way or other. This is how it goes atmost public companies, mutual funds,and money-market funds. Althoughdirectors are elected by the shareholders,shareholders don’t usually get involved.This is particularly true of mutual funds,which rarely make an issue of corporategovernance, probably because theyemploy the same corporate structure asthe companies they invest in.

We approve of AIG’s change regard-ing its audit-committee report, butwon’t give a giant company a pat on theback because it decides to operate in amore forthright manner. (Also, we

don’t know exactly what changes AIGwill make. The new audit-committeereport will appear in the proxy state-ment, which will be distributed nextApril.) Shareholders and policyholdersshould expect the highest standardsfrom AIG.

Although Hank Greenberg gave littleor no thought to AIG’s audit-commit-tee report before we wrote about it,the same cannot be said about everymember of the audit committee.Before proceeding, however, a briefhistory of why audit-committeereports began appearing in proxy state-ments in 2001 is in order.

In 1895 the New York StockExchange recommended that listed com-panies give their shareholders anannual report that included a balancesheet and income statement. Fiveyears later this became a requirementfor companies seeking a new listing.The Securities Act of 1933 and theSecurities Exchange Act of 1934 man-dated important disclosure and createda regulatory authority—the Securitiesand Exchange Commission. Over thenext 68 years, shareholders, activists,gadflies, corporate raiders, legislators,and regulators would seek to makecompanies more accountable, and theaccounting they used more acceptable.It was not until 1977, however, that the

New York Stock Exchange required list-ed companies to have an independentaudit committee comprised of “out-side”—but not necessarily “indepen-dent”—directors.

Change often happens slowly thensuddenly, and nothing can permanentlyalter investors’ mood swings betweengreed and fear. But disclosure, reform,and good regulation can protect intelli-gent investors and increase the markets’efficiency.

On September 28, 1998, ArthurLevitt, then chairman of the SEC, gave aspeech entitled “The Numbers Game”in which he discussed the widespreadpractice of earnings management.“Increasingly, I have become concernedthat the motivation to meet Wall Streetearnings expectations may be overridingcommon-sense business practices,” hesaid. “Too many corporate managers,auditors, and analysts are participants ina game of nods and winks. In the zeal tosatisfy consensus earnings estimates and

Hank Greenberg complies with SEC Order No. 4-460

SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 AUGUST 16, 2002 3

project a smooth earnings path, wishfulthinking may be winning the day overfaithful representation.

“As a result, I fear that we are wit-nessing an erosion in the quality of earn-ings, and therefore, the quality of finan-cial reporting. Managing may be givingway to manipulation; integrity may belosing out to illusion.

“Many in corporate America are justas frustrated and concerned about thistrend as we, at the SEC, are. They knowhow difficult it is to hold the line on goodpractices when their competitors operatein the gray area between legitimacy andoutright fraud. A gray area where theaccounting is being perverted; wheremanagers are cutting corners; and, whereearnings reports reflect the desires ofmanagement rather than the underlying

financial performance of the company.”Levitt noted that the pressure for

companies to meet analysts’ expecta-tions was corrupting peoples’ behavior.“Almost everyone in the financial com-munity shares responsibility for fosteringa climate in which earnings managementis on the rise and the quality of financialreporting is on the decline,” he said.“Corporate management isn’t operatingin a vacuum. In fact, the different pres-sures and expectations placed by, and on,various participants in the financial com-munity appear to be almost self-perpetu-ating.

“This is the pattern earnings manage-ment creates: companies try to meet orbeat Wall Street earnings projections inorder to grow market capitalization andincrease the value of stock options. Theirability to do this depends on achieving theearnings expectations of analysts. And ana-lysts seek constant guidance from compa-nies to frame those expectations. Auditors,who want to retain their clients, are underpressure not to stand in the way.”

Levitt described six practices used tomanipulate or “manage” earnings:accounting hocus-pocus, “big-bath”charges, creative acquisition accounting,miscellaneous cookie-jar reserves, mate-riality, and revenue recognition.

He also outlined a plan of action tostem the abuses, the final item of whichwas strengthening the audit-committeeprocess. “Qualified, committed, inde-pendent and tough-minded audit com-mittees represent the most reliableguardians of the public interest,” he said.

Levitt announced that as part of acomprehensive effort to address earningsmanagement, the New York StockExchange (headed by Richard Grasso),and the National Association ofSecurities Dealers (headed by FrankZarb), had agreed to sponsor a “blue-rib-bon” panel which would “develop aseries of far-ranging recommendationsintended to empower audit committeesand function as the ultimate guardian ofinvestor interests and corporate account-ability.”

On February 8, 1999 the 11-memberpanel released its report, which containednumerous reforms and recommendations.Although several members of the panelmade comments in an accompanyingpress release, we’ll quote one member,Frank Zarb, because, two years later, he

joined AIG’s board, and became a mem-ber of its audit committee the followingmonth. “Corporate governance is a keyissue facing the management of publiclytraded companies,” Zarb said. “The roleof audit committees is critical to thatprocess. These recommendations are athoughtful product of the expertise in thisarea.”

The panel’s numerous recommenda-tions included a written charter for theaudit committee, public disclosure ofaudit-committee activities, and an annu-al letter from the audit committee toshareholders.

According to the blue-ribbon panel,the audit committee was the most impor-tant participant in the financial reportingprocess. “A proper and well-functioningsystem exists,” the panel said, “when thethree main groups responsible for finan-cial reporting—the full board includingthe audit committee, financial manage-ment including the internal auditors, andthe outside auditors—form a ‘three-legged stool’ that supports responsiblefinancial disclosure and active and partici-patory oversight. However, in the view ofthe [panel], the audit committee must be ‘firstamong equals’ in this process, since theaudit committee is an extension of the fullboard and hence the ultimate monitor ofthe process.” [Emphasis added.]

The blue-ribbon panel recommend-ed that the audit committee, in its annu-al report, state that it “believes that thecompany’s financial statements are fairlypresented in conformity with GenerallyAccepted Accounting Principles (GAAP)in all material respects.” This recom-mendation seems so basic that it’s hard tobelieve it wasn’t already a requirement.

The SEC, perhaps feeling outsidepressure from the business community,did not adopt this recommendation. Itnoted a concern about exposing audit-committee members to additional liabili-ty, and mentioned that some commentersaverred that it might be difficult for com-panies to find people willing to serve onaudit committees if the audit-committeemembers were exposed to additional lia-bility. The SEC’s final rule stated that“because of concerns about liability, wedid not propose the disclosure require-ment recommended by the Blue RibbonCommittee, but instead proposed thatthe audit committee indicate whether,based on its discussions with manage-











4 AUGUST 16, 2002 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

ment and the auditors, its membersbecame aware of material misstatementsor omissions in the financial statements.”

Thus, the audit committee is notrequired to say that a company’s financialstatements conform to GAAP; it needonly has to say that it isn’t aware of mate-rial misstatements. Due to the watereddown regulations, a company can issuean evasive, equivocal audit-committeereport yet still comply with the SECguidelines. AIG has done this for thepast two years.

The recent financial and accountingscandals have produced a radical changein attitude. Regulators, legislators, insti-tutions, and even the president of theUnited States are now saying they’regoing to do something. The NYSErecently sent out a 28-page magazinecalled Your Market, one of the purposes

of which was to help restore investorconfidence. “Should you have faith inpublic companies?” asks the headline ofone article. “Without hesitation, theanswer is yes,” it replies. (The NYSEdoes not say why it failed to tellinvestors that they shouldn’t have had somuch faith a couple of years ago, whenthe market was fifty percent higher.)

President Bush, a hands-off, free-market sort of guy, is also concerned with“corporate responsibility,” and talks ofhunting down corporate evildoers andputting them behind bars.

