S P R I N G / S U M M E R 2 0 0 1

STERNbusinessTaking Stock of Our Financial Times

Back to the Future with High-Yield Debt

The Emergence of Online Brokers

Raging Bulls or RagingBull.com?

James Cramer Speaks

How the Very Rich Differ from You and Me

tors, faculty, and students –are improving on a qualita-tive basis as well.

For example, we areaggressively leveraging ourposition as a center of aca-demic excellence located inthe heart of the world’sfinancial capital. Last fall,we broke new ground in thefield of business educationby becoming a part of thenew TRIUM Executive MBAProgram. This exciting part-nership allies Stern with twoof the world’s other premier

institutions: the LondonSchool of Economics andHEC Paris, GraduateBusiness School. And thisspring, Stern sent the entirejunior undergraduate classon a business trip with desti-nations in Italy, Hong Kongand Mexico.

The contents of this mag-azine highlight the fact thatan understanding of financegoes far beyond numbers.You may find yourself drawnin by the charts and figures.But you will be mostimpressed with the sophisti-cation of the analysis, thebreadth of topics covered,and the wide-ranging inter-ests of our contributors.

The same holds for theentire Stern School. Sure, thenumbers are exciting, and ajustifiable source of pride.But the figures are merelythe tip of the iceberg. I inviteyou to discover – and redis-cover – our rich array ofofferings.

George DalyDean

STERNbusinessA publication of the Stern School of Business, New York University

President, New York University ■ L. Jay OlivaDean, Stern School of Business ■ George DalyChairman, Board of Overseers ■ William R. BerkleyChairman Emeritus, Board of Overseers ■ Henry KaufmanAssistant Dean, Marketing

and External Relations ■ Joanne HvalaEditor, STERNbusiness ■ Daniel GrossProject Manager ■ Caren L. PielaDesign ■ Esposite Graphics, New Orleans

Letters to the Editor may be sent to:NYU Stern School of BusinessOffice of Public Affairs44 West Fourth Street, Suite 10-83New York, NY 10012

http://www.stern.nyu.edu

dean

In the world of finance,investors, executives, andanalysts frequently use num-bers as a convenient form ofshorthand. Interest rates,dividend yields, stock prices,weekly sales figures – thisdata can give professionalsan accurate picture of howeconomies and companiesare performing.

In the world of financialeducation, numbers matter,too, which is why we areparticularly gratified by therecent publication of two

rankings. This winter, theFinancial Times adjudgedStern to be the eighth-ratedfull-time MBA program inthe United States, and thetenth-rated program in theworld. And last December, abroad study of facultyresearch productivity waspublished in the Academy ofManagement Journal. Sternranked fourth out of morethan 400 universities for fac-ulty research productivity intop-tier journals; five of oureight individual disciplinesscored in the top ten.

As seen by recent activityin the equity markets, how-ever, numbers can be deceiv-ing. Many $100 stocks nowtrade for $2. And just asmany sure-fire businessmodels have been abandonedas unworkable.

By contrast, our improvedquantitative rankings arebased not on overly opti-mistic projections or flawedassumptions, but on hardfacts. Outsiders may notethat Stern is improving on aquantitative basis. But thosewho spend time on our cam-pus also know that all of ourconstituencies – administra-

a l e t t e r f r o m t h e

SternChiefExecutiveSeries

Interviews withJames Cramer, page 2Hendrick Verfaillie, page 4

8 These Financial Times by Daniel Gross

contents

Cover and related articles illustrations © ARTIST(S)/SIS

Illustrations by James Yang (© artist(s)/SIS) andDave Black

Information graphics byRobert O’Hair

42 InternetMessage Boards:Raging Bulls orRaging Bull?by Robert Tumarkin

48 Endpaper:Golden Oldiesby Daniel Gross

26Business Styles

of the Rich and

Famousby Roy C. Smith

32In Sync: Why Stocks in Some Markets Move Togetherby Randall Morck and

Bernard Yeung

38The Remonopolization of Telecommunicationsby Nicholas Economides

TAKING STOCK OF OUR FINANCIAL TIMES

S P R I N G / S U M M E R 2 0 0 1

10 Déja Vu AllOver Again withHigh-Yield Debtby Edward I. Altman

18The Rise ofOnline Brokersby Chris Stefanadis

24Recountingthe Florida Recountby Aaron Tenenbein

ML: Jim, we'll start with an easyone. What's the stock marketgoing to do in the next year? JC: The market is at a curiousjuncture. Up until oil spiked andthe Euro declined I thought thatwe were very much on targetfor a 1994-style soft landing.Now it looks more like a 1990scenario, when oil went tounsustainably high levels andwe also had rates going higher.That led to a very badChristmas season and it led toa very tough period for banks,as loan loss reserves went updramatically. So we're strad-dling between '94 and '90 rightnow. I think that one-third of thestocks have bottomed already.

ML: Only one-third?JC: Only one-third. I'm talkingabout Alcoa, InternationalPaper, Dow Chemical. Thosestocks have bottomed. I amtrying to accumulate them.AT&T is now my largest posi-tion. Another third, whichincludes some technology thatI regard as classic tech likeIBM, are groping for a bottomright now. And then there'sanother third that I think franklyare sales, if not shorts. Theytend to be the stocks that areselling at a 100 to 200 timesearnings without a sustainablebusiness plan that may not beable to give us any visibilityafter this quarter to next year.

ML: What impact will the elec-tion results have on the stockmarket, if any?JC: First, I have to disclose I'ma card carrying member of theDemocratic party and a largegiver. Gore is closely alignedwith the Rubin faction, whichstands for a hard dollar, lowerinterest rates, and a lessaggressive tax posture. Thereis a sizeable camp of peoplewho are going to begin to ruethe notion that there could be atax cut. So a tax cut, which wenormally would think would bebullish for the markets, isagainst the Greenspan com-pact with Congress, which is toreduce the deficits and keepinterest-rates low. So while I

don't think Bush would be sofabulous for the stock market, Idon't think he would be so bad.

ML: Jim, what's your ownrecord as a stock picker thisyear?JC: We're up 32 percent. It's avery good year for us. We'vecompounded 24 percent net ofall fees.

ML: When you pick a stockwhat do you really look for?What are some of the charac-teristics that either attract youto an investment or repel you?JC: Let's use the example ofAT&T. AT&T traded in the 60s,now it's at 28. It's 40 off itshigh. I like this. I'm buying at

James Cramer is the co-founder, contributing editor, and

director of the TheStreet.com, an online financial publication

that provides financial news commentary and information. In

addition to his many roles at TheStreet.com, Mr. Cramer was

a partner for 13 years in Cramer Berkowitz and Company, a

Manhattan-based hedge fund. A graduate of Harvard

College, where he was president of the Harvard Crimson,

and Harvard Law School, Cramer worked as a journalist and

stockbroker at Goldman Sachs before forming his own

hedge fund. He is a co-founder and former columnist of

Smart Money magazine and a columnist for New York. He

is also the Markets Commentator for CNBC, and appears

frequently on Squawk Box and Business Center.

sternChiefExecutiveseries

James Cramer co-founder, contributing editor and director,TheStreet.com (Interview conducted on October 4, 2000)

2 Sternbusiness

the level where there's been alot of disappointment built intothe stock. I check all the FirstCall notes and I see one luke-warm buy and probably 12holds and even some sells.That means I've got peoplewho are willing to convert. Mostimportantly AT&T is a companythat when cable stocks haverallied 25 percent has said it iswilling to sacrifice its safer longdistance business because itreally is a cable company. Inthe parlance of what we say inthe investment meetings, I thinkAT&T's two down, 30 up. This isthe kind of stock that we thinkworks in this environment.

ML: Does this mean that thatwhen I see a stock where 12analysts are following it and 11have a strong buy and one hasa buy I should be suspicious ofthat stock?JC: I would tell you that I thinkthe stock in that scenario is adangerous stock. When every-body loves a stock there isn'tanybody who can go from a buyto a strong buy, because they'reall clustered at strong buy. We'revery oriented toward finding situ-ations where we think a substan-tial amount of the risk has beenremoved. We're most interestedin Alcoa at 27 with the possibilityof taking out a lot of capacitythan we would be for a situationlike VerticalNet, which everybodyloves.

ML: How's TheStreet.comcoming?JC: You know, TheStreet.comis in many ways a great disap-

pointment for me because I'ma hedge fund manager. And inhedge funds we value stocksfor what they sell at. Whensomeone asks me what'sTheStreet.com worth, I don'tsay, “Well, the brand is worth a$100 million.” I get no satisfac-tion out of what we've accom-plished because of a stockprice [about $4 at the time].That's the downside. Theupside is that we created thisunique product. We give youreal-time information with anopinion that I think is worth asubstantial amount to thosewho read it. I accept the mar-ket's judgment. I think that themarket says that there will notbe a jump up on advertising onthe web; that the web haspeaked. I've accomplished alot in my life, but in the end I'ma $4 stock.

ML: Why did the stock godown so sharply?JC: Last year our companywas run very poorly. That issomething nobody is ever sup-posed to say, but I said it vocallyat board of directors meetings.We had a resulting change inmanagement which I think isnot at all reflected in the stockprice. But we started too high.We went public at 20, andopened at 70. There was justincredible investor enthusiasmthat I thought was way over-done. We proceeded to godown for three reasons. Thestock market which was loved,loved, loved last year at thistime is now disliked by manypeople, because they're losing

money. Reason two is thecohort. Regardless of what wedid, we are still a dot-com andthat is the suicidal suffix.

Finally, we had to switchgears, from growing like wild-fire to becoming profitable. Iwould say that we're in themidst of a very difficult down-sizing that is meant to be ableto produce profitability nextyear at this time.

ML: Turn it around. What isgoing to cause it to comeback? JC: We need to be able to cutcosts more aggressively than

we have. This is a companythat had seven million in rev-enues in the second quarter.We're doing $28-30 million inrevenues a year. Now, I'm fromthe old school which says ifyou can't make money doing$30 million in revenues letsomeone else try it. And I thinkthe mandate that our board hasgiven to management is to finda way to make it profitable at$30 million.

ML: Give me a model thatwould work for a dot-com infor-mation company.JC: Let's look at it from thepoint of the view of the cus-tomer first. The customer hasexpressed a tremendousunwillingness, as we know, to

pay for anything on the web.That said, TheStreet.com willget $3.25 million from subscrip-tion revenues and should beable to generate twice that.And you should have twice thesubscription revenues foradvertising. I think you canbuild a really terrific niche busi-ness that way. I think you canmake five million on $30 millionin revenues. But I can't comeup with a scale business apartfrom Yahoo! and AmericaOnline that works. And I don'tregard Amazon as workingbecause they're losing billionsand they owe a lot of money.

The problem is the advertis-ers just do not want to advertiseon the web. Last week I spent aweek with our top five advertis-ers. In each case they said,“We loved advertising whereyou were in Smart Money, andwe love advertising against yourcolumn in New York, but wedon't feel comfortable doing theweb.”

The people who advertise onthe web don't want to brand onthe web, they want what'sknown as direct response.They want to be able to say“Look, we had this many peo-ple click on our ad.” So we'reheld to a higher hurdle. Theadvertisers expect too muchfrom the web and they want ourrates to come down.

Marshall Loeb, the former managing editor of Moneyand Fortune, conducts a regular series of conversationswith today’s leading chief executives on the Stern campus.

The adver t isers expect too much from

the web and they want our rates to

come down.

cont’d. page 6

Sternbusiness 3

ML: Tell me about the newMonsanto. HV: Basically, we are quite aunique company. We have fullyfocused on biotechnology andgenomics. We believe that hightechnology and genomics havethe potential to revolutionize theway crops are producedthrough the use of chemicalprocesses which are nowbecoming biological processes.We believe that is a lot moresustainable and more produc-tive. We have made very biginvestments in establishing thetechnological capabilities, andalso in acquiring biotech and

genomics companies. We mapthe genes of plants. We aredeveloping information that weput back into seeds that makesthe plant resistant to diseaseand that makes the plant resist-ant to insects. So, instead ofutilizing pesticides, we simplyput information in this combina-tion of information technologyand biology.

ML: Our regulatory environ-ment is different from the regu-latory environment in Europeand in Canada. Does that haveany consequence for the devel-opment of biotechnological

products that help humankind?HV: Yes. In the United States,we have very high demands onthe data that you have to devel-op before you get to registra-tion. The Food and DrugAdministration, theEnvironmental ProtectionAgency, and the U.S.Department of Agriculture arevery good agencies with verystrong reputations. If they saythat something is safe, peoplewill believe it because theyhave a tradition of being right.In Europe, on the other hand,they have regulations wherepeople do not really know what

it is that they have to deliver.As a result, it takes muchlonger before it gets throughthe regulatory process. So, fora company that is making verysignificant investments inresearch and discovery, theU.S. is a much better place toplace your bets than Europe.That is why most of the med-ical, biotech and genomicresearch is being done here inthe United States.

ML: With populations explod-ing, are we going to haveenough food to feed all thepeople in the world?

Hendrick Verfailliepresident and chief operating officerMonsanto

A graduate of the University of Louvain in Belgium,

Hendrick Verfaillie joined Monsanto in 1976 in Brussels. A

chemical engineer by training, he managed a number of

product lines in the European market before transferring to

the corporate headquarters in St. Louis. There he took on

posts of increasing responsibility. As president and chief

operating officer of Monsanto, he was responsible for

developing and executing an integrated strategy across

the life science spectrum, with special emphasis on agricultural food and nutrition. In 2000,

Monsanto agreed to merge with Pharmacia & Upjohn. Several months later, the merged

company spun off its agricultural business in an initial public offering. The newly independ-

ent company – known again as Monsanto – specializes in developing and manufacturing

herbicides, seeds, and biotechnology traits. With 2000 sales of more than $5 billion,

Monsanto is based in St. Louis and has 14,000 employees worldwide. Mr. Verfaillie is now

president and chief executive officer of Monsanto.

sternChiefExecutiveseries

4 Sternbusiness

HV: We think we need newtechnology to be able to feedthe world. There are basicallythree elements that drive thefood demand. Number one isobviously the number of peo-ple. That number is projectedto go up from six billion todayto eight billion by the year2020. Number two is a chang-ing diet. As economies devel-op, people increase the calo-ries they consume, and they gofrom a purely vegetarian tomore of a meat diet, which,again, will impact the numberof crops that are produced.Finally, as the economy aroundthe world improves, peopletend to spend more. So we canmake all kinds of assumptionson those three factors. Our esti-mate is that by the year 2020,we will have to produce 75 per-cent more food than we areproducing today.

ML: How? By increasing theamount of calories produced?HV: Yes. If you look at whathas happened in the 50s and60s, we have had what theycall the Green Revolution. TheGreen Revolution was thearrival of new varieties ofhybrids, fertilizers, and pesti-cides. People were predictingin the 50s that we never wouldbe able to produce enoughfood for the growing popula-tion. Thanks to the GreenRevolution the food productionaccelerated as fast as the pop-ulation did. But now we neednew technology or otherwise,we won't be able to deliver thiscontinuous growth.

Q: What is Golden Rice?HV: It is basically biotechnolo-gy. In the world today, 200 to300 million people suffer fromvitamin A deficiency. Vitamin Adeficiency can lead to nightblindness, and then, eventually,to irreversible blindness.Through biotechnology, wehave been able to develop ricethat will produce very high lev-els of vitamin A. So, by simplyeating the normal foods you eatevery day, you get sufficientvitamin A to avoid this deficien-cy and this illness. That isGolden Rice.

ML: Where is it in use now?HV: It is being developed forAsia. But this is only the first ina long series of possibilitiesthat are being created throughthis new technology. We areworking on, for example, highiron rice and high iron corn.There are two billion people inthe world who suffer from irondeficiency. We can developcrops that basically havehealthier aspects to them to thepoint where you can do pre-ventative health care.

ML: Why then, has there beensuch an emotional, sometimesviolent, outcry against geneti-cally engineered crops?HV: Most of the oppositionstarted in Europe. They havehad a number of food safetyproblems. You may have heardabout mad cow disease. Butthey also have problems withthe safety of their blood bankand with the growth hormonesthat they give children. They

have had one problem afteranother because they do nothave a good regulatory systemas we do have here in theUnited States.

Here in the United States,we love science because weknow that science has tradition-ally brought progress and pros-perity. In Europe, they aremuch more concerned aboutnew technology. So usuallywhat you see is that in Europenew technologies, whether inmedical technology or in food,or even in marketing and busi-ness, are five to ten yearsbehind the rest to adopt tech-nologies. The third element isthat in Europe food is some-thing which is part of that cul-ture. So, when you touch theirfood, they are very concerned.

ML: What is your response tothe protesters?HV: We made a mistake.Biotechnology is really verybeneficial to the farm and thefarmer loves it. But what we for-got is that the consumer playsa big part. In the past, whenwe were developing chemicals,as long as we would discover avery good product that wouldbring a lot of benefits, and wedid a good marketing job, wewould be successful. Now, theconsumer is concerned aboutthe changes that we are mak-ing. We did not pay enough

attention to the consumer,especially in Europe. So wehave started to inform and edu-cate the consumer about thebenefits that biotechnology canbring, such as lower use ofpesticides, examples likeGolden Rice, and sustainability– using less resources to pro-duce the same amount ofcrops. We are now much moreaggressively explaining whybiotechnology is good for theworld.