The SEC now requires CEOs andCFOs of large companies to issuesworn written statements affirmingthat they haven’t cooked their compa-nies’ books.

Speaking about AIG’s recent decisionto expense stock options, Greenberg told

The Wall Street Journal that “the percep-tion out there today, erroneously, is thatnot expensing stock options is wrong. Theperception is more important than the sub-stance.” Since outsiders can’t audit AIG’sbooks, they will always be unable to getall the substance they would like. Theywill have to settle for perception.

Hank Greenberg’s sworn writtenstatement (see page 2) says that AIG’sSEC filings do not contain an untruestatement of a material fact, do notomit any material facts, and are notmisleading.

It also says that he has reviewed hisstatement with AIG’s audit committee. E

The blue-ribbon panel’s audit-committeereport is at http://www.nyse.com/content/pub-lications/NT00006286.html. The SEC’s finalrules are at http://www.sec.gov/rules/final/34-42266.htm#P122_33770.

http://www.nyse.com/content/publications/NT00006286.html

http://www.sec.gov/rules/final/34-42266.htm#P122_33770

The Audit-Committee Report

AIG IS ARGUABLY THE greatest insur-ance organization in the world. As such,it deserves more scrutiny than othercompanies.

In its reports for the years ending 2001and 2000, AIG’s audit committee dis-claimed virtually all responsibility forAIG’s accounting, internal controls, andfinancial statements. It also said that itcould not assure that AIG’s independentaccountants were actually “independent.”(Schiff’s raised this issue last year and cre-ated a bit of a stir.) The language didn’t sitwell with investors, especially thosewhose eyes had been opened by a newwave of reform in corporate governanceand accountability.

As we reported last August, AIG said itwould change its 2003 audit-committee sothat it was not merely a disclaimer of anyresponsibility for AIG’s financials.(Remarkably, we were the only ones to re-port this.)

On Friday, AIG released its 2003 proxystatement, which includes a new and im-proved audit-committee report. Thebroadly evasive language is gone, as is theequivocal twaddle that had rendered thereport meaningless. It has been replacedby something better, but not as good as itcould be. “The [audit] committee hasconsidered and discussed both the au-dited financial statements as well as theunaudited quarterly financial statementswith management and the independentaccountants,” says AIG’s audit-committeereport. The report says that the commit-tee discussed various mandated require-ments with the auditors, consideredwhether the auditors are “independent,”and recommended that the audited fi-nancial statements be included in AIG’sannual report.

For what it’s worth, we’ll contrastAIG’s audit-committee report with Coca-Cola’s. We picked Coca-Cola for one rea-son: its audit-committee is the only one inthe world that Warren Buffett is a memberof. While it’s debatable whether Coca-Cola uses better accounting than AIG, itdoes have a better audit-committee re-port. (The language used by Coca-Cola’saudit committee may even reduce thecommittee’s potential liability.) Here’s animportant excerpt from Coca-Cola’s re-port:

Management has reviewed the audited fi-nancial statements in the annual report with theaudit committee including a discussion of thequality, not just the acceptability, of the accountingprinciples, the reasonableness of significant ac-counting judgments and estimates, and the clar-ity of disclosures in the financial state-ments...Members of the audit committee have

expressed to both management and auditorstheir general preference for conservative policieswhen a range of accounting options is available.[Emphasis added.]

In its meetings with representatives of theindependent auditors, the committee asks... 1) Are there any significant accounting judg-ments or estimates made by management...thatwould have been made differently [by] the au-ditors?

We believe that AIG, its shareholders,and the public would be better served ifAIG’s audit committee adopted languagesimilar to Coca-Cola’s. We’ll have to waituntil next year to see if it does.

Two Audit-Committee Members

ALTHOUGH AIG’S CHAIRMAN and CEO,Hank Greenberg, has dominated AIG fordecades, he does not control the audit

SCHIFF’S




Inside the AIG Proxy Statement

American International Group’s D&O Premium: A Study in Cycles

AIG maintains a Directors & Officers policy for itself, its directors and officers, its subsidiaries, andtheir directors and officers. Although AIG has made several acquisitions since 1992 and is a muchlarger company today, its D&O premium is the same as it was in 1992. AIG has not disclosed theterms and conditions of its policies, or the limits and deductibles.

0

0.5

1

1.5

2

2.5

3

3.5

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

0

0.5

1.0

1.5

2.0

2.5

3.0

$3.5millions

AIG’s D&O Premium


committee (at least in theory). It is madeup of directors who are “independent” ac-cording to the current standards of theNew York Stock Exchange. (Greenberg,for the record, is a member of the NYSE’sCorporate Accountability & ListingStandards Committee.) We don’t knowwhy some members of the audit commit-tee didn’t insist on a better audit-commit-tee report initially.

For example, Frank Zarb, senior advi-sor to Hellman & Friedman, was previ-ously head of the NASD. In 1999, while atthe NASD, he co-chaired a “blue-ribbonpanel” that came up with “far-ranging rec-ommendations intended to empoweraudit committees.” The recommenda-tions were good, but not all were adopted.(Some of the recommendations were toogood.) Zarb is highly knowledgeable aboutfinancial matters and insurance. (He wasCEO of Alexander & Alexander from 1994to 1997.) His behavior on AIG’s audit com-mittee, however, seems to have been farmore passive than the behavior he recom-mended for audit-committee memberswhen he was running the NASD.

Carla Hills, CEO of Hills & Company,is also on the audit committee. (LikeZarb, she is on AIG’s executive commit-tee, as well.) She has had a distinguishedcareer. She was an assistant U.S. attorneygeneral, U.S. Secretary of Housing andUrban Development during GeraldFord’s administration, and the U.S. TradeRepresentative under George Bush. Wehave long been skeptical of her, however.She was a director of Henley Group, runby sleazy wheeler-dealer Mike Dingman.During her tenure on the board, Henleyengaged in dirty accounting, and its boardpermitted Dingman to concoct a variety ofdubious, self-enriching asset shuffles withaffiliated companies. (See “The HenleyManeuver—It Helps the Rich GetRicher,” Barron’s, December 19, 1988, byDavid Schiff.) Hills is also a member of theinfamous Time-Warner board that ap-proved the merger with AOL—perhapsthe worst deal of all time.

Some Notable Connections

MANY OF AIG’S DIRECTORS have closeconnections with AIG or Hank Greenberg(see the chart on page 3). Some have closeand unusual connections. MarshallCohen, for example, joined AIG’s boardin 1992, when he was president of The

Molson Companies. Molson was a share-holder in Coral Re, a suspicious, curiouslycapitalized offshore reinsurance companythat AIG ceded more than $1.5 billion inpremiums to in the 1980s and early 1990s,making it one of AIG’s largest reinsurers.Although Coral was formed to benefitAIG, AIG maintained—despite consider-able evidence to the contrary—that Coralwas not an affiliate. Coral’s initial investorswere offered a most unusual deal: theydidn’t have to put up a cent or risk moneyor collateral, yet they were guaranteed tomake a profit. AIG’s proxy statements didnot mention the connections betweenCoral and Molson and Cohen, eventhough Cohen was on AIG’s board.

AIG’s 2002 proxy statement does notdisclose—and perhaps is not required todisclose—that The Starr Foundation(which owns 56,957,340 shares of AIG, a2.18% interest), gave more than $25 mil-lion to the American Museum of NaturalHistory in 2001. (Greenberg is a trustee ofthe museum.) Seven past or present AIGdirectors were on The Starr Foundation’sboard at that time, and Hank Greenbergwas, and still is, chairman. According to an

IRS filing, he devoted 200 hours to TheStarr Foundation in 2001 and received nocompensation.