ML: But still, there are quite afew educated, knowledgeableprotesters in Britain, France,and Germany. What furtherarguments would you givethem to bring them along?HV: Well, now we are seeinggreat support from Africancountries and Asian countries,because they see the benefits.Just imagine, for a minute, thatyou can develop a crop that istotally tolerant to disease, thatis totally tolerant to insects, thatis tolerant to drought and thatyou can plant anywhere in theworld. Now that seed in Africa,or in Asia, can be planted andcan really increase productivitysignificantly, and thereby helpto feed populations. If you go astep beyond that and havehealthier nutrition, then itbecomes a real Godsend forthose countries. So now theyare standing up and saying

For a company that is making very

s igni f icant investments in research and

discovery, the U.S. is a much better

place to place your bets than Europe.

cont’d. page 6

Sternbusiness 5

James Cramer, cont’d.

ML: Tell me more about therelationship between advertis-ing and subscriptions on thenet?JC: The advertising revenuecan't please Wall Street. Wewent from an all-subscriptionmodel to a free site and a pre-mium site model. We did itbecause we believed that if wewent free for part of our site wewould draw a huge number ofnew users and then be able togo to a whole new set of adver-tisers and say, “Look at thisgrowth that we have.”

When we go to an advertiserthey say, “Well, three millionpeople might go to your frontpage every day, but only10,000 saw our ad.” And I said,“No, only 10,000 people clickthrough it. They see it If youhave the right ad they'll get thebrand message.”

ML: You seem to say thatthere's going to be a pretty bigshakedown in the businessnews and stock informationsites, a handful of survivors,and then this industry will pre-sumably grow and prosper.What do you think?JC: I think that's very right.When you go to see a mediadirector, they’re advertising insix places, because they can'tfigure out which is the best.The moment the top two sitescombine, it’s game over.There's no critical mass in ourbusiness right now. Once thereis critical mass, one will be thewinner. And that one will makea lot of money. But everybody

has had too much ego, andeverybody thinks their stock isgoing to come back again.

Q & A with StudentsQ: When a company likeConseco or Bank One brings ina superstar CEO how much doyou think that really helps thecompany?JC: It's very hard to put a valu-ation on an individual. GaryWendt, who is a man I got toknow when he was at GeneralElectric, moved to Conseco. Hewas a good hire in a really badbalance sheet situation that Istill think is unsavable because Ijust don't think that it's as easyto turn insurance companiesaround as people think. JamieDimon, another man I respecteda great deal when he was atCitigroup, came in to Bank Onein a much easier situation.That’s a really good franchise,and the brand was notdestroyed. So I think there's alot to Bank One, although wedon’t own it right now.

Q: For the last three years, it'sbeen a growth-driven market.Do you see the cycle turningmore toward value stocks? Andif so, will value investing beprofitable?JC: I have always been loatheto talk about value investingversus growth investingbecause those are both disci-plines. But the buying that I seein the momentum stock has nodiscipline whatsoever.Broadcom yesterday tradeddown from being 230 timesearnings to 212 times earnings,which was a terrific opportunityfor four or five investment

banks to reiterate buy ratings.Well, there's no discipline toowning a stock at 212 timesnext year's earnings.

It has been tough to be avalue investor and maintain dis-cipline, because as they strug-gled the most undisciplinedinvestors were up 400 percent.In late 1998 I wrote myinvestors and said listen, we'rehaving a mediocre year. I'mgoing to get rid of all my valuestuff and I am going to play thegame. My assets under man-agement were about $300 mil-lion. And when I sent that letterout, $103 million came out withit. But my old style would haveled to a down year in 1999.Instead, we were up 63 per-cent. So I know what we didwas right.

I just don't see the valuecamp as being able to pull offthe kinds of numbers that thepublic really wants. The valuecamp is the camp that is aca-demic in nature and themomentum growth camp ismore frenetic in nature and isalmost anti-academic. It saysthere really is nothing elsecooking other than the stock.And if the stock goes up I wantto be a part of that.

Q: If you had to put the lessonfrom this under one headlinewhat would it be? JC: In the long run America is anation that is willing to pay asuper premium for growth versusno growth. And that's nowingrained in the American peo-ple so that if you cannot producethat growth it's questionablewhether you want to be public.

Q: I wanted to ask about oneof your favorites, Cisco. CiscoCEO John Chambers came outmany times over the past fewmonths and said everything'sgreat. But the stock is justlanguishing in the 50s. Whatdo you think?JC: It's true, Cisco is myfavorite. I spent three days therein June, which I regarded asvacation days. Cisco is in aperiod of multiple contraction,where everybody's a little tiredof paying a 100 times earningsand there are a lot of peoplewho are saying that Cisco is anover-owned stock. Cisco's ingood shape, but I will not kidyou. Cisco is a ridiculouslyexpensive stock. And for me tobe in it requires every ounce ofconviction. There are somepositions that I own because Ilove management. I will own thatstock until CEO John Chamberseither tells me that business isbad or steps down. ■

Hendrick Verfaillie, cont’d.“Europe, you are well fed, youhave plenty of food. You decidewhat you want to eat, but youare not going to dictate whatwe can eat, or what technologywe can use.”

ML: In its issue of April 10th,the New Yorker reported that"Monsanto has a greater com-mitment to producing geneti-cally modified crops than anyother organization in the world."Is your strategy to continue thatcommitment?HV: Yes. We no longer do anychemical research because webelieve that the future is in biol-

6 Sternbusiness

ogy. We believe that we can doanything through biology thatcan be done through chemistry.Biology is a technology that ismore sustainable. It is environ-mentally more friendly. It hasmuch greater promise becauseit is based on an understand-ing of the human genome, theanimal genome, the plantgenome. We have made a betand we have burned the ships,if you will, because we believethat there is great opportunity.

ML: What biotech productsand other agriculture productsdo you intend to invest themost in? HV: There are three waves oftechnology. The first one is theone on the market today, whichis basically delivering agricul-tural productivity gains. We canmake plants resistant to dis-ease and to insects, makethem higher yielding and resist-ant to herbicides. The secondwave is what we call outputtraits or quality traits. That iswhere you change the qualityof what the output is. That iswhere the Golden Rice andhigh iron comes in. We can putsteroids in the crops that helppeople control cholesterol orthat help people control bloodpressure, simply by eating thefood that you eat every day.The first products will be on themarket within a couple ofyears. The third wave andprobably, eventually, the mostinteresting, is that you canlearn to produce any kind ofchemicals through biotechnolo-gy. We are already doing itwith pharmaceuticals. We now

produce therapeutic proteins inproteins with monoclonal anti-bodies that before you had toproduce through animal mod-els. It can be done anywhere inthe world and it offers very sig-nificant opportunities for moresustainable products. You knowthat whenever you produce achemical utilizing a petroleumbased product that you slowlybut surely are using up theresources in the world. On theother hand, if you do it througha biological process you haverenewable resources as longas you have sun and water.

ML: Why should a Stern grad-uate seek her or his future inthe agriculture business andthe biotechnology business justwhen it's coming under somuch popular attack?HV: Because where there ischallenge, there is opportunity.We are in the process of revo-lutionizing the way agricultureis done. The agriculture chemi-cal business was a $30 billiondollar business. That's high,but really nothing truly exciting.With biotechnology, we now getnot only the input side, but wealso get the output side. Forexample, we have just set up ajoint venture with Cargill to pro-duce new feed. Just like youcan produce healthier food forhumans, you can producehealthier feed for animals. Andsuddenly, we have an entirelynew opportunity driven bybiotechnology and genomics inthe hundreds of billions.

Q & A with StudentsQ: What are your feelings areabout labeling? HV: We are supportive of theFDA position, which is that youonly put something on the labelwhen the nutritional qualitychanges. For example, if therewas a peanut gene broughtinto corn, that would have to belabeled, because peanutscause allergies. If there is ahealth risk, you would have toput it on the label. But if thereis no change in the nutritionalquality of the food that's beingproduced, then there is no rea-son to put that on the label.

Q: I think there is a growingtrend in the U.S. to consumeorganic foods. How is thisaffecting your strategy in theU.S. compared to other areas ofthe world?HV: Less than two percent ofthe food that is being consumedis organic in the U.S. I really likethat there is organic food, thatthere is food produced usingchemicals, and that there isfood produced through biotech-nology. And so the consumerhas a choice of what he wantsto have. And if you believe thatorganic food is better for youand you are willing to pay twoor three times what regular foodwill cost you, then please goahead and do so.

Q: Your company has a strongfocus on the sustainability of ourplanet's limited resources. Is thecompany investing in researchas to possible risk in the equilib-rium of our ecosystem?HV: Yes, we certainly are. If we

did not increase productivity,and the population increases bytwo billion by the year 2020,and on top of that, industryexpands and housing expands,suddenly we are going to lose alot of the forests and wild lands.And so what we are doing is try-ing to increase productivity, sothat we can produce more onfewer acres. Also, throughgenomics and biotechnology,you use a lot less herbicides,pesticides, insecticides, and soon. And so we see that biodiver-sity actually increases.

Q: You also talk a lot aboutyour strategies and the waythat they can help the ThirdWorld and developing coun-tries. How do you propose toensure that these technologicaldevelopments find their waythere and make an impactwhere they're most needed,and where the profits may notbe the highest?HV: Let me just give you a fewexamples. In Africa, we aredeveloping virus-resistant cas-sava and virus-resistant sweetpotato, which are the keyindigenous staples in West andEast Africa, respectively. Wehave brought over scientists tolearn the technology and wehave given them for free thegene to put into their crops. So,we aren't going to make anymoney. But if it helps demon-strate the benefits of biotech-nology very clearly, then wecan introduce virus-resistantcontent. So when we can intro-duce cash crops that areexported, we can make moneyfrom that. ■

Sternbusiness 7

In the United States, finance hasbecome something of a nationalobsession. The Wall Street Journalsells more copies than The New YorkTimes. And when the markets areopen, CNBC routinely draws moreviewers than CNN. As investors haveflooded into the markets – some83 million Americans arebelieved to own stock today – thelevel of popular attention devot-ed to corporate finance and per-sonal finance has intensified.With every passing day, more andmore of our citizens are attunedto the latest change in the FederalFunds rate, the fate of the 30-year bond, and the debate overwhether to invest Social Securityfunds in equities.

To a large degree then, this isan era in which finance reallymatters – and not just to bankersand traders in lower Manhattan.It has become democratized, rele-vant, and pervasive. And many ofthe trends and forces that have roiledthe frequently topsy-turvy world offinance are somewhat bewildering.

One of the most distinguishingfeatures of the personal finance rev-olution has been the growing abilityof individuals to trade stocks andbuy mutual funds online. ChrisStefanadis traces the emergence ofthe online discount brokerage indus-try and describes how they helpedimprove conditions for consumers

(p. 18). The next step in the revolu-tion: “The creation of an onlinefinancial supermarket that offers abroad array of services.” Of course,that’s a tactic many off-line firmshave tried over the years; few havesucceeded.

When they trade stocks, onlineinvestors frequently rely on the newsand analysis they find on websiteslike TheStreet.com, which was co-founded by former hedge fund man-ager James Cramer. As part of ourchief executive lecture series, the vol-uble – and occasionally volatile –Cramer engaged students and facul-ty in a wide-ranging conversationabout the business of picking stocks,and about the business of running anonline magazine devoted to the busi-ness of picking stocks (p. 2).

Aside from logging on toTheStreet.com and its rivals,Internet investors have flocked tobulletin boards and chat rooms,where they can interact and swapideas with other investors. By focus-ing on one site – RagingBull.com –

and systematically analyzingthe action, Robert Tumarkindraws some interesting con-clusions about the relation-ship between the volumeand content of online posts

about certain stocks and theirtrading volume and performance(p. 42).

These days, following stockscan be a 24-hour-a-day preoccu-pation, especially as barriers totrading stocks on foreignexchanges continue to fall. But inthe last decade, investors largeand small have been stung by aseries of crises in emerging mar-

kets such as Mexico, Russia, andthose in Asia. Stockholders were fre-quently frustrated to find that thestocks they had carefully chosenwere pulled down by the activity inthe broader market. In fact, asProfessors Randall Morck andBernard Yeung argue, stocks inemerging markets tend to exhibitgreater price synchronicity thanthose in developed markets – theytend to move in the same direction(p. 32). By deploying some rathersophisticated techniques, the authors

By Daniel Gross

8 Sternbusiness

have managed to offer somecompelling reasons as to whythat may be.

In the 1980s, high-yieldbonds – also known as junkbonds – became one of themost exciting and controver-sial tools in the world of cor-porate finance. Entrepreneursrelied on high-yield bonds tofinance rapid growth, andleveraged buy-out artists andcorporate raiders used them totake over companies. But in theearly 1990s, as junk-bonddefault rates soared, returnsdeclined and the bonds fell outof favor. In the late 1990s,high-yield bonds similarly rodethe crest of exuberant capital mar-kets and then crashed to earth asexcesses entered the system.Professor Edward Altman, a veteranhigh-yield market analyst, providessome valuable perspective on therollercoaster ride of the high-yieldmarket. “Although storm cloudshang over today’s high-yield market,the current situation differs in anumber of important respects from adecade ago,” he notes (p. 10). Andjust as the years after the crisis in theearly 1990s were characterized byoutsized returns, he suggests that“returns will be substantial after thepeak of defaults, and perhaps evenbefore the peak – whenever itoccurs.”

In 2000, several telecommunica-tions companies defaulted on theirhigh-yield debt. Of course, telecom-munications is an industry that haslong depended on engineering foradvances. But in recent years, finan-cial engineering has become anequally important discipline forthese companies, which must navi-gate the shifting shoals of regula-tion and competition. ProfessorNicholas Economides provides anexcellent primer on why once-stolidcompanies such as AT&T have beenmerging, spinning off units, andbreaking up at a dizzying pace (p.38). The unanticipated outcome?“The remonopolization of telecom-munications.”

In his years as a partner atGoldman Sachs, Professor RoySmith ran across his share offinancial moguls. And hishighly readable new book,

The Wealth Creators: The Rise ofToday’s New Rich and Super-Rich,contains some valuable insights asto what separates run-of-the-millbusiness owners from big shots. Inhis article, which is adapted fromThe Wealth Creators, Smith reach-es the (perhaps) surprising conclu-sion that it’s not just a matter ofmoney and financial expertise.

To be sure, numbers don’t lie.But numbers certainly leave room

for debate and interpretation. Twoanalysts can look at the same securi-ty, after all, and one can declare itundervalued while the other declaresit overvalued. And that makesfinance such a rich, complex, andintriguing area of study.

This constant tension, the waythat financial questions invite inter-pretation and foster innovativeanalysis, provides the animatingspirit of this issue. Whether you’re araging bull or a cautious bear, thearticles that follow will certainlychallenge your preconceptions andinfluence the way you think aboutfinance.

DANIEL GROSS is editor of STERNbusiness.

Sternbusiness 9

ILLUSTRATIONS BY DAVE BLACK

STERNbusiness Spring/Summer 2001

10 Sternbusiness

DEJA VU ALL OVER AGAIN?In the summer of 1990, the market for high-yield debt – also known as junk bonds

– stood at a turning point. In 1989, the amount of defaulting issues had reached a new

high of $8 billion, or 4.3% of the $190-billion market. That default rate was almost

twice the average of 2.2% posted between 1978 and 1988. Borrowers had defaulted

on another $4.8 billion in debt in the first six months

of 1990. Drexel Burnham Lambert, by far the leading

underwriter of these bonds, had recently filed for Chapter 11 bankruptcy. And its guru and

leading light, Michael Milken, had been indicted. The new issue market had all but dried up.

T H E A C T I O N I N H I G H - Y I E L D B O N D M A R K E T S I N T W O T U M U L T U O U S

Y E A R S A D E C A D E A P A R T – 1 9 9 0 A N D 2 0 0 0 – S H O W E D S O M E

S T R I K I N G S I M I L A R I T I E S , A N D S O M E I M P O R T A N T D I F F E R E N C E S .

By Edward I. Altman

Sternbusiness 11

© A

RTI

ST(

S)/

SIS

At that time, many marketobservers were pronouncing the junkbond market “finished.” The con-ventional wisdom on the Streets –Wall Street and Main Street – wasthat high-yield bonds had run theircourse and neither new investors norissuers would “play in the junkyard”again. To compound matters, theU.S. government in August 1989enacted a law that forbade Savings& Loans from investing in low-gradebonds and mandated they sell theirholdings by the end of 1993.

The popular stigma attaching tojunk bonds was also reinforced by awidely circulated academic study. Inthe Journal of Finance, Paul Asquithand David Mullins, both then atHarvard, contended that pastresearch, by systematically underes-timating the true default risk ofhigh-yield bonds, had painted anunduly rosy picture of the market.