What makes The Starr Foundation’sgrant to the Museum of Natural Historynoteworthy, aside from its magnitude, isthat Ellen Futter, the museum’s president,was and still is on AIG’s board, and servedon its Stock Option and CompensationCommittee until September 18, 2002.Setting Hank Greenberg’s compensationis one of the committee’s major responsi-bilities. Although Greenberg is very wellpaid, his compensation isn’t particularlyunusual compared to some of the ridicu-lous figures doled out to lesser CEOs thesedays. In 2002, he received $1,000,000 insalary and a $5,000,000 bonus. He also re-ceived 375,000 ten-year options to buyAIG shares at $61.30 per share. Whilethere’s no single correct way to value op-tions, $6,000,000 seems like a reasonablevaluation for this options grant, bringingGreenberg’s total compensation for 2002to $12,000,000.

In February 2003, AIG’s compensationcommittee took unusual action. “In lightof the decline in the market price of AIG

In 1972, AIG’s shares traded at 518% of book value and 32 times earnings. From 1972 to 1974, AIG’sstock fell 66%, as these inflated multiples shrank, even though AIG’s earnings grew. In 2000, AIG’sprice-to-book-value and p/e ratios returned to their euphoric 1972 levels. AIG’s stock is now downabout 50% from its 2000 high. The price-to-book-value and p/e ratios are down significantly, too.

Primary source: Value Line

’720

100

200

300

400

500

600

700

-10

-5

0

5

10

15

20

25

30

35

40Price to Book Value

0

100%

200%

300%

400%

500%

600%

700%P/E ratio

40

35

30

25

20

15

10

5

’76 ’80 ’84 ’88 ’92 ’96 ’00 4/4/03

Price to Book Value

P/E ratio

American International Group: Stock Bubble Bursts, Again


common stock,” says AIG’s proxy state-ment, “the committee determined thatadditional incentives were needed to retain andmotivate employees.” [Emphasis added.]Instead of repricing the previouslygranted options, the committee grantedGreenberg, who owns 45,167,862 AIGshares, 375,000 additional options. Theproxy statement doesn’t disclose whetherthese options carried a lower strike pricethan the previous options.

More Connections

THE CHART ON THIS PAGE, which does notpurport to be complete, shows some of theconnections between Hank Greenbergand AIG’s directors. These connectionsare not listed in AIG’s proxy statementand have been gathered from varioussources. The chart does not show connec-tions between other directors, or betweendirectors, advisory board members, hon-orary directors, and AIG employees.

Four Directors Step Down

THREE AIG DIRECTORS ARE not standingfor reelection this year. They are EliBroad, Edward Matthews, and ThomasTizzio, all of whom work or worked atAIG. Former director, Robert Crandall,the retired chairman of AMR andAmerican Airlines, resigned from theboard on October 9, 2002. AIG’s board willhave sixteen directors.

AIG’s Important D&O Ad

ON FEBRUARY 3, AIG announced a $1.8-billion after-tax charge to increase lossreserves. About twenty-five percent of thischarge was for directors and officers liabil-ity. Hank Greenberg subsequently attrib-uted the charge to a “liability bubble” thatcould not have been foreseen.

In fact, AIG did foresee a difficult lia-bility environment. What it didn’t foreseewas the precise magnitude of the so-calledbubble. That D&O claims have surged isnot just the result of a legal system runamuck; it has much to do with the grosslyabusive behavior of corporate executiveswho have run amuck.

In 1968, the American Home AssuranceCompany, an AIG subsidiary, ran a strikingfull-page ad in The Wall Street Journal. (AIGhas always been an extremely effective andcreative advertiser.) The ad showed a small

No D

egrees of Separation: Hank G

reenberg’s Connections with Am

erican International Group’s D

irectors

AIG’s proxy statement provides very little inform

ation about the company’s directors. The chart below

—w

hich does not purport to be complete—

shows som

e of the non-AIG connections between

Hank Greenberg and AIG’s directors. These connections are not listed in AIG’s proxy statement.

MM..RR

.. GGrreeeennbbeerrgg

AAiiddiinnooffff

CChhiiaa

CCoohheenn

CCoonnaabbllee

FFeellddsstteeiinn

FFuutttteerr

HHiillllss

HHooeenneemm

eeyyeerrHH

oollbbrrooookkeeSSmm

iitthhSSuulllliivvaann

TTssee

WWiinnttrroobb

WWiissnneerr

ZZaarrbb

AAmm

eerriiccaann MMuusseeuumm

Trustee

Director

President

ooff NNaattuurraall HH

iissttoorryy

BBuussiinneessss CC

oouunncciill ffoorr H

onorary Trustee

Vice

IInntteerrnnaattiioonnaall UUnnddeerrssttaannddiinngg

Chairm

an

CCeenntteerr ffoorr SSttrraatteeggiicc

Form

er M

ember

aanndd IInntteerrnnaattiioonnaall SSttuuddiieessV

ice Chairm

an

CCoorraall RR

eeiinnssuurraanncceeSecret A

ffiliate?Shareholder

CCoouunncciill oonn FF

oorreeiiggnn RReellaattiioonnss

Form

er M

ember

Mem

berD

irectorM

ember

Vice

Director

Mem

berM

ember

Vice C

hairman

Chairm

an

FFeeddeerraall RR

eesseerrvvee BBaannkk

Form

er B

oard of F

ormer

ooff NNeeww

YYoorrkk

Chairm

anA

dvisorsC

hairman

IInnssttiittuuttee ffoorr IInntteerrnnaattiioonnaall D

irectorD

irectorEE

ccoonnoommiiccss

NNeeww

YYoorrkk SSttoocckk EE

xxcchhaannggeeD

irectorD

irectorF

ormer

Director

PPrroojjeecctt HH

ooppeeD

irectorA

dvisory B

oard

TThhee AA

ssiiaa SSoocciieettyyT

rusteeT

rusteeT

rusteeT

rustee

TThhee SSttaarrrr FF

oouunnddaattiioonnC

hairman

Director

Director

TTrriillaatteerraall CC

oommmm

iissssiioonnM

ember

Mem

berM

ember

Mem

ber

UUSS--AA

SSEEAA

NN BB

uussiinneessss CCoouunncciill

Vice C

hairman

Husband is

Vice C

hairman

UUSS--CC

hhiinnaa BBuussiinneessss CC

oouunncciillD

irectorD

irector


photograph of a conservative-looking busi-nessman beneath the headline, “I justmight sue every company director readingthis newspaper.” Here’s the text of the ad:

I am not a madman.This is not a joke.If you are a director of a major company, I’ve

got you where I want you.At my mercy. All I have to do is own a few

shares in your corporation and I can sue you andevery other director and officer in the company.

What can I sue you for?I can sue you for sending me a dividend pay-

ment that I think is unwarranted.I can sue you because I think your salary is

too high, or for conflict of interest or for missinga few director’s meetings. I can blame and sue you because of a misstatement in your com-pany’s financial report—or should I say our com-pany? I can’t begin to list all the reasons I can

sue you for. And here’s the saddest part. I’m notalone. There are 24 million other people outhere just like me. There are 24 million stock-holders in the United States and that’s 24 mil-lion potential stockholder suits. And even if youshould win a stockholder suit—you lose. Whenyou take into consideration lawyers’ costs,wasted time etc. At this point, you must be feel-ing helpless. You’re not.

There is a company that can help you.American Home Assurance Company. They didn’t invent stockholder suits, but they havecome up with some interesting solutions to them.They feature a type of insurance that every di-rector or officer in the United States should con-sider. Directors and Officers Liability Insurance.They have a booklet which tells all aboutDirectors and Officers Liability Insurance forthose that qualify. You can get it by writing toDept. A-14, American Home AssuranceCompany, 102 Maiden Lane, New York, N.Y.10005. Send for it and talk it over with your in-surance agent or broker. He and American HomeAssurance Company are good friends to havewhen you have 24 million potential enemies.

AIG’s ad, which pitted corporate big-wigs against their shareholders (“potentialenemies”), generated a tremendous re-sponse from major corporations and in-surance brokers. D&O coverage may havebeen unusual back then, but it’s ubiqui-tous now.

Perhaps some academic will analyzeD&O lawsuits and figure out whether di-rectors and officers were emboldened be-cause they were able to buy coverage. Didthe availability of insurance create moralhazards that led to greater abuses? DidAIG’s ad, unwittingly, awaken corporate di-rectors and officers as well as shareholdersand class-action lawyers?

In 1968, many public companies werenot especially accountable to their share-holders—their owners. Although share-holders have gained some power, they’restill disenfranchised. When directors missboard meetings, act as rubber stamps,overpay their CEOs, have conflicts of in-terest, and permit their companies’ finan-cial statements to be materially misstated,it’s not surprising that they’re sued.

Call it a liability bubble. Or call it abubble in directors’ and officers’ malfeasance. E

Editor and Writer . . . . . . . . David SchiffProduction Editor . . . . . . . . . Bill LauckForeign Correspondent . . Isaac SchwartzCopy Editor . . . . . . . . . . . . John CaumanPublisher . . . . . . . . . . . Alan ZimmermanSubscription Manager . . . . . . Pat LaBua










Some of AIG’s “independent” di-rectors have had unusual financialrelationships with AIG or its affil-iates, and may not be as indepen-

dent as they appear to be. AIG’s failure todisclose these relationships to its share-holders raises questions about the com-pany’s corporate governance and businesspractices.

Schiff’s has written extensively aboutAIG, for many reasons: AIG is the largestinsurance organization in the world; it’s agreat company; Hank Greenberg is a bril-liant guy; and, as Churchill said of Russia,AIG is a riddle wrapped in a mystery in-side an enigma. In our April 7 issue webroke a story, “Inside the AIG ProxyStatement,” that discussed, among otherthings, the undisclosed financial connec-tion that an AIG director, Ellen Futter,had with The Starr Foundation, a charita-ble foundation affiliated with AIG andcontrolled by the folks who control AIG.

Futter, president of The AmericanMuseum of Natural History, joined AIG’sboard in March 1999. Between 1999 and2001, the Starr Foundation gave $36.5million to The American Museum ofNatural History. AIG’s proxy statementshave not disclosed this , or that Greenberg isa trustee of The American Museum ofNatural History.

The Starr Foundation was created byAIG’s founder and is controlled by HankGreenberg and current and former AIGofficers. It owns fifty-six million AIGshares (2.1% of AIG) and is located inAIG’s headquarters.

The fact that The Starr Foundation pro-vided an enormous amount of funding tothe museum after Futter joined AIG’s boardis a matter that AIG’s shareholders deserveto be made aware of by AIG. Shareholders

have a right to know how indebted theircompany’s directors may be to Greenbergand the other AIG officers who control Starrand AIG. (Futter, who received about$145,000 in director’s fees from AIG lastyear, was on the company’s CompensationCommittee until September 18, 2002. Shethen switched to the Nominating andCorporate Governance Committee.)

If there were doubts whether AIG’sundisclosed relationship with Futter wasof interest to a broader audience thanSchiff’s readers, those doubts were erasedon Friday by an excellent front-page arti-cle in The Wall Street Journal entitled“Giving at the Office: On CorporateBoards, Officials From Nonprofits SparkConcern / When Directors’ Positions

Help Them Raise Funds, Danger ofConflict Follows: Aiding Ms. Futter’sMuseum.” The article, by David Bankand Joann Lublin, noted that Futter wason the boards of four companies (includ-ing AIG) that made substantial contribu-tions to her museum, and that these con-tributions “create[d] a potential conflictof interest: the possibility that moneyflowing from companies and their execu-tives will make nonprofit officials be-holden to the corporate management theyare supposed to monitor.” The long arti-cle, which contained a picture of Futteron the front page, left out a key fact we’dwritten about earlier—that AIG didn’tdisclose its unusual relationship withFutter to its shareholders. continued

SCHIFF’S


June 23, 2003Volume 15 • Number 10 I N S U R A N C E O B S E R V E R


AIG’s Secret Connection with Director$36.5 Million from Starr

AIG and the Starr Foundation: Secret Connection with AIG Director

When AIG’s founder, Cornelius Vander Starr,died in 1968, he left his estate to The StarrFoundation, which now has $3.2 billion inassets (all of it in AIG stock). The Foundationpays about $220 million of grants annually tomore than 1,000 individuals and almost asmany organizations.

Hank Greenberg, AIG’s chairman and CEO,is chairman of The Starr Foundation, whichoperates out of AIG’s headquarters. TheFoundation’s other directors are current orformer AIG directors or officers.

Ellen Futter, president of The AmericanMuseum of Natural History, joined AIG’s boardof directors in 1999. Between 1999 and 2001,The American Museum of Natural History wasthe second largest recipient of funds from TheStarr Foundation, receiving $36.5 million, or6.6% of all grants paid by Starr.

AIG did not disclose these grants to itsshareholders, nor did it disclose that HankGreenberg was a trustee of The AmericanMuseum of Natural History. AIG considersFutter to be an “independent” director.

1993 1994 1995 1996 1997 1998 1999 2000 20011993 1994 1995 1996 1997 1998 1999 2000 20010

50

100

150

200

$250millions

The Starr Foundation’s annual distributions

Source: The Starr Foundation’s IRS Returns, Form 990-PF

2 JUNE 23, 2003 SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000

Not surprisingly, spokesmen for Futter,Starr, and AIG said, respectively, thatFutter’s role as an AIG director is inde-pendent of her job as president of the mu-seum, that the donations Starr gave to themuseum had nothing to do with Futterbeing on AIG’s board, and that The StarrFoundation is independent of AIG.

We’ll throw out a rhetorical question:Might the fact that The Starr Foundationgave $36.5 million to Futter’s museummake her, as an AIG director, disinclinedto differ with Hank Greenberg? Accordingto an IRS filing, Greenberg is quite in-volved with The Starr Foundation; he de-voted 200 hours to it in 2001. Does this,plus $36.5 million—a huge sum for themuseum—have the potential to colorFutter’s decisions in any way?

Definitive answers to these questionsare probably unknowable. But AIG’sshareholders have a right to know aboutpotential conflicts of interest their direc-tors have. They have a right to know aboutinformation that might affect a director’sindependence. They have a right to knowwhat other financial or “charitable” con-nections AIG may have with its directors.

We’ve often been amazed by HankGreenberg’s tin ear on the subject of dis-closure. Although he wasn’t available totalk to us when we called today, we sus-pect that he would have said that AIG isa great company that has done nothingwrong; that Starr and AIG are generous,and that this whole matter is being blownout of proportion. He might also say thatwhen you’re as big as AIG it’s impossibleto disclose every little thing. And, oh yes,the New York Stock Exchange doesn’t re-quire disclosures about grants that AIG—or an AIG affiliate—gives to nonprofit or-ganizations that AIG’s directors are in-volved with.

Greenberg has done things his ownway for a long time, and has been extraor-dinarily successful. When AIG was re-porting rapidly growing earnings everyyear and the markets were going up, fewseemed bothered by little things such asdisclosure, corporate governance, and ac-counting transparency. Investors had faithin AIG because it was, after all, AIG, andit was run by Hank Greenberg, who al-ways hit his numbers.