At the time, I was convinced thatthe market’s problems were tempo-rary. In a 1990 article, I wrote: “Thesystem needs to be “cleansed” of theexcesses of the past few years. Thenext wave of junk bond issues – andthere will almost certainly be one(although whether the issuers will bepublicly or privately placed is not atall clear) – will reflect more conser-vative capital structures and financ-ing strategies. Prices of leveragedtransactions will come down and theproportion of equity underlying suchlevels will rise.)”

In the 10 years since, many ofthose predictions have become reali-ty. The default rates noted byAsquith and Mullins did not presagea permanent increase. Instead,default rates averaged less than 2%per year from 1992-1998. Perhapsmost important, during the 1990s –and, indeed, over the entire 25-yearlife of the modern high-yield market– investors have essentially gotten

what they bargained for. They haveearned a rate of return that, atroughly 200 basis points over thereturn on 10-year Treasuries, is com-mensurate with junk bonds’ interme-diate level of risk – higher than thatof investment grade bonds, but lowerthan that of common stocks.

Some of the same phenomenathat we observed in 1989-1990 haveagain surfaced in 1999-2000.Default rates soared to 4.15% in

1999, with a record $23.5 billion inbonds falling into default. And lastyear, borrowers defaulted on another$30.2 billion (Figure 1), a secondconsecutive record year of defaults.With defaults rising, investors’required yield spread over Treasuriesrose to 944 basis points as ofDecember 31, 2000. By one metric –the return on high-yield bonds rela-tive to Treasury note returns – theyear 2000 was the worst performing

2000 $597,200 $30,248 5.065%1999 $567,400 $23,532 4.147%1998 $465,500 $7,464 1.603%1997 $335,400 $4,200 1.252%1996 $271,000 $3,336 1.231%1995 $240,000 $4,551 1.896%1994 $235,000 $3,418 1.454%1993 $206,907 $2,287 1.105%1992 $163,000 $5,545 3.402%1991 $183,600 $18,862 10.273%1990 $181,000 $18,354 10.140%1989 $189,258 $8,110 4.285%1988 $148,187 $3,944 2.662%1987 $129,557 $7,486 5.778%1986 $90,243 $3,156 3.497%1985 $58,088 $992 1.708%1984 $40,939 $344 0.840%1983 $27,492 $301 1.095%1982 $18,109 $577 3.186%1981 $17,115 $27 0.158%1980 $14,935 $224 1.500%1979 $10,356 $20 0.193%1978 $8,946 $119 1.330%1977 $8,157 $381 4.671%1976 $7,735 $30 0.388%1975 $7,471 $204 2.731%1974 $10,894 $123 1.129%1973 $7,824 $49 0.626%1972 $6,928 $193 2.786%1971 $6,602 $82 1.242%

YEARPAR VALUE

OUTSTANDING (a)PAR VALUEDEFAULTS

DEFAULTRATES

ARITHMETIC AVERAGE DEFAULT RATE 1971 TO 2000 2.713% 2.484%1978 TO 2000 2.948% 2.683%1985 TO 2000 3.719% 2.829%

WEIGHTED AVERAGE DEFAULT RATE (b) 1971 TO 2000 3.482% 2.558%1978 TO 2000 3.503% 2.563%1985 TO 2000 3.582% 2.565%

MEDIAN ANNUAL DEFAULT RATE 1971 TO 2000 1.656%

Notes

(a) As of mid-year.

(b) Weighted by par value of amount outstanding for each year

Source: Authors' Compilation and Salomon Smith Barney Estimates

F I G U R E 1

StandardDeviation

H I S T O R I C A L D E F A U L T R A T E S - S T R A I G H T B O N D S O N L YE X C L U D I N G D E FA U LT E D I S S U E S F R O M PA R VA L U E O U T S TA N D I N G

1 9 7 1 - 2 0 0 0 ( $ M I L L I O N S )

12 Sternbusiness

year in the history of the market.The benign credit cycle of 1993-1998, when the default rate wasbelow 2.0% each year, has clearlygiven way to a more turbulent andstormy environment.

So, will the next 18 months turnout to be as difficult and tumultuousas 1990-1991? Will default rates riseto approximately 10% as they did inboth 1990 and 1991? And, will themarket almost cease to function? Or,will returns rebound to almost theunbelievable annual level of over40%, as they did in 1991?

After the FallIn the early 1990s, the high-

yield bond market fell dramatically,and then rose even more dramatical-ly. In both 1990 and 1991, default

rates exceeded 10.0% of the market– much larger than the previous highof 5.8% in 1987. The total amountof debt defaulting in each of thesetwo years was over $18 billion. Thepundits who predicted the demise ofthe market looked like sages whentotal returns to high-yield investorsturned out to be -8.5% in 1990 –only the second year since 1978 thattotal returns were negative. Andsince Treasuries earned a positive6.9% return that year, the returnspread of junk bonds was a shocking–15.4%. At the end of 1990, theaverage historical annual return(starting from 1978, when the datawere first compiled) to high-yieldinvestors fell to 9.96% per year, andthe return spread was a mere 0.19%per year. Clearly, this was inadequate

compensation for the added risk ofhigh-yield bonds.

Then came the turning point in1991. Despite a second consecutiveyear of a default rate over 10%,high-yield investors earned a totalreturn of 43.2%, the highest everrecorded in the history of the market(Figure 2). Investors realized thatthe worst was over and that theexcesses of the 1980s had beenpurged. What remained were, for themost part, viable companies whosebonds were not going to default. Asthe operating performance of thesecompanies continued to improve, theprices of their bonds made a spectac-ular recovery. The relationshipbetween default rates and totalreturns is shown in Figure 3.

There is a striking parallel

F I G U R E 2

2000 (5.68) 14.45 (20.13) 14.56 5.12 9.44 1999 1.73 (8.41) 10.14 11.41 6.44 4.97 1998 4.04 12.77 (8.73) 10.04 4.65 5.39 1997 14.27 11.16 3.11 9.20 5.75 3.45 1996 11.24 0.04 11.20 9.58 6.42 3.16 1995 22.40 23.58 (1.18) 9.76 5.58 4.18 1994 (2.55) (8.29) 5.74 11.50 7.83 3.67 1993 18.33 12.08 6.25 9.08 5.80 3.28 1992 18.29 6.50 11.79 10.44 6.69 3.75 1991 43.23 17.18 26.05 12.56 6.70 5.86 1990 (8.46) 6.88 (15.34) 18.57 8.07 10.50 1989 1.98 16.72 (14.74) 15.17 7.93 7.24 1988 15.25 6.34 8.91 13.70 9.15 4.55 1987 4.57 (2.67) 7.24 13.89 8.83 5.06 1986 16.50 24.08 (7.58) 12.67 7.21 5.46 1985 26.08 31.54 (5.46) 13.50 8.99 4.51 1984 8.50 14.82 (6.32) 14.97 11.87 3.10 1983 21.80 2.23 19.57 15.74 10.70 5.04 1982 32.45 42.08 (9.63) 17.84 13.86 3.98 1981 7.56 0.48 7.08 15.97 12.08 3.89 1980 (1.00) (2.96) 1.96 13.46 10.23 3.23 1979 3.69 (0.86) 4.55 12.07 9.13 2.94 1978 7.57 (1.11) 8.68 10.92 8.11 2.81

YEAR HY HYRETURN (%)

TREAS SPREAD SPREADPROMISED YIELD (%)

TREAS

ARITHMETIC ANNUAL AVERAGE:

1978-2000 11.38 9.51 1.88 12.90 8.14 4.76

COMPOUND ANNUAL AVERAGE:

1978-2000 10.73 8.83 1.90

*End of year yields

Source: Salomon Smith Barney Inc.'s High Yield Composite Index

A N N U A L R E T U R N S , Y I E L D S A N D S P R E A D S O N T E N - Y E A R T R E A S U R Y ( T R E A S ) A N D H I G H Y I E L D ( H Y ) B O N D S( 1 9 7 8 - 2 0 0 0 )

Sternbusiness 13

between the increasing default ratesin 1989-1990 and 1999-2000; notethe dip in returns in 1990 and 2000and finally the resurgence in 1991.The question remains whether therewill be a comparable resurgence in2001.

Despite the high returns in 1991,however, the high-yield marketshrunk rapidly, from a high of $189billion in 1989 to $163 billion in themiddle of 1992. Between 1987 and1989, an average of $30 billion inhigh-yield bonds were sold eachyear. But in 1990 and 1991,the volume of new issues was$1.4 billion, and $10 bil-lion, respectively.

Since 1991, however,the growth in new issueshas been nothing short ofspectacular, with over$100 billion of new issuesin each of the last threeyears of the decade, and$45 billion in 2000.Between 1997 and 1999,new issuance of high-yieldbonds accounted for over halfof the bonds issued by industrialcompanies. At the end of thedecade, about $600 billion of high-yield bonds were outstanding, ascompared to under $200 billion atthe start of the decade. This $600billion represents roughly a third ofthe entire corporate bond market inthe U.S.

During the 1990s, the annualreturn spreads between junk bondsand Treasuries rose from near zero atthe end of 1990 to almost 2% peryear last year. As reported in Figure2, total compound annual returns onhigh-yield bonds for the 23-yearperiod from 1978 through 2000averaged 1.9% per year over thereturns of 10-year U.S. Treasuries.This means that a $1,000 investmentin high-yield bonds in 1978 would

have been worth over $10,400 at theend of 2000, as compared to just$7,000 for 10-year Treasuries. Andif one subtracts the average annuallosses from defaults of about 2.45%per year over the period 1978-2000from the average promised yieldspread (4.76%) over that same peri-od, the result (2.31%) is quite closeto the realized annual

return spread. Thus, one can attemptto predict future relative returns inthe high-yield market by comparingcurrent yield spreads to actual lossesfrom the primary risk component –defaults.

Deteriorating Credit QualityBut even as issuance of high-yielddebt continued to soar in recentyears, there were signs of trouble.The default rate in 1999 spikedsharply to 4.15% from 1.6% in 1998– the first time the rate topped 4%since 1991. One of several apparentreasons for the increase in defaults in

1999 was the seeming deteriorationin credit quality of newly issuedbonds. Over one fourth of the 125issues that defaulted in 1999 hadbeen outstanding less than 12months before they defaulted – and55% had been outstanding less than24 months. These percentages com-pare with just 4% and 20% from theperiod 1991-1998, and 7.7% and

24.3% for 1971-1999. In 2000,that proportion of defaulted

debt that had defaulted with-in two years of issuance

dropped to 38%. Still, asmuch as 69% of thenon-performing high-yield bonds had beenissued within the previ-ous 36 months.

To better under-stand these mortalitystatistics, however, it isimportant to analyze thepurpose of the financ-ing. Whether companies

are using junk bonds tofund LBOs (as they did in

great numbers in the late1980s, but not in the late

1990s), growth opportunities, orjust to refinance debt (as they have

done in most years), can tell us agood deal about whether these one-or two-year mortality results aretruly symptomatic of a decline incredit quality or can be explained byother factors.

There was, in fact, a decline incredit quality between 1997 and1999. As mentioned earlier, high-yield new issu-ance as a percentageof all corporate bond issuanceincreased dramatically over the samethree-year period. And within thehigh-yield sector, the percentage ofnew issues rated B and CCC alsoincreased. Indeed, in 1999, B ratedbonds comprised 66% of high-yieldissuance and 31% of all new corpo-rate bond issuance! CCC-ratedbonds were particularly evident in

14 Sternbusiness

the 1998 cohort, with$9.3 billion repre-senting 10% of allhigh yield issuance –a jump from previousyears.

At the start of2000, I said thatinvestors would likelyrequire additionalpromised yields tocompensate them forthe uncertainty aboutpossible higher defaultrates in the next fewyears. The large spikein yield spreads in2000 – 447 basispoints – seemed tobear out my predic-tions. (Actually, thedefault risk spikewas only 315 basispoints, since Treasuriesdeclined by 132 basispoints in 2000.)

Other Reasons for theIncrease in Defaults

In addition to the deteriorationin credit quality and the earlieroccurrence of defaults, four otherfactors contributed to the sizeableincrease in 1999 and 2000: (1) therecent increase in new issuance; (2)the Russian default in 1998; (3) anumber of “sick” industries despitethe economy’s overall strength; and(4) banks’ reluctance to refinance orgive additional waivers to the mar-ginal firm.

Because of the huge newissuance during 1997-1999, someincrease in default rates is expectedas these new issues age. In theabsence of any other developments,two simple principles known as“regression to the mean” and themortality or “aging” effect wouldhave led us to expect both the defaultamounts and the default rate in the

last two years to increase vis-à-visprior years.

But the surge in the default rateto over 4% in 1999, and over 5% in2000, was caused by additional fac-tors. One important consideration,though difficult to document withstatistics, is the ability of distressedfirms to refinance their indebtedness.Borrowers found it increasingly diffi-cult to refinance in the aftermath ofRussia’s 1998 default and the flight-to-quality that ensued. Without theRussian contagion, the default ratewould surely have been lower.

In recent years, there werenotable concentrations of defaults ina number of chronically or newly ail-ing industrial sectors. Such sectors asenergy, retailing, communications,healthcare, leisure/entertainment,and shipping were hit hardest. In2000, newly hard-hit sectors havebeen telecommunications, steel, and

movie theatre complex companies, aswell as some large asbestos-relatedcompanies.

The energy sector’s difficultiesreached their peak fairly early in1999, while retailing and textileshave long experienced chronic prob-lems. Industries such as communica-tions and healthcare became new“leaders” in defaults, reflecting thefrenetic new issuance in the formerand the overcapacity and govern-mental regulation of fees in the lat-ter. In sum, despite a vigorous econ-omy driven by technology and pro-ductivity growth, a number of sec-tors have been ailing, and going for-ward some will continue to flounder.

Finally, there is anecdotal evi-dence of an increasing trend of banksand other lenders who are no longerwilling to waive violations ofcovenants after just a few prior vio-lations. There appears to be pressure

F I G U R E 3

D E F A U L T R A T E S A N D T O T A L R E T U R N S ( 1 9 8 6 - 2 0 0 0 )

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000

60.00%

45.00%

30.00%

15.00%

0.00%

-15.00%-0.03

0.03

0.06

0.09

0.12Default Rate Default Rate Returns % Return

0

Sternbusiness 15

from the Federal Reserve Board inthe last two years for banks to recordhigher loss reserves and actualcharge offs when bank profits are athigh levels. Coupled with some indi-cations of a slowing of the UnitedStates economy and higher interestrates, on lower quality issues, thesefactors are acting to increase thelikelihood of defaults on bank loansand publicly held bonds. The Fed,however, seems to have finally let upon this pressure by lowering interestsrates by 50 basis point in January2001 and again by 50 in mid-March.

The Difference BetweenThen and Now

Although storm clouds hang over

today’s high-yield market, the cur-rent situation differs in a number ofimportant respects from a decadeago. Viewed from a purely statisticalstandpoint, 10% default rates in thenear future are certainly possible,but not likely. Statistical analysiswould suggest there is something likea 2.5% probability of default ratesreturning to their 1990 and 1991highs.

The market, however, is notanticipating such a dire scenario,since yield spreads were 9.44% as ofDecember 31, 2000, as compared toover 10.5% at the end of 1990. It isalso important to recognize that ahigh percentage of those distressedissues in 1990 were the result of

LBOs and other highlyleveraged transactions(HLTs). Although HLTsmade a strong come-back in the ’90s, theyare far more conserva-tively financed todaythan their ’80s coun-terparts. Defaults fromhighly leverage restruc-turings in 1999-2000did not account forany material amountof defaults. And theoutlook is for thissource to continue tobe less important. Iexamined the propor-tion of total new high-yield bonds issued fora number of statedreasons, includingacquisitions, leveragerestructurings (e.g.,LBOs), capital expendi-tures and other generalcorporate investments,and the refinancing ofexisting debt, between1986 and 1999. Andwhile the latter catego-

ry has been the most important use ofnew debt financing every year since1986, the levels of refinancing in1997-1999 are not exceptionallyhigh – about 44%. That’s below theaverage over this 14-year period. Onereason for this is that, althoughTreasury rates did fall in these yearsfrom 1996 levels, the yields on high-yield debt actually increased, makingrefinancing more expensive.

Overall, in recent years about20% of high-yield bond newissuance was used for acquisitionsand only 4-5% for leveraged restruc-turings. This compares to 10-15%for acquisitions and well over 30%for LBOs and recapitalizations in theyears leading up to the market’s

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

1990 1992 1993 1995 1998 1999 4Q00

F I G U R E 4

D I S T R E S S E D * A N D D E F A U L T E D D E B T A S A

P E R C E N T A G E O F T O T A L H I G H - Y I E L D D E B T M A R K E T

28%

31%

17%

14%

7%

26%

8%

15%

6%

7%

9%

5%3%

2%

*Defined as yield-to-yield maturity spread greater than or equal to 1000bp over comparable treasuriesSource: Salomon Smith Barney and NYU Salomon Center

16 Sternbusiness

problems a decade ago. Since lever-aged restructurings can lead tounsustainable levels of debt and pos-sible financial distress, the new issuemarket was decidedly more risky inthe earlier period.

One of the important similaritiesbetween 1990 and 2000, however, isthe proportion of the market that isdistressed. If we define distressedbonds as those with a yield-to-worst at least 10%(1000 basis points)above the risk freerate, 28% ofhigh-yield bondissues were in this precar-ious position at the end of1990, as compared to about30% as of December 31, 2000.

In Figure 4, we see that inearly 1990 the proportion of dis-tressed and defaulted bonds was41%, with 28% distressed (thetotal market includes defaultedbonds in this case). A bit morethan one-third of the 28% actu-ally defaulted in each of theyears 1990 and 1991. At theend of 1999 the distressed propor-tion was 9% of the market, and itgrew to 17% in mid- 2000, and to31% by the end of 2000, admittedlya dramatic increase. A good deal ofthis increase, however, is due to thebig decrease in treasury yields(Figure 2). If one-third of the dis-tressed proportion again defaults inthe next 12 months, we will have a10% default rate. Incidentally, 10%of a market that is over $600 billionworks out to a default total of over$60 billion for the next 12 months.I do not believe this will occur,however, even with the sudden cri-sis in California utilities and arenewed scare of asbestos-relatedbankruptcies.

I believe that the default rate willbe in the 6.5-7.5% range over thenext 12 months and will not reach

the higher levels that Moody’s andsome other analysts are forecasting.And, I am persuaded by the past andby the market’s dynamics, thatreturns will be substantial after thepeak of defaults, and perhaps evenbefore the peak – whenever it occurs.Indeed, in the first two months of

2001, returns have been over 8.0%on high-yield bonds.

A Word on the EconomyThe relationship between overall

economic activity and default rateshas always been tricky. Clearly,depressed economic growth anddeclining corporate profits and cashflows are related, in a negative sense,to default rates. But, the lead-lagrelationship is not very stable overtime. Still, the economic recession atthe start of the 1990s clearly was anadditional factor that helped pushdefault rates to double-digit levels.Despite a slowdown in economicgrowth, few economists are forecast-ing a recession in the next year ortwo. And, with the recent Fed inter-est rate cuts, we do not foresee the

same economic pressures on defaultrates in 2000 and 2001 as occurreda decade ago. Admittedly, there isgreat uncertainty today.

Déjà vu All Over Again?So is it déjà vu all over again?

Yes and no.Despite the apparent similarities,

there are sufficient differencesbetween 1990 and 2000. As a

result, the current marketdownturn will be less severe

and less dramatic than its1989-1990 predecessor.The high-yield bond

market will weather thisdownturn, just as it did in the

early 1990s. The present deterio-ration in credit quality will run itscourse, as investors refuse to contin-ue providing capital to undercapital-ized entities. New issue activities willno doubt fall off, and defaults willprobably continue at levels that,although unsettling, are not cata-strophic. Just so, subsequent returnswill be impressive, although not of1991’s magnitude. Indeed, in theearly weeks of 2001, new issues ofhigh-yield bonds surged as interestrates were lowered. A key questionwill be if this increase can be sus-tained. No doubt, the market will behit by periodic setbacks, and morebad news in the coming months.

But, as long as the vast majorityof issuing entities in the high-yieldmarket remain viable enterprises,the market for high-yield bonds willretain its position as an importantmajor source of finance for compa-nies worldwide and a legitimate andprofitable asset class for investors.

EDWARD I. ALTMAN is the Max L. Heine pro-fessor of finance at NYU Stern and a consultantto Salomon Smith Barney in the high-yield anddistressed debt areas.

Sternbusiness 17

18 Sternbusiness

© ARTIST(S)/SIS

With the NASDAQ notching one of its worst performances ever, 2000 may not have

been a stellar year for technology investors. Nonetheless, investors continued to embrace

technology and the Internet – especially when trading stocks. Today, there are an

estimated 18 million online trading accounts in the U.S. And online equity trades

constitute about 25% of industry-wide trades.

First offered in the 1980s, mainly

through proprietary networks, online trading caught fire in the 1990s. By allowing

consumers to bypass their personal brokers and place orders online, thus saving on

trading costs, online discount brokers attracted millions of customers in the past

several years. The spectacular growth of the

Internet in the mid-1990s encouraged brokers

of all sizes to focus on the provision of Internet trading. And as investors became

increasingly comfortable with Internet trading, firms ranging from AmeriTrade to

Web Street Securities hastened to offer their clients the ability to trade stocks online.

By Chris Stefanadis

ONLINE DISCOUNT BROKERS HAVE MADE LIFE EASIER FOR THE GROWING ARMIES OF DO-IT-YOURSELF

INVESTORS – AND HARDER FOR THE ESTABLISHED FIRMS WHOSE DOMINANCE THEY HAVE CHALLENGED.

P L A C E S

T R A D I N G

Sternbusiness 19

Where did online brokers comefrom? And, more importantly,where are they going? Havingemerged from a period of explo-sive growth, online trading isreaching a stage of more maturedevelopment. And just as furi-ously as they built scale,online brokerage firms arenow pondering and executingstrategic initiatives intendedto ensure their survival.

The Evolution of OnlineBrokers

The first discount brokersemerged in the late 1970s as aresult of deregulation in the U.S.securities industry. From 1792, theyear in which the New York Stock Exchange (NYSE) wasestablished, to 1975, NYSE members charged for theirservices on the basis of a minimum commission schedule.The NYSE had the authority, subject to permission fromthe Securities & Exchange Commission (SEC), to setminimum commission rates on stock transactions. Thisfixed commission regime limited price competitionamong brokers. Consequently, fixed commissions led tohigh rates, market fragmentation, and an oversupply ofancillary services.

Robert Schwartz, a finance professor at the ZicklinSchool of Business, points out that in large orders, thebroker’s profit could be as high as 90% of the commis-sion. To circumvent the fixed commission structure, largetraders were sometimes induced to turn to regionalexchanges, or to the third and fourth market, thus frag-menting the market system. Furthermore, the absence ofprice competition led to fierce non-price competition andan oversupply of potentially redundant ancillary servic-es. Commissions alone often paid for an entire packageof products, including order handling, advisory services,and research reports. Therefore, brokers effectivelyoffered indirect rebates to customers in the form of serv-ices, rather than a direct rebate in the form of dollars.

In 1968, the NYSE appealed to the Securities &Exchange Commission for what it thought would be aroutine rate increase. But the U.S. Justice Departmentunexpectedly intervened, questioning not only the needfor an increase, but also the very existence of a fixedcommission structure. After an investigation, the SEC

finally eliminated fixed commis-sions on May 1, 1975 – now

referred to simply as “May Day.”The effect of deregulation

on prices was dramatic. Asbrokers started to competeon price, rates fell sharply.At first, however, the newstructure benefited mostlymajor institutional investors.For example, Schwartz notesthat the commission rate forlarge institutional orders(orders larger than 10,000shares) fell to 0.31% of princi-pal in December 1978, from0.57% in April 1975 – a 45%

reduction. Individual investors didn’t get such breaks

until discount brokers emerged. In the mid-1970s, firmssuch as Charles Schwab and TD Waterhouse sprung up,operating with a fundamentally different business modelthan established brokers. These discount brokers soughtto lure customers by providing inexpensive trading com-missions, not by providing a range of ancillary services.Since they eschewed the creation of large researchdepartments, discount brokers were able to offer moreaffordable, no-frills service. Aggressive marketing andattractive pricing allowed them to create a new marketniche and drove the sector’s growth through the early1990s.

Technological FactorsThe tremendous spread of the Internet in recent

years transformed discount brokers into online brokersand encouraged a host of new entrants into the field.

Online brokerage services were first introduced inthe early 1980s. But they were initially offered throughproprietary networks, like CompuServe and the GeneralElectric Network for Information Exchange (Genie). Inthe mid-1990s, the growing use of the Internet inducedonline brokers to launch Internet trading. In the yearssince, several discount brokers, as well as pure electron-ic brokers, entered the new business segment and foughtaggressively for market share.

The Internet offered such firms essentially two tech-nological advantages.

First, online brokers can provide less expensive tradeexecution than their offline counterparts. Placing orders

20 Sternbusiness

TRADING PLACES

online allows investors to circumvent personal brokers,reducing transaction costs. As a large number ofinvestors established Internet connections, the webbecame a ubiquitous network that can be used as a com-munication channel between a brokerage firm and itscustomers. Online trading also lets brokerage firms auto-mate their order placement process, thereby economizingon personnel time and effort.

Secondly, the Internet contributed to the emer-gence of online trading by becom-ing a medium for the transmis-sion of information. Largegroups of consumers becameincreasingly sophisticated andmore able to direct their ownfinancial affairs without thehelp of a personal broker.The Internet facilitatesthe diffusion of informa-tion, eroding one of themain advantages of pro-fessional brokers: theiraccess to superior informa-tion.

The use of the Internethas led to a sharp drop intrading costs. Bill Bernhamand Jamie Earle of CS FirstBoston point out that the aver-age commission charged by the top-10 online tradingfirms fell from $52.89 at the beginning of 1996 to$15.75 in 1999 – a 70% reduction. The attractive pric-ing has stimulated great growth. According to the ABABanking Journal, the number of online accounts rosefrom 0.3 million in 1995 to 9.3 million in 1999. GregSmith and Adam Townsend of Chase H&Q noted that atthe end of 2000 there were approximately 18 milliononline trading accounts in the U.S. Furthermore, theABA Banking Journal pointed out that daily onlinetrades as a percentage of all equity trading soared from8.5% in 1997 to 15.9% in 1999. And Smith andTownsend note that today online equity trades accountfor approximately 25% of industry-wide trades.

Levelling OffThere is reason to think that this explosive growth

trend may not continue. To begin with, many of the mostactive investors have already embraced Internet trading.According to Henry McVey of Morgan Stanley Dean

Witter, less than 5% of all online accounts account formore than 50% of all online trades.

Economic theory points out that in building a net-work, firms often tend to establish the most valuableconnections first. In this respect, online firms initiallyfocused on attracting the busiest traders. However, asnew accounts start contributing less trading activity thanexisting ones, online brokers may find expansion an

increasingly costly task. It is easier to attracta small number of large customers than a

large number of small customers. A McKinsey & Co. report found

that 63% of all investors that openedonline accounts earlier than a year

ago now execute at least 13 tradesa year. However, only 41% ofinvestors that went onlinerecently are likely to execute 13or more trades a year. In thisrespect, practitioners point outthat the acquisition cost per cus-tomer has increased considerablyin the online brokerage industry.

Furthermore, the explosivegrowth of online trading hasattracted many entrants to theindustry, leading to intense compe-

tition. In addition to discounters andpure electronic firms, traditional full-service bro-

kers like Merrill Lynch and USB Paine Webber offeronline services. The proliferation of online firms maythus lead to even deeper price discounts and tighterprofit margins.

Exploiting AssetsThe rapid development period for electronic trading

may be ending. However, several online brokers havealready invested in important assets, such as strongbrand names and large installed customer bases. Onlinebrokerages believe that these assets may allow them tomaintain their growth by leveraging their power to otheronline financial sectors.

In particular, electronic commerce sectors, like theonline brokerage industry, offer large economies of scale.When the consumer network is large, a firm can enjoyscale advantages and supply a greater number of ancil-lary and complementary products to customers at alower price.

Furthermore, customers of e-commerce firms often

There’s reason to

believe financial super-

markets may not be viable in

the long run. Customers, for

example, may be reluctant to

aggregate their accounts

because of security and

privacy concerns.

Sternbusiness 21

face “lock-in” as the costs of switching from one compa-ny to another may be substantial. Changing to anotherbrokerage firm requires transferring one’s assets andfamiliarizing oneself with the web site of the new firm.As customers complain, switching brokers can sometimestake months. Even when switching costs appear to below, in comparison with the commissions for stock trans-actions, they may be significant.

The existence of scale economies and customer lock-in thus induces e-commerce firms to sacrifice early prof-its to aggressively seek new customers. For e-commercefirms, building a large customer base quickly is of utmostimportance, since the market is one of scale. In the earlystages, firms invest heavily in price discounting, adver-tising and marketing to attract more customers. Once aweb site becomes successful in selling one product, it canoften branch into others.

It follows that online brokers with a strong brandname and a large customer base possess valuable assets.The importance of the online brokers’ installed customerbase is highlighted by the fact that stockholdings cur-rently make up a larger share of overall householdwealth than ever before. In the first quarter of 2000,stockholdings accounted for 40% of household wealth,up from 35% in the first quarter of 1998. The control ofsuch assets can potentially be the powerhouse that driveselectronic brokerages’ future development, allowingthem to expand to other activities and maintain a highgrowth rate.

Branching Out – Figuratively Thus far, online brokers have focused mainly on

their core product – trading over the Internet. However,the provision of more integrated financial services is anatural evolution from their core activity of providingelectronic trading. A number of online brokers havetherefore launched ambitious expansion initiatives, someof which are aimed at providing one-stop financialservices: online checking accounts with ATM cards andprinted checks, savings accounts, certificates of deposit,credit cards, loans, bill payment services, insuranceplans, portfolio management, financial planning, andsecurities underwriting. The companies that manage topull together electronically all these services on a singleweb site will likely gain a significant competitiveadvantage.

Integration of financial services may take threeforms. First, a brokerage firm may expand its own oper-ations by supplying a broader gamut of products itself.

The broker, for example, may open a banking or aninsurance division, as E✴ Trade has done with itsE✴ Trade Bank. Alternatively, a brokerage firm mayform an alliance with a range of outside suppliers, offer-ing their products on its web site, as Charles SchwabCorp. has done with its mutual fund supermarket. Then,the site serves as an integrator, pulling together a groupof different financial companies under one roof. Or,thirdly, the broker may act as a “screen scraper,” gather-ing information and completing transactions on behalf ofits customers at all rival institutions. With the customer’sconsent, a screen scraper can poseas the customerherself, gain-ing access toa customer’saccounts atall financial companies. The screen scraperorganizes and displays this information on asingle web site, through which customers can handleall their financial affairs. AmeriTrade, for example, haslaunched a subsidiary, OnMoney, for the sole purpose ofproviding screen scraping services.

The objective of several brokers is, thus, the creationof online financial supermarkets that offer a broad arrayof services. Once a customer visits a supermarket, shecan handle all her personal financial needs. Competitionfor the provision of integrated services is one of the maindriving forces in the current wave of mergers and strate-gic alliances in the online industry.

Branching Out – LiterallyAside from possessing cutting-edge online capabili-

ties, some electronic brokerage firms also intend toestablish a limited physical presence. Customers maywant to have access to their supermarket through var-ious channels, like, for example, at a physical branch,by phone, or over the Internet. Integrated shoppingcan be the key to attracting mainstream customers.For instance, since customers still rely on cash for sev-eral daily activities, online financial firms will probablyneed to develop the ability to deliver cash or acceptcash deposits. Several online brokers, thus, aim atsupplementing their potential expansion to one-stopInternet services with the creation of a modest phys-ical distribution channel. Charles Schwab’s “clicksand mortar” strategy is a good example of how thephysical and online distribution channels can belinked.

22 Sternbusiness

TRADING PLACES

Benefits of One-stop Financial Services Financial supermarkets can afford their customers

greater convenience and superior product quality. In arecent American Banker/Gallup consumer survey, forinstance, about 50% of consumers expressed interest inaggregating their financial accounts and having onlineaccess to this information.

One-stop services can become more user-friendlythan separate products by putting every type of transac-tion just a “click” away from the customer. A customer,for example, can easily transfer money between herchecking and stock trading accounts or use money fromher checking account to buy an insurance product.

More important, online financial supermarkets willlikely enjoy the advantages of superior information.When a customer uses a single web site for most of herfinancial transactions, the site will be able to map the

financial profile of the customer and offer appro-priate, or even tailor-made, products. For

example, a financial supermarket will be in a better posi-tion to help customers set financial goals consistent withtheir risk preferences, choose a suitable portfolio ofassets, or seek the most appropriate loan arrangements.Combining their superior access to information withtheir online technological capabilities, financial super-markets will have the opportunity to offer “integratedpersonal financial management,” managing online theircustomers’ assets and liabilities on a continuous basis.Each customer could eventually have a personalizedweb page.

Competition from Other SectorsOnline discount brokers believe they are in a good

position to thrive in the market for integrated financialservices. Several brokers have already built a strongbrand name and a large customer base through their coretrading products and aggressive marketing campaigns.They now expect to leverage this asset to other onlinesegments, providing one-stop services.

However, in the new technological environment,firms from other sectors are also investing in the provi-sion of integrated financial services. Aside from onlinebrokers, the companies that actively contend for marketshare in this segment include online banks, traditionalbanks and brokerages, and general e-commerce portals.In the presence of such intense competition, it is not clearthat online brokers will actually reap the benefits of inte-gration.