The Journal’s article reported that theNYSE and Nasdaq are considering rulesunder which a director would not be con-sidered “independent” if his corporationreceived more than a certain percentageof its revenues from the company whoseboard he’s on. It isn’t clear if these pro-posals, or others that might be adopted,would affect AIG. Ultimately, having in-dependent directors will not create goodcorporate governance; good directors will.As for “independence,” almost everyoneon a corporate board got there because thepeople running the company wanted thatperson there. People who might shakethings up—even if that’s what’s needed—don’t get asked to be on corporate boards.

Whether Ellen Futter serves AIG’sshareholders well is not the issue. The issueis AIG’s lack of disclosure about the mate-rial financial connections between AIG,Starr, The American Museum of NaturalHistory, and Futter. This absence of dis-

closure is troubling and raises many ques-tions, including, “What is AIG hiding?”

AIG’s proxy statement also lacks dis-closure about directors’ compensation:“Certain directors who are not employeesof AIG also serve as directors of varioussubsidiaries of AIG and receive fees fortheir service in that capacity.” The proxydoesn’t say which directors serve on whichsubsidiaries, or how much they get paid.

AIG’s shareholders deserve betterdisclosure. Then they can make an in-formed decision about whether theywant to vote for a director whose organi-zation received $36.5 million from afoundation affiliated with AIG. E

Please go to the next page to read “No Degreesof Separation: Hank Greenberg’s Connectionswith American International Group’sDirectors.”











continued

SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 JUNE 23, 2003 3

No Degrees of Separation: Hank Greenberg’s Connections with American International Group’s Directors

AIG’s proxy statement provides little information about AIG’s directors.The chart below—which does not purport to be complete—showssome of the connections between Hank Greenberg and AIG’s directors.

These connections are not disclosed in AIG’s proxy statement. The chartdoes not show connections between other directors, or between direc-tors, advisory board members, honorary directors, and AIG employees.

American Business Center for Federal US- US-Museum Council Strategic Council on Reserve Institute The The ASEAN Chinaof Natural For Int’l and Int’l Coral Foriegn Bank of for Int’l Project Asia Starr Trilateral Business BusinessHistory Understanding Studies Reinsurance Relations New York Economics NYSE Hope Society Foundation3 Commission Council Council

Hank Trustee Honorary Former Vice Secret Former Vice Former Director Director Director Trustee Chairman Member Vice DirectorGreenberg1 Trustee Chairman Affiliate? Chairman Chairman Chairman

Aidinoff1,2 Member

Chia Trustee

Cohen Share-holder

Conable2 Member Director

Feldstein Director Board of MemberAdvisors

Futter President Member FormerChairman

Hills1,2 Member Vice Director Trustee Member Husband is DirectorChairman Vice Chairman

Hoenemeyer1,2

Holbrooke Director Trustee Member

Smith Director

Sullivan

Tse Advisory DirectorBoard

Wintrob

Wisner Vice Chairman Member

Zarb1,2 Member FormerDirector

Sources: Various, including www.elitewatch.netfirms.com 1. Executive Committee 2. Audit Committee 3. Some disclosure in AIG’s proxy

On Wednesday, February 11,American International Group,whose advertising slogan is“We Know Money,” issued a

press release containing its fourth-quarterand full-year earnings and related finan-cial information. It was a swell press re-lease—except for the fact that it was mis-leading, deceptive, and inconsistent withthe way AIG had highlighted its earningsin previous press releases.

If AIG’s intention was to dupe thepress and the public, it appears to havesucceeded. News organizations across theglobe reported the figure AIG high-lighted—68% growth in earnings—whereas, based on its previous re-leases, 14.7% growth would havebeen a more appropriate figure.(The media typically report compa-nies’ earnings by reprinting or summa-rizing press releases.)

This was not the first time that AIGpresented its earnings in a deceptive way.Schiff’s recently conducted a study of thecompany’s quarterly-earnings releases andannual reports during the 1998-to-2003period and determined that from thefourth quarter of 1999 through the fourthquarter of 2003, AIG used four definitionsof earnings, switching back and forthamong those definitions ten times. Theseswitches improved the appearance ofAIG’s growth rate and made declines inearnings seem like increases. [See thechart on pages 5 and 6.]

Since the fourth quarter of 1999, AIGhas issued 16 earnings releases and fourannual reports. In 19 of these releases andreports, AIG highlighted the better num-bers that were created by switches in theways it defined its earnings. Perhaps it’schance, but these switches never made

AIG’s earnings or growth rate appearlower (even though they were lower inmany cases). The figures that AIG high-lighted gave a misleading impression inten earnings releases and annual reports.

It is appropriate, when a company pre-sents its financial results, for it to do so ina consistent manner: results from one re-porting period should be comparable withthose of the previous year (as they say, ap-ples should be compared to apples). If acompany constantly changes its methodof reporting, then it may be difficult—orimpossible—to track its progress, or lackthereof.

AIG’s shares trade on the New YorkStock Exchange. The NYSE’s “Listed

Company Manual” states that,“Unfavorable news should be re-ported as promptly and candidly asfavorable news.” It continues:

“Reluctance or unwillingness to re-lease a negative story or an attempt to

disguise unfavorable news endangers man-agement’s reputation for integrity. Changesin accounting methods to mask such oc-currences can have a similar impact.”

We don’t know if AIG deliberately dis-guised unfavorable news (i.e. maskinglower earnings and a lower growth rate);but AIG’s earnings releases and annual re-ports have, in fact, disguised unfavorablenews. We can’t help but note a remark-able coincidence: that the numerousswitches AIG made in its earnings pre-sentations improved the earnings orgrowth rate that AIG highlighted 95% ofthe time. What are the odds that AIG—bysheer chance—switched its standards tentimes in four years, and that theseswitches—by sheer chance—improvedthe appearance of AIG’s earnings orgrowth rate 19 out of 20 times? (The oddsthat a coin flip will turn up heads 19 out of20 times are about 50,000-to-1.)

At one time AIG’s quarterly earnings’releases and annual reports highlightedthe company’s “net income” according toGenerally Accepted AccountingPrinciples (GAAP). Net income is a com-pany’s actual “bottom line,” but it isn’t al-ways the best way of looking at an insur-ance company’s results. Analysts oftenmake adjustments to the bottom line inorder to get a clearer picture of actual per-formance. It is common to exclude the ef-fect of realized capital gains and losses onearnings. The reason for this is that thetiming of gains and losses is generally dis-cretionary, and realized gains and lossesusually bear no relationship to a com-pany’s investment results in a given quar-ter or year. A company might have real-ized losses in a year in which its invest-ment portfolio appreciated, and it mighthave realized gains in a year in which itsportfolio declined. Since unrealized gainsand losses aren’t run through the incomestatement (they’re a balance-sheet entry),it can be fair and useful to present “pro-forma” earnings excluding realized gainsand losses—even though this doesn’t con-form to GAAP.

From 1992 through 1999, AIG had re-alized capital gains in 30 quarters and

SCHIFF’S


February 26, 2004Volume 16 • Numbers 3 & 4 I N S U R A N C E O B S E R V E R


AIG: The Art of Manipulation?Deceptive Earnings Releases

THE ANNUAL

Thursday, April 15,2004

in New York City (434-977-5877)

http://www.snlcenter.com/schiff/spring2004/

INSURANCE CONFERENCE

REGISTER NOW

SCHIFF’S

http://www.snlcenter.com/schiff/spring2004/

SCHIFF’S INSURANCE OBSERVER ~ (212) 724-2000 FEBRUARY 26, 2004 2

small realized capital losses in two quar-ters. (The losses were 1.03% and 0.47% ofpretax income, in the fourth quarters of1998 and 1999, respectively.)