Furthermore, there are concerns that financialsupermarkets may not be viable in the long run.Customers, for example, may be reluctant to aggregatetheir accounts because of security and privacy concerns.Or, consumers may prefer to buy financial products from“specialists,” rather than from supermarkets, becausethey are willing to deal with “experts,” rather than withgeneralists.

Despite these implications, several online brokersfeel comfortable coexisting with firms from other sectorsin the provision of one-stop services. Electronic brokersexpect that their nimble structure and state-of-the-artefficiencies, as well as their brand name and customerbase will enable them to prosper even in the face ofintense competition from other sectors.

It is too early to predict how the new technologicalenvironment will reshape the online brokerage industry.And it is equally difficult to divine which of today’splayers will emerge as long-term survivors. One thing iscertain, however. The real long-term winners of the tech-nological, regulatory, and attitudinal changes that helpedcreate online brokerages will be investors, who will reapthe rewards of better and more inexpensive products andservices.

C H R I S S T E F A N A D I S has a Ph.D. in economics and financefrom NYU Stern and is a research economist at the Federal ReserveBank of New York.

The views expressed in this paper are those of the author and do not nec-essarily reflect the position of the Federal Reserve Bank of New York or theFederal Reserve System.

Sternbusiness 23

he never-ending presidential campaign of2000, which stretched for 36 days beyond thetraditional November end point, was one forthe history books. And as such, it attracted

the attention of a range of academics: law professors,historians, political scientists, and communicationsexperts.

But another group of academics also found themselvesriveted to the drama in Florida: statisticians. It led to abonanza of interest among professional statisticians. In par-ticular it led to discussions in many statistics classes, includ-ing my own, on the applications of sampling and other issues.A rash of articles and even web sites discussing the issues,appeared in the days following the election. Contrary to howthe public felt, we statisticians found the debate and thedelay incredibly interesting. As a matter of fact, many of uswere sorry that it ended.

To recap briefly: On election night, the television net-works, acting on estimates by the Voters News Service (VNS)prematurely said Democratic candidate Al Gore was the win-ner of Florida and its 25 electoral votes. That was at roughly7:50 p.m. election night. Two hours later, they retracted thisprojection, stating that Florida was too close to call. Early thenext morning, Republican candidate George W. Bush wasnamed the winner of the Sunshine State and its crucial electors,which were enough to give him 271 electoral votes and thePresidency. About one hour and 40 minutes later this projec-tion was retracted and once again Florida was too close to call.This led to the unprecedented and bizarre situation where theDemocratic candidate, Gore, retracted his concession, madeprivately to Bush.

According to Warren Mitofsky, who was the founder andthe former director of VNS, the reasons for these occurrenceswere as follows: the first projection, that Gore won Florida,was caused by statistical error. We have to remember that

projections, based on exit polls, are subject to error with alow probability of being incorrect. Projection mistakes dooccur and this was such a case. However, the second projec-tion, that Bush won Florida, was caused by a data error inDaytona Beach, which is in Volusia county. Twenty thousandextra votes were mistakenly credited to Bush by VNS shortlyafter 2 a.m. on November 8, 2000. The county noted the cor-rection before 3 a.m. that morning, decreasing Bush's lead bya total of 25,000 votes.

The next day, it was revealed that the actual margin wasremarkably close, about 1,700 votes out of six million cast(or less than 0.03% of the total votes cast). Indeed, Florida’selection law required that if the margin of victory for anycandidate was less than one half of one percent of the totalvotes cast, a recount would be mandatory. As the margin wasmuch less than 1%, a machine recount had to be carried out.The recount was carried out, resulting in an even narrowergap of roughly 350 votes, with Bush still in the lead (withouta count of an expected two to three thousand overseasabsentee ballots).

he recount process was further complicated by sev-eral concurrent lawsuits, charges and counter-charges of election-day shenanigans, ballot irregu-larities, and political bias. One of the issues was the

confusing use of the butterfly ballots in Palm Beach County,which may have resulted in a large number of Gore votersmistakenly voting for Reform Party Candidate PatrickBuchanan. Buchanan received a much larger than expectednumber of votes in this county (approximately 2,500 more).As a result of these lawsuits, the public has been exposed tostatisticians who have testified as expert witnesses in variouscourt cases.

One of the key issues to be resolved was how to count theso-called undervote. The undervote represented the ballotswhich were cast but for which the machines, for one reason

T h e r o l e o f s t a t i s t i c s i n F l o r i d a ’ s B u s h - G o r e r e c o u n t d r a m a

By Aaron Tenenbein

T

T

24 Sternbusiness

T O C O U N T O RN O T T O C O U N T:T H A T W A S T H E Q U E S T I O N

Sternbusiness 25

ILLUSTRATION BY DAVE BLACK

or another, did not record a vote for President. And here’swhere the statisticians came in. One of the main questionswas the use of probability and statistics to determine whetheror not to carry out a hand count of the Florida undervote.

Apart from the complicated legal propositions, therewere three statistical questions necessary to answer in orderto establish whether a statewide recount in Florida should beundertaken. First, was there a substantial number of ballotscast for the presidency for which no vote was recorded bymachine? Second, could a substantial percentage of theseballots be recovered, so as to classify ballots for which a votefor President was recorded? Third, was there a probability,however remote, that the election result in Florida could beaffected by a full count of the votes?

n testifying for the Gore-Lieberman campaign in oneof its Florida state lawsuits, Yale statistics professorNicolas Hengartner weighed in on the subject. Hedemonstrated that the first two conditions were met.Arguing for a hand count of the ballots which were not

counted by the machine in three Florida counties,Hengartner showed that there was an undercount, whichranged from an average of 1.5% in counties using old-stylepunch cards to an average of 0.3% in counties using moresophisticated optical scanning devices. Furthermore, hereported that varying percentages of these ballots (in theneighborhood of 25%) could be identified as votes for Presi-dent by using hand counts, based on data obtained in thecounties, which had already carried out hand counts. He didnot testify on the third issue. However, the vote was so closethat the probability of overturning the election of Florida,although difficult to determine, was not negligible. Cross-examination of Hengartner involved the issue of whether theundercount was machine related. In other words, did thepunch card machine cause this undercount? This is a muchmore difficult question to answer. This issue is important toresolve in order to determine the future use of punch cardballots in elections. However, in this present situation, it isnot relevant because the important issue is whether or not anundercount did occur and not the reasons for this occurrence.And the evidence did establish that there was an undercount.

Ultimately, the testimony of this statistician failed tocarry the day. The Florida Supreme Court did rule that astatewide manual recount of all ballots should be undertak-en. But the U.S. Supreme Court in December brought thecounting to a stop, paving the way for Florida’s state govern-

ment to bring thematter to a close andaward the state’s electors toBush.

To this statistician’s mind,a full statewide recount wouldhave been the optimal solution.Any partial recount – i.e.recounting the votes in selectedcounties – could have beenbiased. And while a partialrecount may have favored onecandidate over the other, itwould not have satisfactorilyanswered the vexing question as towho really won the popular vote inFlorida. A hand recount may nothave been a perfect solution.Uniform procedures as to how toclassify the undervote as a vote mayhave been difficult to establish, but itwas not impossible. And it is certainlypreferable to ignoring the undercount.

A courtroom setting is not the idealplace to demonstrate the application ofprobability and statistics. Part of the prob-lem is the fact that the concepts of statisticsare difficult for the layperson to understand.The issues became even murkier in a courtroomwhere dueling attorneys raised questions onissues that are already confusing. The confusion,which the public must certainly have felt, isechoed by a famous quote by Benjamin Disraeli,a nineteenth century Prime Minister ofEngland, who said: “There are three kinds oflies: lies, damned lies, and statistics.”

I will make two predictions. First, this issue willnot go away. Currently, there are severalgroups counting the undervote under variouscriteria. Second, in future elections, the useof punch card balloting will disappear.

A A R O N T E N E N B E I N is professor of statisticsand actuarial science at NYU Stern.

I

THERIGHTSTUFF

26 Sternbusiness

What will turn today ’s entrepreneurs into tomorrow ’s

big-time business people? After tracking the careers

of dozens of contemporary moguls, one author concludes

that it has as much to do with the content of their

character as with the make-up of their balance sheets.

By Roy C. Smith

The 1999 Forbes 400 list included 251individuals whose source of wealth wasdescribed as “self-made.” These self-madeentrepreneurs are among the greatest cre-ators of new wealth in the country. Theyinclude billionaires whose money comesfrom medical devices, computer software,railroads, testing laboratories, real estate,home building, stock market investments,trading stamps, oil and gas, computerassembly, direct sales organizations, retail-ing, heath care, mobile phones, music andrecords, newspapers and media, insurance,cable TV, public storage, plastics, and adozen other businesses.

These days, of course, entrepreneurs areregarded as something far greater than justordinary businessmen – hired hands andadministrators. Indeed, today’s entrepreneursmay have become the cultural replacement forthe famous American “rugged individualists”of the last century, the ones who tamed theWest and built great industries from nothing.They can be extreme risk-takers, and colorful,self-confident persons of strong character andpersonality. Think of Bill Gates, Steve Jobs,Michael Dell, Richard Branson, Ted Turner,and Donald Trump.

Stripped of all the hype and filler, howev-er, “entrepreneurs” are simply people who, asindividuals or in small groups, have started oracquired businesses and attempted to growand/or alter them to a point where they couldcash in successfully on the rewards. Butachieving entrepreneurial success may be thegreatest challenge available to anyone in busi-

ness. According to the Small BusinessAdministration, there were about 6.6 millioncorporations in the United States in 1996, thevast majority of which were small businesses.Only the tiniest sliver ofthem will grow into big busi-nesses.

Academics have beenstudying self-made business-men, hoping to find amethodology to teach toyoung entrepreneurs. Wehave not yet found, ofcourse, a simple, repeatableformula for turning small orsubstantially restructuredbusinesses into gold mines.And many academics believethat great entrepreneurialsuccess is as much influencedby luck as by skill, or by for-tuitous (if unwise) risk tak-ing as by any other quality.

Nonetheless, I believedthere is something to belearned from studying whatthe successful entrepreneursactually did to become suc-cessful. And while conduct-ing research for my recentlypublished book, The Wealth Creators: TheRise of Today’s New Rich and Super-Rich, Ilearned a good deal about factors that seemto increase the probabilities of success. Andthere are in fact several sine qua nons – fac-

The first characteristic of

a successful entrepre-

neur is perhaps the most

obvious one: a wilingness

to take risk. Regardless

of other attributes, this

first, primordial require-

ment must be present.

Sternbusiness 27

© A

RTI

ST(

S)/

SIS

available in large quantities.The market was totally new at the time,

and there were no barriers to entry, so a largenumber of small refiners and oil producerssprung up. The fledgling firms cut prices, andconspired with or against each other to try tomake progress. Since business conditions inthe industry became chaotic, Rockefellerchanged his central idea – instead of justrefining crude oil, he would focus on consoli-dating the refining, transportation and mar-keting components into a new industry. Thisdecision, a vision of opportunity and a whollydifferent way to develop it, played to all ofRockefeller’s organizational and administra-tive strengths – his comparative advantages –that enabled him to become a success.

Starting a business with an idea that is notnew at all, something that just presents anoth-er choice for the consumer, can be a verytough grind for the entrepreneur. Anotherbank branch on the corner, or a new videotaperecorder, may take forever to gain any kind ofmarket share and could divide the total prof-its available in the market into increasinglysmaller pieces. Something somewhat new,however, can capture the market’s attentionwithout having to completely reeducate it. Itcan quickly change market dynamics, increas-ing total demand for, say, tennis shoes becausethey are no longer perceived as tennis shoesbut as performance enhancing footwearfavored by professional athletes with a differ-ent shoe for every sport. Of course, the idea,or vision, has to be strong enough to alter thedynamics. But it is clear that many of themost powerful new business ideas have notbeen all that new.

Making It Big – Thinking BigAny new (or somewhat new) business

vision ought to lead to a business with a largepotential market. The national market forkerosene must have seemed enormous to theyoung Mr. Rockefeller, who became a billion-aire long before the automobile assured thefuture of the oil business. Ray Kroc, a travel-ing milkshake-machine salesman, became abillionaire because he realized that a smallhamburger stand could be cloned into thou-sands of McDonald’s stores nationwide

tors without which entrepreneurial successprobably cannot happen.

The Vision to Take RisksThe first characteristic of a successful

entrepreneur is perhaps the most obvious one:a willingness to take risk. Regardless of otherattributes, this first, primordial requirementmust be present – the willingness to step offinto the void, risking most of what other peo-ple think of security and well being. An entre-preneur must first quit his or her “day job.”

A second requirement is the ability toidentify and develop viable plans for captur-ing an attractive business opportunity – pos-sessing the vision. After all, when an entre-preneur decides to enter a business, an equi-librium already exists between those productsand services that are in circulation, and thosethat could be, but aren’t. Unless a new prod-uct or service can overcome the existing bar-riers to entry into the market, it has nochance.

The entrepreneur seeks away around this equilibriumby finding something new ordifferent that will reset theequilibrium more advanta-geously. It doesn’t have to bean entirely new product oridea (like Edison’s electriclight, which took a long timeto introduce and requiredexpensive power plants andtransmissions lines). But it hasto be sufficiently new tochange the original configura-tions of the market.

John D. Rockefeller didnot invent the oil lamp. Butafter the discovery of crude oilin Titusville, Pennsylvania, in1859, the dry goods merchantforesaw the opportunity ofproducing large quantities ofkerosene from petroleum to besold as a cheap, efficient fuelfor illumination. At the time,whale oil was the principalsource of lamp illumination,but it was expensive and not

Starting a business withan idea that is not newat all, something thatjust presents anotherchoice for the consumer,can be a very tough grindfor the entrepreneur...Something somewhatnew, however, can cap-ture the market’s atten-tion without having tocompletely reeducate it.

28 Sternbusiness

THE RIGHT STUFF

through a franchising process. Only the con-cepts that can be applied nationally, or evenbetter, globally, have the potential to be trulybig.

Ted Turner, the founder of CNN, saw thatthe value of cable television ultimatelydepended on what went through the cables.He knew that many new channels would bemade available through the cable hookups,and that these channels would be offered tosubscribers who would pay to get certainkinds of programming. Turner already ownedsome sports teams and WTBS, an Atlanta-based UHF broadcasting station. In 1976, hepioneered the “superstation” concept byarranging to transmit his UHF signal by satel-lite to content-starved cable system operatorsall over the country. The idea caught on. Andin 1980, Turner introduced the all-newschannel for his system – CNN. Critics thoughtTurner’s idea was a bit crazy, because mostpeople could not imagine tuning in just tonews all day long. But the cable operators,who were selling a package of several chan-nels for a fixed cost, were eager to add it totheir range of offerings. The rest is history. By1985, Turner’s cable offerings reached 80% ofthe American homes equipped with cable andCNN was frequently the item that was most indemand. Before long Turner had tens of mil-lions of subscribers essentially paying to haveCNN in their house for a few minutes eachday – especially after the 1991 Gulf War.

Making It Happen – ExecutionDistributing a product to the national or

global market quickly is a very complex anddifficult undertaking. At all times, an entre-preneur must be able to make the critical saleat the right time, or deliver on promises made.Some “breaks” may appear to be just luck,and luck is an important factor in all equa-tions for success. But the good field operatorhelps good luck along by constantly findingother ways to accomplish things, and by mar-shalling talent and resources just where theyare needed at just the right time. In otherwords, successful entrepreneurs must execute.

In 1995, Marc Josephson, who trained asan engineer, formed a new company to con-nect New York City’s new Information

Technology Center (ITC) at 55 Broad Streetto the Internet “backbone” network. TheITC was to become the focal point of NewYork’s effort to create a new media and tech-nology industry, and it needed efficient, reli-able, and, above all, fast access to theInternet.

Years before, New YorkCity had been encircled infiber-optic cable. Several ofthese cable circuits wereessentially idle, but containedbandwidth ample for anyknown purpose. Josephsonfigured out how to connect 55Broad to the cable circuitdirectly, and was able to passthe bandwidth advantages onto the building’s occupants atlow cost. To capitalize on hisdiscoveries, he set up a smallcompany to lay the necessarywiring and lease Internetaccess to the tenants directly,which he could do at muchless cost than the regularaccess providers. Josephsonoffered to wire Manhattan’s Jacob JavitsConvention Center and then persuadedRockefeller Center to let him wire up thewhole complex, which he then had to leaseto the tenants, one office at a time.

With such high-profile assignments,Josephson was a happy entrepreneur. But hewas faced with the challenge of building outhis business as quickly as possible so anoth-er Internet access provider didn’t figure outwhat he had done and offer the same service.To launch his business nationally, he neededa team of competent engineers and capital.Through a lawyer friend he met an “angel”investor, who invested a few million dollarsof seed capital. By the end of 1998, twoyears after finishing 55 Broad, Josephson’scompany, now called IntelliSpace, had wired50 buildings for online services in New YorkCity; there would be 140 a year later. Thebuilding space wired by the company wasgrowing at 400% per year, and Josephsonwas shaping up plans to replicate itsbusiness in Philadelphia, Boston and

Sternbusiness 29

The good field operator

helps good luck along by

constantly finding other

ways to accomplish

things, and by mar-

shalling talent and

resources just where

they are needed at just

the right time. In other

words, successful entre-

preneurs must execute.

Washington, DC. In 1999 IntelliSpace com-pleted a $35 million second round of financ-ing. Josephson was able to attract the invest-ment largely because his investors had confi-dence in his engineering skills and executiveability to carefully plan out a step-by-stepprogram for building out the business.

Margins MatterThe biggest ideas, no matter how well

they are executed, will certainly fail if ade-quate margins cannot be earned. HenryFord’s mass manufacturing of Model T’s wasan operation designed to exploit economies ofscale so his cars could be sold at prices lowenough to turn the automobile into a massconsumer product. Although the cars weresold cheaply, the volume was large enough forhim to capture healthy operating margins.These permitted Ford to pay generous wagesto increasingly skilled assembly line workersand to invest in facilities to make almosteverything that went into the car at Ford’shuge central factory at River Rouge. Theinvestments increased his margins further,permitting even more investments to improveand expand the business.

One small business I know calledMaxFlight, founded in 1994,makes aircraft flight simulatorsfor amusement parks. Thefounder, Frank McClintic, aformer helicopter pilot and anatural, tinkering mechanic,was familiar with aviationtraining simulators andthought he could improve ontheir motion characteristics.However, he believed that sell-ing the product to the militaryor the airlines would meanhaving to lower margins tounworkable levels. So he decid-ed to change his product tomake it into a cheaper, glitziermodel for the amusement busi-ness, where it could be sold bythe ride. The product becamean enclosed audio-visualenhanced, aircraft cockpit thatrotated on all axes. The rider

would be treated to dramatic virtual carriertakeoffs and landings, or realistic dogfights. Bigpark operators thought they could sell a lot ofrides, and the public liked it enough to pay upto $10 for a five minute ride.

McClintic knew he had to keep his costslower than his competitors, so that he could cre-ate a barrier to potential market entrants.Mostly, his costs were in the sophisticated partsthat the machine required. So McClintic focusedon lowering these costs, scouring technical pub-lications for lower-cost replacement parts. Hefound several, including surplus governmentequipment that he could buy at a distressedprice. These efforts substantially lowered hiscosts and maintained his margins well into dou-ble digits. Like Henry Ford, McClintic knows hemust continually re-engineer his machines andimprove their performance capability to keepahead of his larger competitors.

Timing Michael Bloomberg stepped into the finan-

cial data market with his Bloomberg machinejust at the right moment. He understood thebusiness well enough to know of the machine’scoming importance and that his competitorsmight be slow to develop a similar product – but

only for a while. He moved withinhis two to three year window andsuccessfully launched his product.Just so, Charles Schwab was anearly exploiter of the opportunitiesin discount stock brokerage, andgained a significant market shareand $1.8 billion of net worth in theprocess.

In early 1998, Greg and GlennMorello, two brothers in NewJersey, decided to expand theirBridgewater Autobody repair busi-ness from two shops into a dozenor more. Their idea was to gainsome economies of scale by operat-ing at a larger size, and then tocontract with auto insurance com-panies for bulk-purchases ofrepairs. The insurance companieswanted to be sure that the repairshops they used were reliable and

30 Sternbusiness

THE RIGHT STUFF

honest, but they particularly wanted to beable to contract for repairs at a lower cost.The two brothers believe they have a year ortwo to be able to pull their new companytogether, before an aggressive or betterfinanced auto-body repair chain comes intotheir market and forces them out. Time iscritically important for them.

In the fragile world of the entrepreneur, agood idea performed too early or too late isnot worth nearly as much as one performedat the just right time. Indeed, a good ideamay be worth very little if badly timed. Formost hard-pressed entrepreneurs who takeone busy year at a time, good timing, howev-er, can mean the difference between beingquickly established in a marketplace, or notat all. But that early market position has to bereinforced and defended against strong com-petitive efforts, better products, and, ofcourse, trend changes.

The Right StuffTom Wolfe’s best-selling book about

American astronauts, The Right Stuff,describes the skills and the characters of thefirst Project Mercury team, men like AlanSheppard, John Glenn, Scott Carpenter,Gordon Cooper, Wally Shirra, Deke Slayton,and Gus Grissom. Wolfe sensed that as differ-ent as they all were from one another, they allhad something that seasoned aviators knewto be “the right stuff.” They behaved likepeople think fighter pilots should – they werefearless, of course, and somewhat reckless,though always confidently so. They hadextremely quick reflexes, remained coolunder pressure, and never showed any con-cern that they might end up among the grue-some statistics of their profession. Not every-one had the right stuff, but you had to have itto make it to the top in combat aviation or thetest pilot business. It was hard to define, buteveryone knew it when they saw it.

Entrepreneurs, too, must have the rightstuff if they are going to make it big. For peo-ple in business, the right stuff is the specialsoftware inside the product that makes amaz-ing things happen, that allows them to havevision, to think big, to execute, and to have

good timing. Indeed, the quintessentialAmerican big time entrepreneur is a compos-ite of experiences, skills, toughness of charac-ter and self-control that are, like our astro-nauts, unique to their profession.

They have to have a certain mindset thatmost business people do not have. They wantto bet on themselves and their abilities, evenif the odds look pretty long. They are notespecially concerned with security, or appear-ances, or creature comforts. They performwell under pressure, and adapt optimisticallyto even harsh disappointments.They believe totally in what theyare doing, but are prepared tochange things often, so that whatthey end up doing may not havebeen what they started out believ-ing. They often demonstrate a dis-dain for large, bureaucratic work-ing environments, and the lines ofauthority that go with it. They canbe hard and ruthless competitors,but, when successful, much moregenerous with their time andmoney than those who inheritwealth and slowly feed the oldmoney charities. In short, they arehighly driven to succeed, and todo things “their way.” They havea lot of attitude.

Most real estate investors will tell you thatthe three most important factors in determin-ing the success of a real estate investment are“location, location, and location.” Mostinvestors in small businesses will tell you thatfor them the three most important factors are“the CEO and the management team,” infirst, second and third place. Regardless of thebrilliance of its product idea, or the potentialsize of its market, an entrepreneurial compa-ny’s survival frequently depends on whetheror not its leader has the right stuff.

R O Y C . S M I T H is the Kenneth G. Langone professorof entrepreneurship and finance at NYU Stern. This articleis adapted from his new book, The Wealth Creators: TheRise of Today’s New Rich and Super-Rich (St. MartinsPress).

In the fragile world of

the entrepreneur, a

good idea performed

too early or too late is

not worth nearly as

much as one per-

formed at the just

right time. Indeed, a

good idea may be

worth very little if

badly timed.

Sternbusiness 31

By Randall Morck and Bernard Yeung

When asked to predict activity in the stock market, J.P. Morgan replied that stock prices would fluctuate.

Modern finance theory ascribes meaning to these fluctuations. The stocks of successful, well-run, or

lucky companies rise. Those of unsuccessful, misgoverned, or unlucky companies fall. While portfolio

managers view the volatility of individual stocks as a problem to be overcome through diversification,

corporate executives watch their stocks rise or fall with euphoria or dismay. A soaring stock price helps

a company grow, by raising bond ratings and bringing in more money from additional share offerings.

A plummeting stock price unsettles creditors and raises the dilution cost of each dollar of new equity.

his realignment of stockprices and, more impor-tantly, the redirection ofcapital flows that it causes,

are thought to underlie the growthand prosperity of modern free marketeconomies. An economy that investscapital in poorly run or ill-conceivedenterprises cannot provide as high astandard of living as an economy thatputs capital where it is most useful.That this mechanism works well is aprime argument by those who opposegovernment intervention in the econo-my. That stock fluctuations are ameaningless throw of dice is a primeargument made by those who mistrustcapitalism and all it stands for.

In a recent study, we uncovered apuzzling regularity in the fluctua-tions of stock prices across the mar-kets of different countries. In theUnited States, and most other devel-oped countries, stocks tend to movein a relatively unsynchronized man-ner. Ford rises, while GM simultane-ously falls, for example. But stocks inemerging markets, like those ofLatin America, Asia, and EasternEurope, exhibit a uniformly differentsort of volatility. In contrast to theaction in developed countries, stocksin emerging markets tend to move upor down together – en masse.

We began by analyzing stockreturns data from the first 26 weeks

in 1995 to measure the degree ofprice synchronicity in some represen-tative stock markets. We calculatedthe fraction of stocks that moved inthe same direction in a given country,filtered out those whose prices didn’tmove, and charted the number thatrose and fell. As seen in Table 1, inemerging markets like China andPoland, over 80% of stocks oftenmoved in the same direction in agiven week. For example, in Chinaover 70% of stocks moved in thesame direction in 18 weeks. InPoland, 100% of traded stocksmoved in the same direction duringfour of the 26 weeks, and 70% movedin the same direction in 20 weeks. By

Stocks in emerging markets tend to move in the same direction whilethose in developed markets tend to fol low divergent paths. Why?

IN SYNC

32 Sternbusiness

© A

RTI

ST(

S)/

SIS

T

contrast, Denmark, Ireland, and theUnited States lacked any instances inwhich more than 57% of the stocksmoved in the same direction duringany week.

We then correlated the full 1995year price movements with the sizeof a country’s per capita GrossDomestic Product between 1992 and1994. Per capita GDP, of course, is ageneral measure of economic devel-opment. But it can also serve as aproxy for the economic structuresand attributes that make up devel-oped economies. And we found that,in general, high-income countrieshave asynchronous stock prices, withthe U.S. having the lowest fraction of

stocks moving together. In contrast,low-income economies have higherdegrees of price synchronicity, withPoland, China, Taiwan, Malaysia,and Turkey leading the pack. A clos-er look at the data reveals that thenations we investigated groupedthemselves into two data clusters:high-income countries with low syn-chronicity and low-income countrieswith high synchronicity.

Why should there be such a dif-ference between the fluctuations wesee in the stock markets of high andlow income countries? We consideredseveral possible explanations. First,firms in low-income countries mighthave more correlated fundamentals.

After all, low-income economies tendto be small, undiversified, subjectto unstable macroeconomic policy,and characterized by intercorporateequity cross-holdings. All these fac-tors can turn events that affect oneindustry into market-wide events.Second, low-income economies usu-ally provide poor and uncertain pro-tection of private property rights. Ifthis reduces the transparency ofcompanies in these countries toinvestors, the fine tuning of individ-ual companies’ stock prices that wesee in developed economy marketsmight not happen. If so, the stockmarkets of these countries might befailing in their primary social task,

Sternbusiness 33

the allocation of capital. We concluded that measures of

fundamentals correlation do notexplain our finding, but that meas-ures of private property rights pro-tection do. Indeed, variables likemarket size, country size, economicdiversification, macroeconomic poli-cy stability, and intercorporate earn-ings correlation are, at best, onlyvaguely related to stock price asyn-chronicity. The measures of develop-ment that are most closely related tostock price asynchronicity are meas-ures of the integrity of government,the efficiency of the judicial system,and the rule of law.

Economic ExplanationsTo determine what explains the

highly significant negative correla-tion between stock price synchronic-ity and per capita GDP, we investi-gated which particular developmentmeasures are most correlated withstock price synchronicity.

irst, we considered severalpossible economic explana-tions. The negative correla-tion between stock price co-

movement and per capita incomecould be due to the fact that thecompanies in low-income economiestend to have more correlated eco-nomic fundamentals. Or, it could bethat unstable market fundamentalsare caused by macro-economicinstability – erratic or unpredictablegrowth. In such economies, volatilemarket fundamentals may over-whelm variations due to firm-specificfactors, so that stock prices tend tomove together.

Another factor could be countrysize. Since economic activity in smallcountries is geographically localized,nearby geopolitical instability orenvironmental catastrophes likeearthquakes might have market-wide effects that would not be as evi-dent in a larger country. For exam-ple, Finland’s economy shrank by

15% in the early1990s as then e i g h b o r i n gSoviet Union dis-integrated andFinland’s role asa gateway toRussia temporar-ily lost value. Inaddition, stocksin large countriesmight move moreindependent lythan those insmall countriesbecause of indus-trial and econom-ic diversity. Forexample, if oilprices fall, theprospects of Ohiomanufacturingfirms brighten,while those ofTexas oil compa-nies dim. In con-trast, stocks in asmaller oil-pro-ducing country,like Venezuela,might move moresynchronously asoil prices change.In some economies, listed firms couldbe concentrated in just a few indus-tries, which tend to move in sync.And since the stock markets in someeconomies may be dominated by afew very large companies, a highdegree of stock price synchronicitymay result if most other listed firmsare suppliers or customers of thesedominant firms.

We tested these factors by con-structing indexes to stand as proxiesfor the variables and by using regres-sion analysis on our data. And wefound some intriguing results. Forexample, price synchronicity is nega-tively correlated with a country’s geo-graphical size – the bigger the coun-try is, the less likely it is that its stocksmove together. We also found that

price synchronicity is positively corre-lated with both GDP growth varianceand earnings co-movement. However,these correlations are all statisticallyinsignificant. And we found thatgreater economic and industrialdiversity is not consistently correlatedwith less stock price synchronicity. Soclearly, our basic result cannot be duesimply to the fact that low-incomecountries tend to be small and undi-versified. Overall, in fact, these corre-lations suggest that no one structuralvariable, on its own, satisfactorilyexplains the link between per capitaGDP and stock price synchronicity.We checked to see if these structuralvariables, acting in concert, mightexplain the link. But these resultswere similarly inconclusive.

34 Sternbusiness

sample 308 stocks 38 stocks 6,889 stocks

Week

CHINA POLAND U.S.%Up %Down %Same%Up %Down %Same%Up %Down %Same

T A B L E 1

THE FRACTION OF STOCKS WHOSE PRICES GO UP, GO DOWN, ANDREMAIN THE SAME DURING EACH OF THE F IRST 26 WEEKS OF 1995

1 32 61 7 97 3 0 47 29 242 4 89 6 5 95 0 47 38 153 6 88 7 59 31 10 49 37 134 7 88 5 3 92 5 54 32 145 84 8 7 3 97 0 33 53 156 7 50 42 100 0 0 44 43 147 59 31 10 15 77 8 57 30 138 18 73 9 10 90 0 48 38 149 71 22 7 82 13 5 42 43 1510 93 4 4 95 5 0 44 42 1411 9 88 3 3 95 3 33 52 1512 41 51 7 0 92 8 50 37 1313 89 7 4 15 67 18 41 44 1514 84 9 6 100 0 0 50 35 1515 21 73 5 100 0 0 47 37 1516 18 75 7 56 38 5 45 40 1517 29 63 8 90 10 0 41 44 1518 5 92 3 8 92 0 50 35 1519 35 56 9 41 49 10 46 40 1420 29 60 11 87 10 3 49 37 1421 89 8 3 0 100 0 42 44 1422 21 76 4 92 5 3 46 39 1523 16 79 5 74 23 3 47 39 1424 55 37 8 36 51 13 44 41 1525 4 84 12 41 49 10 52 34 1426 73 20 7 82 5 13 47 39 14F .

Source: Datastream

IN

SYNC

Another Explanation:Institutional Development

After looking at these economicdevelopment measures, we turned tomeasures of institutional, legal, andpolitical development. In many coun-tries, after all, governments and courtsserve as mercantilist devices for divert-ing wealth to an entrenched elite.Through legislation, licensing require-ments, and nationalization, govern-ment can inhibit the growth and devel-opment of businesses. In these environ-ments, political events, or even rumorsabout political events, may cause largemarket-wide stock price swings andgenerate high levels of stock price syn-chronicity. Scholar Ray Fisman in1999 estimated that as much as 25%of the market value of manyIndonesian firms was related to politi-cal connections. This conclusion wasbased on an analysis of stock pricemovements in response to rumorsabout President Suharto’s health.

ndeed, bad government mightincrease stock price syn-chronicity through channelsthat are not directly associat-ed with economic fundamen-tals, like corporate earningsor GDP. Finance theory holds

that professional investors – risk arbi-trageurs – expend resources to uncov-er proprietary information aboutstocks and earn an acceptable returnby using that information to tradeagainst less-informed investors. Suchtrading by many risk arbitrageurs,each possessing unique proprietaryinformation, is thought to capitalizeinformation into share prices of indi-vidual companies.

But risk arbitrage of this sort maybe less economically attractive incountries that protect private propertyrights more poorly. Economic funda-mentals can be obscured by politicalfactors in many low-income countries.Political events may be hard to fore-cast in low-income nations whose gov-ernments are often relatively opaqueand erratic. And risk arbitrageurs whodo make correct predictions may not

be allowed to keep their earnings.Because firm-specific risk arbitragecould be relatively unattractive insuch countries, informed tradingmight be correspondingly thin.