In 1998 and 1999, AIG’s earnings re-leases highlighted the company’s “net in-come” but also provided a pro-forma fig-ure—“income, as adjusted”—which ex-cluded realized capital gains or losses.

In the first quarter of 2000, AIGchanged the way it highlighted its growthrate in its press releases; it began exclud-ing realized capital losses. (In every quar-ter since then AIG has had realized capi-tal losses. These losses have often beensizable—greater than 10% of pretax in-come.) “AIG’s First Quarter 2000 IncomeExcluding Realized Capital Gains(Losses) Rose 15.5%,” stated the headlineof the company’s press release. If AIG hadused its previous practice of highlighting“net income,” the headline would havedeclared that income increased by 12.3%.

There’s a big difference between a15.5% growth rate and a 12.3% rate. Over20 years, $100 compounded at a 15.5%rate will grow to $1,785, versus $1,018 forthe same sum compounded at a 12.3%rate. All things being equal, companieswith higher growth rates (or the appear-ance of such) invariably trade at muchhigher P/E ratios than those with some-what lower growth rates. For decades,AIG has been viewed as a “growth” com-pany, and its stock has usually traded at amuch higher P/E ratio and price-to-bookratio than have the stocks of most otherinsurance and financial-services compa-nies. (Often, the higher P/E ratio was jus-tified.)

AIG’s practice of highlighting the pro-forma growth in earnings by excluding re-alized gains and losses isn’t troubling perse. In AIG’s 2000 annual report, chairmanand CEO Hank Greenberg noted that“we [AIG] and the investment commu-nity look at our results” this way. What istroubling, however, is that AIG did notsubsequently highlight its earnings thisway when doing so resulted in a lower rateof growth.

AIG continued to highlight the pro-forma “income, as adjusted” growth ratethrough the second quarter of 2001. TheWorld Trade Center loss occurred the fol-lowing quarter. AIG then highlighted itsresults using a pro-forma method it called“core earnings,” which excluded under-writing losses related to the World Trade

Center attack. In the next four quartersAIG made at least three more switches inthe method it used to come up with fig-ures that it highlighted. First it used in-come excluding capital losses; then it usednet income. Finally, when it took a $1.8billion loss-reserve charge in the fourthquarter of 2002, it used a new variation of

pro-forma “core earnings” that excludedthe loss-reserve charge.

In his letter to shareholders in the 2002annual report, Greenberg dragged a redherring across the issue of the loss-reservecharge, calling it “an extraordinary reserveadjustment.” He wrote that “no actuarialcalculation could have predicted the ex-

REGISTER NOW

‘Grand Delusion’PRESENTING THE ANNUAL


Thursday, April 15, 20048:30 am - 5:30 pm

New York City

Registration fee: $695 per person.Call (434) 977-5877 for more

information, or reserve a place now.www.snlcenter.com/schiff/spring2004/

David Schiff editor of Schiff’s Insurance Observer, will start off witha look at the seamy side of the insurance business. Throughout the day he will,as always, interrogate the speakers and force them to answer brazen questions.

Over the decades, William R. Berkley, CEO of W.R. BerkleyCorporation, has demonstrated that he knows how to build value in hard mar-kets and in soft markets. When Bill spoke at our 1999 conference, he was acute-ly aware of the risks—and opportunities—that lay ahead. Since then, his businesshas been on a roll and his company’s stock (which we had recommended) hasmore than quadrupled. Bill will give us his atypical perspective in his typicallyeloquent manner.

Betsy McCaughey is a thinker, author, and expert on health policy. Her1994 critique of the Clinton health plan, “No Exit,” caused a ruckus and helpedkill the plan. Betsy, who was an unusually independent Lieutenant Governor ofNew York, has published two books on U.S. constitutional history and is writ-ing a book on health care. She will tell all, including how “medical courts willsolve the malpractice crisis.”

Milberg Weiss Bershad Hynes & Lerach LLP didn’t invent the class-action lawsuit, but, as the largest contingency-fee-based law firm representingplaintiffs, it has certainly perfected it. Senior partner Melvyn Weiss is a lead-ing practitioner in the fields of securities, insurance, environmental, antitrust,and consumer litigation. Mel’s comments may leave some members of theinsurance industry feeling afraid—very afraid.

Lunch: Decent food; fine conversation.

Many insurance companies don’t have the data to price risk properly. DanielFinnegan, president of Quality Planning Corporation, is a statisticianwho knows how to compile, analyze, and use data in ways that can create a signifi-cant underwriting edge. “There’s enormous room for the improvement of predic-tion,” he notes matter-of-factly. Daniel will take us into the world of rating error,black boxes, credit scoring, database analysis, geo-positioning systems, privacyissues, and probabilities. And that’s just the beginning. continued on next page

9:00 a.m.

9:30 a.m.

10:40 a.m.

11:20 a.m.

Noon

1:00 p.m.

www.snlcenter.com/schiff/spring2004/


plosion of litigation in the United States,which has resulted in an enormous in-crease in the frequency and severity of li-ability claims and judgments.”

The $1.8 billion charge, however, was-n’t for events that occurred 25 years ear-lier; it was for losses during the 1997 to2001 accident years. The only thing thatmade the charge “extraordinary” was thatAIG doesn’t usually make such large mis-takes. The reserve charge was not attrib-utable to an isolated legal judgment or todiscontinued operations; it was for excesscasualty (including excess workers’comp), directors and officers liability, and“other casualty” (including healthcare li-ability).

AIG’s actuaries may not have pickedup on Greenberg’s so-called “explosion oflitigation,” but Greenberg did. He’s criti-cized “tort law” and the “legal system” fordecades. He has written that “courts in ourcountry continue to broaden the standards

of legal responsibility and increase the sizeof awards,” and raised the “persistent mat-ter of excessive liability awards by courts.”(The quotations just cited are fromGreenberg’s 1977 letter to shareholders,AIG’s 1985 annual meeting, Greenberg’s1986 letter to shareholders, and his 1989letter to shareholders, respectively.)According to Tillinghast-Towers Perrin’s“U.S. Tort Costs: 2003 Update,” inflation-adjusted tort costs per citizen grew from$716 in 1990 to $809 in 2002.

AIG’s $1.8 billion reserve charge wasthe result of underwriting mistakes overfive years. Was it really appropriate to treatthese mistakes as “extraordinary” itemsthat deserved to be factored out of high-lighted earnings in 2002? If it was appro-priate, then wouldn’t it have also been ap-propriate for AIG’s fourth-quarter 2003earnings release to compare 2003’s earn-ings with the pro-forma earnings AIGhighlighted in 2002? (If AIG had done

that, it would have reported a 14.7% in-crease in earnings rather than a 68% in-crease.)

Page one of AIG’s 2002 annual reportcontains a bar chart of the company’s netincome each year from 1998 to 2002. Thefigure for 2002 adds back the $1.8 billionloss-reserve charge, making the com-pany’s growth have a smooth upwards tra-jectory. Although AIG’s chart didn’t in-clude the charge in 2002, the chargeshould go somewhere. If it was appropri-ate to add back $1.8 billion to 2002 earn-ings, then $1.8 billion should have beensubtracted from the 1997-to-2001 years asan acknowledgment that earnings hadbeen overstated during that period.

Inconsistent ReportingDuring 2003, AIG continued to switch

the way it highlighted its earnings andgrowth. In the first quarter it highlighted“income, as adjusted” (excluding capitallosses). For the next three quarters itswitched to “net income”—despite thefact that Greenberg had written that the“income, as adjusted” method was theway AIG looked at its results.

We’ve noticed one constant in the wayAIG has highlighted its earnings orgrowth: 19 out of 20 times the companyused the figures that made its results lookbetter.