If weak property rights discourageinformed risk arbitrage, they mightalso create systematic stock price fluc-

tuations. Scholars believe that aninsufficient level of informed tradingcan “create space” for noise trading –trading that reacts to political events,rumors, and the like. Once the pro-portion of noise traders in the marketrises above a critical level, it mightcrowd out risk arbitrageurs, who are

Sternbusiness 35

T A B L E 2

Per capital Gross Domestic Product and stock return synchronicity

Panel A ranks countries by per capita GDP. Panel B ranks countries by stock return synchronicitymeasured by the fraction of stocks moving together in the average week of 1995.

Panel C ranks countries by stock market synchronicity measured as the average R2 of firm-levelregressions of bi-weekly stock returns on local and U.S. market indexes in each country in 1995.

Japan 2276 33,190 United States 57.9 United States 0.021Denmark 264 27,174 Canada 58.3 Ireland 0.058Norway 138 25,336 France 59.2 Canada 0.062Germany 1232 24,343 Germany 61.1 U.K. 0.062United States 7241 24,343 Portugal 61.2 Australia 0.064Austria 139 23,861 Australia 61.4 New Zealand 0.064Sweden 264 23,861 U.K. 63.1 Portugal 0.068France 982 23,156 Denmark 63.1 France 0.075Belgium 283 21,590 New Zealand 64.6 Denmark 0.075Holland 100 20,952 Brazil 64.7 Austria 0.093Singapore 381 20,131 Holland 64.7 Holland 0.103Hong Kong 502 19,930 Belgium 65.0 Germany 0.114Canada 815 19,149 Ireland 65.7 Norway 0.119Finland 104 18,770 Pakistan 66.1 Indonesia 0.140Italy 312 18,770 Sweden 66.1 Sweden 0.142Australia 654 17,327 Austria 66.2 Finland 0.142U.K. 1628 17,154 Italy 66.6 Belgium 0.146Ireland 70 14,186 Norway 66.6 Hong Kong 0.150New Zealand 137 12,965 Japan 66.6 Brazil 0.161Spain 144 12,965 Chile 66.9 Philippines 0.164Taiwan 353 10,698 Spain 67.0 Korea 0.172Portugal 90 9,045 Indonesia 67.1 Pakistan 0.175Korea 461 7,555 South Africa 67.2 Italy 0.183Greece 248 7,332 Thailand 67.4 Czech 0.185Mexico 187 3,944 Hong Kong 67.8 India 0.189Chile 190 3,361 Philippines 68.8 Singapore 0.191Malaysia 362 3,328 Finland 68.9 Greece 0.192Brazil 398 3,134 Czech 69.1 Spain 0.192Czech 87 3,072 India 69.5 South Africa 0.197South Africa 93 2,864 Singapore 69.7 Columbia 0.209Turkey 188 2,618 Greece 69.7 Chile 0.209Poland 45 2,322 Korea 70.3 Japan 0.234Thailand 368 2,186 Peru 70.5 Thailand 0.271Peru 81 1,920 Mexico 71.2 Peru 0.288Columbia 48 1,510 Columbia 72.3 Mexico 0.290Philippines 171 880 Turkey 74.4 Turkey 0.393Indonesia 218 735 Malaysia 75.4 Taiwan 0.412China 323 455 Taiwan 76.3 Malaysia 0.429Pakistan 120 424 China 80.0 China 0.453India 467 302 Poland 82.9 Poland 0.569

Source: Datastream

Panel A Panel B Panel C1995per capitaUS$ GDP R2

%stocks movingin step (fj)listed stockscountry country country

I

more risk-averse. As a result, a stockmarket without a sufficient amountof informed trading could be charac-terized by large systematic priceswings – in other words, greater pricesynchronicity.

Measuring Good GovernmentIf including a measure of good

government in our regression analy-sis renders per capita GDP insignifi-cant, that might be evidence that alack of property rights protectionunderlies the high degree of stockprice synchronicity. To capture theextent to which a country’s politi-cians respect private property rights,we constructed a good governmentindex as the sum of three indexesfrom the International Country RiskGuide (ICR). The “corruption index”is an assessment of corruption ingovernment. Low scores indicatethat “high government officials arelikely to demand special payments”and that businesspeople may have topay bribes in order to get import andexport licenses, loans, or tax assess-ments. The “risk of expropriationindex” gauges the risk of outrightconfiscation or forced nationaliza-tion. The “repudiation of contractsby government index” measures therisk of a “modification in a contracttaking the form of a repudiation,postponement, or scaling down” dueto “budget cutbacks, indigenizationpressure, a change in government, ora change in government economicand social priorities.”

he good governmentindex, like our syn-chronicity measures,tends to be quite highfor developed countries

and quite low for emergingeconomies. The results show thatbetter protection of private propertyrights “explains” stock price syn-chronicity, so much so that its inclu-

sion renders per capita GDP insignif-icant in explaining synchronicity. Inaddition, countries with higher percapita incomes have higher goodgovernment indices. And the goodgovernment index is significantlycorrelated with market size, a find-ing that is consistent with more insti-tutionally advanced economies hav-ing markets on which more stockstrade. We also found that the goodgovernment index remains signifi-cantly negatively correlated withstock price synchronicity even aftercontrolling for market size and thestructural variables.

Haves and Have-NotsOur result leads to a conjecture

that the presence of a non-corruptgovernment that honors and respectsprivate property rights makes itattractive to conduct informed riskarbitrage which results in moreinformed stock prices. Without aninstitutional environment that hon-ors property rights, informed riskarbitrage recedes and noise tradinggenerates large systematic swing inall stock prices that is not closelyrelated to economic fundamentals.

The “good government” index,however, is a measure of institution-al development. It is unlikely to havea fine-grained incremental impact onthe matter. Indeed, our data tends togroup into two clusters: high-incomecountries with low stock price syn-chronicity and low-income countrieswith high synchronicity. Substitutingthe good government index for percapita GDP also clearly reveals twoclusters. And this clustering suggeststhe possibility of a threshold effect. Ifinstitutional development, as meas-ured by our good government index,is below a critical level, a differentregime governs stock prices, and ahigh degree of synchronicity isobserved.

We used the mean of the goodgovernment index as the dividingline in creating these subsamples,yielding a developed economy groupof 22 countries and an emergingeconomy subsample of 15 countries.Then, we tested whether our resultshold within both subsamples, ormainly describe differences betweenthe two subsamples.

Among emerging economies,stock market synchronicity is not cor-related with either the logarithm ofper capita GDP or the good govern-ment index. So overall, synchronicityin the emerging markets is generallyhigh but not much worsened by mar-ginal decline in the protection accord-ed private property. Interestingly, asin our conjecture, higher stock pricesynchronicity in emerging economiesis mainly associated with greatersystematic variation.

In the developed country sub-sample, however, the situation ismore complex. Synchronicity is mar-ginally higher when the goodgovernment index is lower. Moreimportantly, high synchronicity indeveloped countries is associatedboth with low levels of firm-specificvariation and high levels of market-wide variation. This finding moti-vates a closer look at the developedcountries to clarify the determinantsof stock price synchronicity there.

Capitalization of Firm-SpecificInformation

It is possible that a country’sinstitutions might affect the relativeamounts of firm-specific versusmarket-wide information that arecapitalized into stock prices set byrationally informed risk arbitrageurs.In other words, certain attributesmight encourage people to place theirbets on individual stocks rather thanthe market as a whole.

Our focus is that, unlike in emerg-

36 Sternbusiness

IN

SYNC

T

ing economies, lower synchronicity indeveloped economies is associatedwith greater firm-specific variation. InChina, for example, the whole markettends to move up or down dramatical-ly, and so stocks move together. Bycontrast, in the U.S., stocks tend tomove more independently of oneanother. Scholars have concluded thatmost of the variation in U.S. stockprices reflects the capitalization ofproprietary firm-specific information– people processing the data they haveabout individual companies to influ-ence the stock price.

e considered twofactors that mightmake firm-specif-ic risk arbitragemore attractive in

economies: (1) better accountingdata; and (2) better protection forpublic investors from corporateinsiders. If accounting data are moreuseful, more firm-specific publicinformation is available to allinvestors. And that may let risk arbi-trageurs make more precise predic-tions regarding firm-specific stockprice movements. A lack of respectfor the property rights of publicinvestors by controlling shareholdersmight discourage risk arbitragebased on firm-level information andhence impedes the capitalization offirm-level information in stock pricesin some developed countries.

To test these hypotheses, we ranthe numbers again using only datafrom developed countries in oursample. First, we substituted a directmeasure of the sophistication of eachcountry’s accounting standards inplace of the good government index.The measure was created by scholarsbased on 1990 data compiled by theCenter for International FinancialAnalysis and Research, Inc. Andwhen we did, it turned out that goodaccounting standards are negatively

correlated with synchronicity. Butsince the significance levels are in theneighborhood of 20%, the account-ing standards index itself is uniformlystatistically insignificant.

Next, we employed a directmeasure of the extent to which pub-lic shareholders’ property is protect-ed from appropriation by corporateinsiders – the anti-director rightsindex. This index is a scorecard ofshareholders’ rights against directorsin various countries. For such rightsto provide effective protection, acountry must have functional politi-cal and legal systems. It is thereforeplausible that the anti-director rightsindex might be relevant only incountries where the rule of law pre-vails. Notice that among countrieswith strong property rights protec-tions in general there exists quite abit of variation in protecting theproperty rights of public investorsagainst corporate insiders.

We therefore ran regressionssubstituting the anti-director rightsindex for the good governmentindex. We found that while the anti-director rights index is insignificantin the whole sample and emergingeconomy subsample, it is negativeand highly statistically significant inthe developed economy subsample.

Numbing the Invisible Hand?So what do we conclude from

these results? Yes, stock returns aremore synchronous in emergingeconomies than in developedeconomies. But while some economiccharacteristics may contribute tostock return synchronicity, they don’tentirely explain the outcome. Rather,it seems that the level of institutionaldevelopment is highly correlatedwith stock price synchronicity.

In particular, less respect for pri-vate property by government is asso-ciated with more market-wide stock

price variation, and therefore alsowith more synchronous stock pricemovements. Since these market-wideprice fluctuations are uncorrelatedwith fundamentals, we conjecturethat poor property rights protectionmight deter informed risk arbitrageand noise traders create arbitraryfluctuations. However, we wouldwelcome other possible explanations.

In developed economies, provid-ing public shareholders with strongerlegal protection against corporateinsiders is associated with lowersynchronicity. We conjecture thateconomies that protect publicinvestors’ property rights might dis-courage intercorporate income-shift-ing by controlling shareholders.Better property rights protectionthus might render risk-arbitragebased on firm-specific informationmore attractive, which leads to asyn-chronous stock price movements.

Overall, our results suggest thatstock markets in emergingeconomies may be less useful asprocessors of economic informationthan stock markets in advancedeconomies. The function of an effi-cient stock market is to processinformation, and thereby guide capi-tal towards its best economic use.But stock price movements in emerg-ing economies are mainly due toeither politically driven shifts inproperty rights or noise trading;numb invisible hands in their stockmarkets may allocate capital poorly,thereby retarding economic growth.

R A N D A L L M O R C K is Stephen A.Jarislowsky distinguished professor of finance atthe University of Alberta.

B E R N A R D Y E U N G is Abraham Krasnoffprofessor of international business and professorof economics at NYU Stern.

This article is adapted from an article in theJournal of Financial Economics, October 2000Volume 58, pp. 215-260.

Sternbusiness 37

W

C O M I N G A P A R T ,

38 Sternbusiness

C O M I N G T O G E T H E R : T H E AT & T B R E A K U P ( R O U N D T H R E E )A N D T H E R E M O N O P O L I Z AT I O N O F T E L E C O M M U N I C AT I O N S

By Nicholas Economides

In the past year, the s tock pr ices o f te lecommunicat ions companies

have fa l len sharp ly. Indeed, the shares o f te lephone companies , once

regarded as bor ing but dependable , have o f la te d isp layed the vo la t i l i -

ty more common among sof tware companies and In ternet re ta i le rs . For

example , AT&T, once the u l t imate w idows-and-orphans s tock, los t ha l f

i t s va lue in 2000. Jus t two years a f te r invest ing in cable TV assets , CEO

C. Michae l Armst rong is now p lann ing to break AT&T in to severa l par ts .

n the meantime, a series ofmergers by local telephonecompanies – the Baby Bellscreated after the 1980s govern-

ment-mandated break-up of AT&T– has led to a substantial remonopo-lization of the telecommunicationssector. Add in the cross-media AOL-TimeWarner merger, the rapidgrowth of the Internet, and thestratospheric bids for Europeanspectrum to be used for wirelesstelecommunications, and the tele-communications landscape is both

confusing and treacherous toinvestors.

Why are these once-reliablecompanies seeing their fortunes shiftdramatically? And why have somany shrewd investors been caughtunaware by the plummeting stockprices?

At the most basic level, thestocks of some telecommunicationscompanies fell sharply in 2000because the Internet- and technolo-gy-related investment bubble wasfinally pricked. As investors realized

that early expectations for Internetgrowth were much higher than justi-fied, stock valuations were appropri-ately adjusted.

But the deeper answer to whyAT&T is breaking up while its for-mer offspring are acquiring eachother lies in an understanding of thenew market dynamics created byadvances in technology, and of suc-cessive government attempts to stim-ulate competition through the enact-ment of new regulatory schemes.

First, a bit of history. For the

I

Sternbusiness 39

© A

RTI

ST(

S)/

SIS

past four decades, rapid technologi-cal change in computers and trans-mission technology has consistentlydriven production costs of telecom-munications services steeply down-wards. At the same time, the regula-tory environment, which was estab-lished to protect consumers frommonopolistic abuses, insteadkept most of the benefits oftechnological change fromreaching consumers. Fordecades, in fact, telecommuni-cations services price decreaseshave been much slower thancost decreases. In other words,companies did not exactlyrush to pass on the full bene-fits of rapid technologicalchange to their customers inthe form of lower prices.

he 1981 govern-m e n t - i m p o s e dbreakup, which cre-ated Seven Baby

Bells to provide local serviceand left parent Ma Bell(AT&T) as a long-distanceprovider, was intended to remedy thisstate of affairs. By allowing compa-nies other than AT&T to compete inthe then lucrative long-distance mar-ket, the breakup did indeed createhuge benefits to consumers. Theadvent and growth of long-distanceplayers like MCI (later WorldCom)and Sprint lowered rates, to the pointwhere residential and business cus-tomers now pay long-distance ratesthat are a small fraction of the 1981price.

However, the AT&T breakuphad another effect that was less ben-eficial for consumers. It fossilized the

monopoly status of the seven localtelephone companies that werecarved out of AT&T: Ameritech, BellAtlantic, Bell South, NYNEX, PacificBell, Southwestern Bell, and U.S.West. To be sure, prices did fall dra-matically for long-distance rates. Butthey didn’t fall quite as much as

technological change and competi-tion would imply. That’s because thelocal telephone companies – the so-called Baby Bells – were allowed tocharge “access fees” that were asmuch as ten times greater than costto let long-distance calls travel the“last mile” along their lines to theconsumer. As part of the 1981 AT&Tbreakup, local telephone monopolieswere barred from entering the long-distance market since their hugeaccess fees would have given themthe ability to undercut long-distanceprices and easily drive the long-dis-tance providers out of business.

Fifteen years after the AT&Tbreak-up, the government again triedto remedy the competitive situationthrough a sweeping action. TheTelecommunications Act of 1996 wassupposed to level the playing field byallowing competition in local mar-kets. Once that happened, the local

monopolies could compete withtheir former parent in offeringlong-distance service.

However, things haven’tquite worked out as intended.Five years after the passageof the landmark legislation,less than four percent of thelocal telecommunications mar-ket belongs to new entrants.Instead, each of the eight largelocal monopoly telephone com-panies at the AT&T 1981breakup – the seven Baby Bellsplus GTE – continues to controlmore than 96% of its market. Asimportant, these firms haveconsolidated with one another,to the point where there are nowonly four large local telecommu-

nications monopolies: Bell Atlantic,NYNEX, and GTE merged toform Verizon; Southwestern Bell,Ameritech, and Pacific Bell gottogether to form SBC Communi-cations; U.S. West was acquired byQwest, and Bell South remains inde-pendent.

The Telecommunications Act of1996 ordered the local telephonecompanies to lease parts of theirnetwork to new entrants, so thatcompetition would take place inlocal markets. But the local tele-phone companies have failed to doso. And this has led to the sub-

For decades, telecommu-

nications services price

decreases have been

much slower than cost

decreases. In other words,

companies did not exactly

rush to pass on the full

benefits of rapid techno-

logical change to their

customers in the form of

lower prices.