On February 18, we discussed our ob-servations with AIG and asked why thecompany changed its methodology sooften, noting that the changes improvedAIG’s figures 85% of the time. (We sub-sequently determined that they improvedthem 95% of the time.)

Two days later AIG provided a politeresponse: “AIG gives a thorough ac-counting in its quarterly earnings newsreleases, and it reports its results in com-pliance with all SEC and accounting reg-ulations.” Because this was such a briefresponse, we’ll add the following: AIGis the world’s leading international in-surance and financial-services organiza-tion, with operations in approximately130 countries and jurisdictions. Its earn-ings releases include GAAP financial in-formation.

On December 4, 2001, the SEC is-sued a release containing caution-ary advice regarding the use of

pro-forma financial information in earn-ings releases: continued

It may surprise some to learn that Schiff’s has a hero. His name is JosephBelth and he is, of course, the editor of The Insurance Forum. Joe, whosearticles, speeches, and testimony have shaken up the life-insurance industry, isthe author of numerous books and journal articles and professor emeritus ofinsurance at the Kelley School of Business at Indiana University. He’ll tell youwhat’s bothering him these days.

Jay Brown is CEO of triple-A-rated MBIA Inc., which specializes in cred-it-enhancement insurance. He was previously CEO of Talegen Holdings, andbefore that, CEO of Fireman’s Fund. Jay, who’s an actuary, has a contrariannature and a keen appreciation of risk—desirable attributes for one running acompany with $6 billion of equity and $500 billion of financial guarantees out-standing. He will offer his thoughts about insurance, credit, financial guaran-tees, risk versus reward, and more.

David Schiff will discuss where he sees value and solvency (or the lack there-of), and have his say on the great insurance issues of the day.

Attendees will socialize with their fellow insurance mavens and observers, dis-cussing the day’s events and making deals over cocktails while taking in theview from the top of the New York Athletic Club.

There will be an additional reception and dinner for those who wantmore of a good thing. The venue is the Coffee House, a convivial and somewhatworn-at-the-edges private club devoted to “agreeable, civilized conversation.”Attendance is limited to 36 people.

1:45 p.m.

2:45 p.m.

3:45 p.m.

4:45 p.m.

6:00 p.m.

‘Grand Delusion’PRESENTING THE ANNUAL


Thursday, April 15, 20048:30 am - 5:30 pm

New York City


...It is appropriate to sound a warning topublic companies...who present...their earningsand results of operations on the basis of method-ologies other than Generally AcceptedAccounting Principles (“GAAP”). This presen-tation in an earnings release is often referred toas “pro forma” financial information. In thiscontext, that term has no defined meaning andno uniform characteristics...

...Public companies may quite appropriatelywish to focus investors’ attention on criticalcomponents of quarterly or annual financial re-sults in order to provide a meaningful compari-son to results for the same period of prior yearsor to emphasize the results of core operations...

Because “pro forma” information is...derivedby selective editing of financial information com-piled in accordance with GAAP, companies shouldbe particularly mindful of their obligation not tomislead investors when using this information...

Companies must pay attention to the mate-riality of the information that is omitted from a

“pro forma” presentation. Statements about acompany’s financial results that are literally truenonetheless may be misleading if they omit ma-terial information.

In 2003, the SEC issued the final rulesfor Regulation G, which deals with theuse of non-GAAP financial measures.AIG’s 2003 fourth-quarter earnings re-lease contains a “comment” onRegulation G. The company acknowl-edges that its press release contains non-GAAP financial measures, and that a “rec-onciliation of such measures to the mostcomparable GAAP figures” is included inaccordance with Regulation G. AIG saysits press release “presents its operationsin the way it believes will be most mean-ingful and useful, as well as most trans-parent, to the investing public and otherswho use AIG’s financial information inevaluating the performance of AIG.”

If, in the past, AIG also presented itsoperations in the manner it believed wasmost meaningful, useful, and transparent,that raises questions, including the fol-lowing: Why did AIG find it meaningfuland transparent to make so many switchesin the way it highlighted its earnings andgrowth rate? Why did AIG highlight “netincome” in the third quarter of 1999, “in-come, as adjusted” excluding capitallosses in the third quarter of 2000, “coreearnings” excluding certain underwritinglosses in the third quarter of 2001, and“net income” in the third quarters of 2002and 2003? Is it a coincidence that theseswitches made AIG’s results or growthrate appear better than they otherwisewould have been in 19 of 20 instances?

AIG’s “comment on Regulation G”notes that “the determination to realizecapital gains or losses is independent ofthe insurance underwriting process...Realized capital gains or losses for any par-ticular period are not indicative of quar-terly business performance.” It goes on tosay that “providing only a GAAP presen-tation of net income and operating incomemakes it much more difficult for users ofAIG’s financial information to evaluateAIG’s success or failure in its basic busi-ness, that of insurance underwriting, andmay, in AIG’s opinion, lead to incorrect ormisleading assumptions and conclusions.The equity analysts who follow AIG ex-clude the realized capital gains and lossesin their analyses for the same reason...”

In other words, AIG seems to be say-ing that if it “only” provides GAAP net in-

come figures that might be misleading be-cause “income, as adjusted” to exclude real-ized gains and losses is the more importantmeasure of performance.

If it might be misleading to provideonly GAAP figures, then isn’t it mislead-ing (and downright sneaky) to highlight thegrowth rates for GAAP “net income”when, in fact, the growth rates for “in-come, as adjusted” (excluding capitalgains and losses) are considerably lower?

With that in mind, let’s examine AIG’sJuly 24, 2003 earnings release. The head-line reads, “AIG Reports Second Quarter2003 Net Income Rose 26.4% to $2.28Billion.” To report 26.4% growth seemsspectacular. But why would AIG highlight“net income” instead of the figure it hassaid is more meaningful: “income, as ad-justed” (excluding gains and losses)? Did thefact that “income, as adjusted” grew13.9%—about half as much as “net in-come”—have anything to do with AIG’sdecision to highlight the higher, mislead-ing figure? Ask Hank Greenberg. Andwhile you’re at it, ask him why AIG’s 2003third-quarter headline declared that “NetIncome Rose 26.9%,” when, in fact, “in-come, as adjusted” rose 15.4%?

It would have been nice if AIG ex-plained how it came to pass that in 19out of 20 instances it highlighted the

earnings or growth rate that was most fa-vorable. The company could have told usit was chance. It could have explainedwhy highlighting “net income” some ofthe time and highlighting ever-changingpro-forma figures other times really wasthe best way to present its performance ina fair, honest manner. It could have saidthat it was “trying to put a positive spin onits results,” then tried to provide somereason why that was not deceptive.

AIG has long been respected and val-ued for the steadiness with which its earn-ings have grown. Beginning in 2000, AIG’searnings releases and annual reports havegiven the impression of greater growth andconsistency than that which actually oc-curred. AIG is a gigantic company andHank Greenberg is a brilliant man. Butpeople should be wary of companies thatdon’t present their results fairly. AIG’smanner of highlighting the most favorablefigures and growth rates raises a sad ques-tion: Should AIG be trusted?

Please refer to the charts on the follow-ing pages.


Editorial OfficeSchiff’s Insurance Observer300 Central Park West, Suite 4HNew York, NY 10024Phone: (212) 724-2000Fax: (212) 712-1999E-mail: [email protected] Website: InsuranceObserver.com

Publishing HeadquartersSchiff’s Insurance ObserverSNL c/o Insurance Communications Co.One SNL PlazaP.O. Box 2056Charlottesville, VA 22902Phone: (434) 977-5877Fax: (434) 984-8020E-mail: [email protected]









What Figure is Highlighted? Effect Result Comments

1999 - 1Q Net Income Neutral1999 - 2Q “ “ Neutral1999 - 3Q “ “ Neutral1999 - 4Q “ “ NeutralAnnual Report “ “ Neutral

2000 - 1Q Income, as adjusted (excludes Improve Inconsistent Reports 15.5% growth in “adjusted” income instead of 12.3% growth Capital Realized Losses) in net income

2000 - 2Q “ “ “ Improve Inconsistent Reports 13.1% growth in “adjusted” income instead of 10.2% growth in net income



Annual Report 1) Net Income; 2) Net Income, Improve Inconsistent Hank Greenberg mentions adjusted income in letter to share- as adjusted (excludes Realized holders. (Says it’s “the way we and the investment communityCapital Losses) look at our results.”) Reports 14.8% growth in adjusted” income

instead of 11.5% growth in net income

2001 - 1Q Income, as adjusted (excludes Improve Reports 15.2% growth in “adjusted” income instead of 13.8% growthRealized Capital Losses) in net income

2001 - 2Q “ “ “ Improve Reports 15.8% growth in “adjusted” income instead of 15.6% growth in net income

2001 - 3Q Core Income (excludes 9/11 Improve Misleading 9/11 WTC loss. American General restructuring charges. WTC loss, Realized Capital AIG now reports “core income” instead of “adjusted” income. Losses, and Acquisition & (“Core income” is reported ahead of “net income.”) Headline Restructuring Charges) of AIG’s release: “Core income rose 14.1% to $1.92 million.”

This is an unfair comparison and inconsistent with prior reporting. An accurate headline would have been, “Adjustedincome declines 18.5%.”

2001 - 4Q 1) Net Income; Improve Misleading Headline cites “net income.” Text shows “core income” increased 2) Core Income (excludes 9/11 13% in 2001. Core income increased 5% when 9/11 WTC loss WTC loss, Realized Capital is counted.Losses, and Acquisition & Restructuring Charges)

Annual Report 1) Core Income Improve Misleading Does not show “adjusted” income, which is lower than “core (excludes 9/11 WTC loss, income.” On page 1, a five-year graph of “Net Income” uses Realized Capital Losses, “core income” for 2001. Since high-severity, low-frequency and Acquisition & Restructuring losses like 9/11 WTC do occur, treating them as extra-Charges); 2) Net Income ordinary or non-recurring—which AIG has done—smoothes core

earnings. In a table showing an 11-year summary of consolidatedoperations, the bottom line—net income—omits the charge for theWTC losses.

In his letter to shareholders, Hank Greenberg writes about“Reaffirming our Corporate Values,” and says, “Every year we workhard to improve our disclosure...We will always...adher[e] to thehighest ethical standards, and provid[e] a thorough and accuratepicture of our operations and financial performance.”

Annual report shows so-called core income increasing by 13%. It only increased 5% when all underwriting losses are included.

(table continues on next page)

AIG: The Art of Manipulation? Deceptive Earnings Releases and Annual Reports

From the fourth quarter of 1999 through the fourth quar-ter of 2003, AIG used four definitions of earnings, switch-ing back and forth among those definitions ten times.These switches improved the appearance of AIG's growthrate and made declines in earnings seem like increases.

During this period, AIG issued 16 earnings releases andfour annual reports. In 19 of the 20 releases and reports,AIG highlighted the better numbers that were created byswitches in the ways it defined its earnings. The highlight-ed figures were misleading ten times.

The table below tracks AIG's quarterly releases andannual reports. The first column describes which figuresAIG highlighted in its release or annual report. Notewhere AIG switched the way it highlighted its earnings(i.e. net income, income as adjusted, core earnings,etc.). The second column notes the effect the switcheshad on the earnings that AIG's highlighted. The third col-umn notes the result of the switches (i.e. inconsistent,misleading, etc.). The last column contains our com-ments.


What Figure is Highlighted? Effect Result Comments

2002 - 1Q Income, as adjusted (excludes Improve AIG’s earnings are released three days after AIG, whose Realized Capital Losses) stock is under pressure, issues press release claiming to “have

observed considerable short selling in [AIG’s] stock.” AIG requests that the NYSE and SEC “investigate this activity.”

2002 - 2Q Net Income Improve Misleading AIG’s earnings are released one day after AIG’s stock hits its year’s low of $46.71 (down from an all-time high of $103.75 on December 8, 2000). Investors are skeptical of complex “black-box” financial companies like AIG, and are worried whether AIG can continue to achieve the steady growth it is known for.

By reporting “net income” instead of “adjusted” income, AIG showsa 37% increase in earnings instead of a 9.8% increase.

2002 - 3Q “ “ Improve Misleading In the prior year’s third quarter, AIG used “core income,” which excluded the WTC losses and various restructuring charges. In the second quarter of 2002, however, AIG begins highlighting “net income.” Because the third quarter of 2001had been bad, AIG can expect to show sensational growthfor the year by highlighting “net income.”

AIG reports that “net income” increased 60.8% during the first nine months of 2002. Core income—which AIG used in the previous year’s third quarter—increased 11.3%.

2002 - 4Q 1) Net income; Improve Misleading AIG announces a $1.8 billion loss-reserve charge. In the previous2) Income,as adjusted seven quarters AIG had gone from reporting “adjusted” income(excludes Realized Capital to “core income” to “net income” to “adjusted” income then Losses and a $1.8 billion back to “net income.” Loss-Reserve Charge) By highlighting “net income” for 2002, AIG once again

portrayed its earnings in a misleading way. In 2001, Greenberg told shareholders that income adjusted to exclude realized capital gains and losses was the best way to view AIG’s results.

In 2002, AIG reports that “net income” increased 2.9% and “income as adjusted” (excluding the reserve charge) increased 11.9%. If AIG had reported “adjusted” income (excluding capital gains and losses) it would have shown a 4.2% decline for the year.

Annual Report 1) Income, as adjusted Improve Misleading On page 1, a five-year bar chart of “Net Income” uses “core income”(excludes Realized Capital for 2001 and 2002. (Core income was a much higher figure.)Losses and a $1.8 billion Commenting on the $1.8 billion charge to increase loss Loss-Reserve Charge); reserves, Greenberg, who has been complaining about the legal2) Net Income system for more than 30 years, blames society: “No

actuarial calculation could have predicted the explosion of litigation in the United States.”

The truth: AIG underestimated its losses during 1997-2001.

2003 - 1Q 1) Income, as adjusted (excludes Neutral First time in three years that AIG doesn’t highlight earnings in theRealized Capital Losses) most favorable way. “Income, as adjusted” (excluding realized

capital gains and losses) is an apples-to-apples method of lookingat the change in earnings from year to year.

2003 - 2Q Net Income Improve Misleading Misleading reporting resumes. Headline says “net income rose 26.4%.” In fact, “adjusted” income rose 13.9%.

2003 - 3Q 1) Net Income; Improve Misleading Highlights earnings both ways, but first states that “net income 2) Income, as adjusted rose 26.9%.” Misleading because “adjusted” income, which (excludes Realized Greenberg has said is the way AIG and the investment communityCapital Losses) look at the numbers, rose 15.4%.

2003 - 4Q “ “ “ Improve Misleading AIG’s headline says “net income” increased 68%. AIG doesn’t provide an “income, as adjusted” figure (excluding the loss-reservecharge) like it did the previous year. Had it done so, one would have seen that, on an apples-to-apples basis, earnings increased 14.7%—far less than the 68% figure AIG trumpeted. Since AIG had highlighted the “adjusted” income figure in the previous year’s release and in its annual report, it should have included compara-ble figures here.

AIG: The Art of Manipulation? (continued)

Documents

SCHIFFS - Schiff's Insurance Observer€ states that “situations which could result in conflicts of interest or the appearance of a conflict of interest should be avoided whenever