40 Sternbusiness

T

stantial failure of theTelecommunications Act of1996, as well as to the demise ofthe hopes of long-distancecompanies to become majorcompetitors in local markets. Itdoes appear likely that localtelephone companies such asVerizon may eventually be per-mitted to offer long-distanceservice – as Verizon already doesin New York State – and there-fore be able to sell both localand long-distance service to thesame customer. But it seems veryunlikely that long-distance compa-nies will be able to capture signifi-cant market shares in local marketsby leasing parts of the local telecom-munications companies’ networks.

his state of affairs helpsexplain AT&T’s strategicmoves of the past fewyears. Facing great diffi-

culty in entering local markets underthe terms of the TelecommunicationsAct of 1996, AT&T CEO C. MichaelArmstrong decided two years ago toget into the local telephone marketsthrough broadband cable televisionconnections. In other words, it wouldoffer local service through the high-capacity coaxial cables that run intomillions of American homes. Thus,AT&T spent billions of dollarsacquiring cable television companiesTCI and MediaOne, as well as astake in TimeWarner. With a deep-pocketed and aggressive firm on thescene, and with the ability to offertelephone service over cable wires, itappears likely there could now besubstantial competition in local tele-phone service. As an added benefit,

the cable TV connection gives AT&Tthe possibility to sell high capacityInternet service and other broadbandservices, such as interactive video.

Given this set of circumstances,it may seem contradictory for AT&Tto even consider divesting its cabletelevision and wireless assets. Hasthe company lost its nerve andvision? I think not. AT&T’s divesti-ture plan was formed in response topressure from financial markets andlarge institutional shareholders.Despite its strategic moves and largeinvestments in cable TV assets,financial markets apparently contin-ued to value AT&T stock as if it wereonly a long-distance telephone com-pany. Moreover, long-distance priceshave been under tremendous pres-sure because a great deal of networktransmission capacity was built upby several competitors over the lastthree years due to rampant Internetgrowth in the United States.

Thus, management came tobelieve that the value of AT&T as asum of the values of its independentparts (cable-broadband, wireless,business services, and residential

long distance) is indeed higherthan the present value of theunified AT&T.

But if it makes good finan-cial and strategic sense forAT&T to break itself intoseveral parts once again,what is driving the seeminglycontradictory mergers of thelocal telephone companies thatemerged from the 1981 AT&Tbreakup?

t the 1981 AT&Tbreakup, the localtelephone companies

were allowed to remain monopolistsin the local markets. The 1996Telecommunications Act attemptedto create competition in local mar-kets and failed. Presently the localtelephone companies are poised toenter the long-distance market with-out significant decreases of theirmarket shares in local markets.Looking forward to the time whenthey will be allowed to sell long-distance services, local telephonecompanies have merged to expandtheir customer base footprint andbecome stronger competitors in thenext battle among carriers that sellboth local and long-distance services.Twenty years after the governmentbroke up the longstanding MA Bellmonopoly, the remonopolization oftelecommunications is almost here.

N I C H O L A S E C O N O M I D E S isprofessor of economics at NYU Stern.

For more on these issues, see ProfessorEconomides’ Economics of Networks website,http://www.stern.nyu.edu/networks, which hasbeen ranked by The Economist as one of the topfive economics sites on the Internet.

Five years after the pas-

sage of the landmark

1996 legislation, less

than four percent of the

local telecommunica-

tions market belongs to

the new entrants.

Sternbusiness 41

T

A

42 Sternbusiness

RAGING BULLS OR

RAGING BULL:INTERNET MESSAGE BOARD ACTIVITY AND MARKET EFFICIENCY

By Robert Tumarkin

T h e I n t e r n e t i s c l e a r l y p l a y i n g a n e v e r - i n c r e a s i n g r o l e i n f i n a n c i a l

m a r k e t s a n d p e r s o n a l f i n a n c e . T h e s i x l a r g e s t I n t e r n e t b r o k e r a g e s

c u m u l a t i v e l y b o a s t e d o v e r 1 2 m i l l i o n a c c o u n t s i n 1 9 9 9 a n d g r e w s i g -

n i f i c a n t l y i n 2 0 0 0 . A n d i n v e s t o r s n o w b e n e f i t f r o m a w i d e a s s o r t m e n t

o f f i n a n c i a l i n f o r m a t i o n a v a i l a b l e o n l i n e , r a n g i n g f r o m S e c u r i t i e s

a n d E x c h a n g e C o m m i s s i o n d o c u m e n t s t o f i n a n c i a l s i t e s l i k e t h e

M o t l e y F o o l . • O n e o f t h e m o r e i n t e r e s t i n g p h e n o m e n a o n t h e w e b

h a s b e e n t h e g r o w t h o f s t o c k - r e l a t e d c h a t r o o m s a n d b u l l e t i n b o a r d s ,

w h i c h f a c i l i t a t e d i s c u s s i o n a m o n g t h o u s a n d s o f i n v e s t o r s . B u l l e t i n

b o a r d s , w h i c h a r e n o t l i v e f o r u m s , a l l o w u s e r s t o p o s t m e s s a g e s f o r

r e t r i e v a l b y o t h e r s a t a l a t e r t i m e . A t y p i c a l s i t e c o n t a i n s d i s t i n c t b u l -

l e t i n b o a r d s f o r e a c h m a r k e t s e c u r i t y t h a t u s e r s c a n d i s c u s s .

ILLUSTRATIONS BY DAVE BLACK

ecently, the press has sensa-tionalized the activity inthese forums, linking it to

egregious examples of stock-pricemanipulation. For example, inFebruary 1999, the stock price of asmall Milwaukee-based toy compa-ny, Alottafun Inc., soared 382%based on speculation started inInternet chat rooms. Despite suchexamples, the vast majority of thediscussion involves investors honest-ly expressing their opinions on secu-rities markets.

Over the past several years, ana-lysts and academics have tried to fig-ure out a means to place appropriatevalues on Internet stocks. And somehave focused on whether message-board activity has any bearing onstock prices. So, I thought it wouldbe useful to examine the relationshipamong the volume, quantity, andquality of the opinions expressed onmessage boards about individualstocks, and the movement in stockprices. In other words, I sought todetermine whether message boardactivity helps predict stock returnsand/or trading volume.

I chose to focus on a single site –RagingBull.com – in part because itis extremely popular. Between Apriland November 1999, the site’s mem-bership tripled in size to 300,000,while averaging six million daily pageviews. In addition, RagingBull.comhas some unique attributes. The sitehas categories for different tickersymbols. And it includes an “option-al disclosure” feature, which letsusers clearly indicate their opinionon the short and long-term prospectsof the stock by selecting long, short,

or no position. Similarly, users canissue strong buy, buy, hold, sell, andstrong sell ratings for both the short-term and the long-term.

Because much of the discussionon sites like RagingBull.com revolvesaround high-technology and Internetcompanies, I chose to examine thepostings and activity in a group of 73Internet service companies drawnfrom Zacks’ Internet Services sectorgroup. This group included well-known, large firms like Yahoo!(market capitalization $114.8 billionat the time) and many obscurecompanies, like Biznessonline.com(worth $53.1 million at the time).The sample had a medium marketcapitalization of $1.12 billion. For this study, I downloaded some181,633 messages posted betweenApril 17, 1999, the day when theopinion-disclosure feature was addedto RagingBull, and February 18,2000. Of that total, 43,794 (24.1%)contained short-term opinions,37,810 (20.8%) had long-term opin-

ions, and 52,812 (29.1%) included a general “voluntary disclosure.” Moststocks did not have a huge numberof postings each day. The mean stockmessage board had an average of 7.6posts daily, while the median mes-sage board had 2.5 messages perday. The maximum average numberof daily postings was 103.6 – forCMGI Inc.

The next task was to calculatethe average short-term opinion.Messages with short-term strong-buyrecommendations were assigned avalue of +2. Similarly, messages withshort-term buy, hold, sell, andstrong-sell recommendations wereassigned values of +1, 0, -1, and –2,respectively. These opinion valueswere averaged on a daily basis to cal-culate the daily average opinion foreach stock. The mean daily averageopinion was 1.56, while the medianwas 1.64 – somewhere between abuy and strong buy.

I also weighted opinions for each

R

44 Sternbusiness

RAGING BULL

TABLE 1 : DESCRIPTIVE STATISTICS FOR INVESTOR OPINION

ON EVENT DAYS

Raw Change in Weighted Opinion Adjusted Change in Weighted Opinion

Group

Average Value 27.97 3.11 -2.85 1.57 0.70 -0.76

Maximum 177.60 6.00 -0.33 2.00 1.14 -0.08

Minimum 6.00 0.25 -12.00 1.18 0.08 -1.81

StrongPositives

StrongPositives

StrongPositives

StrongPositivesNegative Negative

stock on a daily basis. Each messagewith a short-term opinion wasassigned a value according to theaforementioned scale. These opin-ions were added to calculate thedaily weighted opinion, which wasthen averaged for each stock. Themean average daily weighted opinionwas 6.09, while the median was3.44. The standard deviation of theaverage daily weighted opinion valuewas 9.49. The maximum was 56.64(CMGI Inc.) while the minimum was1.14 (TheGlobe.com).

ext, I calculated thearithmetic average andstandard deviation ofdaily returns for each

stock during the sample period. Themean arithmetic average of dailyreturn for the stocks was 0.677%and the median was 0.648%. Themaximum average daily return was2.53% (Be Free Inc.) and the mini-mum was -0.58% (FlashnetCommunications). The average stan-dard deviation of daily returns was7.59% and the median was 7.39%.The maximum standard deviationwas 13.37% (Cobalt Group) and theminimum was 4.80% (Cybercash).As might be expected, given thevolatility of the Internet sector at thetime, the average return and stan-dard deviation of returns are veryhigh compared to average values inthe stock market during the sampleperiod. The final piece of data wascompiling the average trading vol-ume for each stock.

I then subjected this data to twomethodologies: a vector autoregression(VAR) analysis and an event study.

N

8.00%

6.00%

4.00%

2.00%

0.00%

-2.00%

-4.00%-5 -4 -3 -2 -1 0 1 2 3 4 5

FIGURE 2: ABNORMAL RETURNS AROUND THE EVENT DAY(Categorized by Raw Weighted Opinion)

Strong PositivesWeak PositivesNegatives

Cum

ulat

ive

Abn

orm

al R

etur

ns

Day

7.00%

6.00%

5.00%

4.00%

3.00%

2.00%

1.00%

0.00%

-1.00%

-2.00%

-3.00%

-4.00%

FIGURE 3: ABNORMAL RETURNS AROUND THE EVENT DAY(Categorized by Adjusted Weighted Opinion)

Cum

ulat

ive

Abn

orm

al R

etur

ns

200%

150%

100%

50%

00%

-50%

-5 -4 -3 -2 -1 0 1 2 3 4 5

Cum

ulat

ive

Abn

orm

al R

etur

ns

Day

Day

FIGURE 4: ABNORMAL TRADING VOLUME AROUND THE EVENT DAY(Categorized by Raw Weighted Opinion)

VAR AnalysisI also performed a vector autore-

gression (VAR) analysis – on a stock-by-stock basis – to examine the gen-eral relationship among stockreturns, trading volume, messagepostings, and weighted opinion. Thisanalysis showed that none of thesefactors was useful in predicting stockreturns one day into the future. Theanalysis did show, however,that high trading volume daystended to precede days of hightrading volume – and that lowtrading volume days tended toprecede days of low tradingvolume. In other words, theopinion represented in bul-letin-board messages were nothelpful in predicting dailystock returns. This is consis-tent with market efficiency.

And days with high tradingvolume and positive weightedopinions are followed by dayswith greater message activity.Finally, weighted opinion isdependent on the number ofmessages and opinions postedon the previous day. Positiveopinion days tend to follow days withpositive opinions. The dependence ofweighted opinion on the number ofmessages posted is consistent with thesimple summation method used tocalculate weighted opinion and theobservation that each message boardhad positive average daily weightedopinions.

The Event StudySince it was clear that bulletin-

board opinions had no effect onstock prices in general, an eventstudy was conducted. The event

study attempted to answer the ques-tion: Does an unusual level of bul-letin-board discussion measurablyimpact stock prices? Days withunusual levels of discussion weretermed “event days” and weredefined as those with message post-ings that exceeded the previous five-day average by at least two five-daystandard deviations. (Event days in

which fewer than 10 messages wereposted were excluded from thesample.)

I examined two opinion metricsto determine the strength of opinionchanges on the event day. The rawchange in weighted opinion was cal-culated as the difference between theevent-day weighted opinion and theaverage weighted opinion over theprevious five days. The adjustedchange in weighted opinion was cal-culated as the raw change in weight-ed opinion divided by the standarddeviation of weighted opinion over

the previous five days.The event study found a total of

293 event days. Forty-seven of thesedays had opinions lower than theprevious five-day average, and wereclassified as “negatives,” while 241of the event day opinions weregreater than the previous five-dayopinion average, and were dubbed“positives.” (Five event days had

opinion equal to the previ-ous five-day average).These positives were fur-ther split in half. The“strong positives” categorycontained those with eventdays with the strongestopinion change. The“weak positives” containedthe remaining event days –those with the weakestpositive opinion change.Table 1 presents descrip-tive statistics for thechanges in opinion on theevent days.

Adjusted Returnsand AbnormalTrading Volume

Because of the high and volatilereturns in this sector, it was neces-sary to adjust the daily returns forindustry returns. So, I adjusted thereturns on my chosen 73-stock port-folio using the Philadelphia StockExchange (PSE) Internet Index. Theindustry adjusted return was definedas a stock’s daily return less thereturn on the PSE Internet Index.

Abnormal trading volume, whichis defined as the percentage change intrading volume on a given day com-pared to the average trading volume,was computed for each ticker and

46 Sternbusiness

RAGING BULL

Over the past several years,

analysts and academics

have tried to figure out a

means of placing appropri-

ate values on Internet

stocks. And some have

focused on whether mes-

sage-board activity has any

bearing on stock prices.

each day during the sampleperiod. A 20-trading-day peri-od preceding the day in ques-tion was used to calculate theaverage trading volume.

The Results So what did I find?

Figures 2, 3, and 4 show theindustry-adjusted returns andabnormal volume for a five-day period surrounding theevent day. It is apparent thatonly strong-positive-opinionevents classified using the rawchange in weighted opinionshow a statistically significantpositive drift up to the eventday (Figure 2). Returns for weak-positive-opinion events are statisti-cally flat leading up to the eventday. Negative-opinion event daysseem to show a downward drift upto the event day, but the phenome-non is not statistically significant.On the event day, both strong andweak positives have statistically sig-nificant, positive industry-adjustedreturns (Figures 2 and 3).Negative-opinion event days have aslightly negative industry-adjustedreturn, which is not statisticallysignificant. Returns for all the opin-ion groups are statistically flat afterthe event day.

Similarly, trading volume is nor-mal leading up to the event day. Buton and one day after the event day,there is a sharp increase in tradingvolume (Figure 4). The stronglypositive raw change in weighted-opinion group shows the most signif-icant increase in trading volume.For that group, approximately 160%more shares are exchanged on theevent day than on the previous 20

days. Trading volume retreats tomore normal levels approximatelytwo days past the event day.

he results show that mes-sage board activity isdefinitely linked to stockprice movements. However,

abnormal message board activitydoes not help predict future stockprice movements over a one-day orfive-day window in the future. Thisobservation is consistent with marketefficiency. On the event day, strong-positive and weak-positive event daysshowed statistically significant returnsin excess of the industry index.Therefore, abnormal message-boardactivity is coincident with abnormalstock returns. Using this methodology,however, it is impossible to determinewhether activity on the messageboards causes or is the result ofabnormal returns on the stock.

ConclusionsSo, can one predict future stock

returns and performance based solelyon message board activity? Not really.

The event study showsthat returns followingabnormal Internet mes-sage-board activity arestatistically insignifi-cant. However, statisti-cally significant positivereturns precede the dayswith strong positiveopinions and abnormalmessage board activity.Furthermore, stockreturns and message-board opinions on daysof abnormal message-board activity appear tobe related.

These results aresignificant because they counter theconventional wisdom that Internetservice stocks are valued irrational-ly. In general, message-boardactivity and opinion do not appearto impact stock prices in a signifi-cant, industry-adjusted fashion.Furthermore, abnormal message-board activity does not appearto predict significant abnormalreturns. In sum, at least in the peri-od I studied, the valuation ofInternet service stocks appeared rea-sonable and consistent with marketefficiency.

R O B E R T T U M A R K I N , a 2000 NYUStern graduate, is an analyst at MezzacappaManagement, a New York-based hedge fund.

The author wishes to acknowledge the assis-tance of Professor Robert Whitelaw.

The research discussed in this article was com-pleted with a grant from the I. GlucksmanInstitute for Research in Securities Markets atStern. The paper in its entirety will be publishedin a forthcoming issue of the Financial AnalystsJournal.

T

Sternbusiness 47

The results show thatmessage board activity isdefinitely linked to stockprice movements. However,abnormal message boardactivity does not helppredict future stock pricemovements over a one-dayor five-day window in thefuture.

endpaper By Daniel Gross

No issue dealing withfinance would be completewithout some mention ofgold. After all, the metal –chemical symbol AU –holds a special place in thepopular imagination.

There’s the Golden GateBridge in San Franciscoand the Golden Rule. TheGolden Gloves and Gold’sGym. High-quality servicesare gold-plated. In theOlympics – as in so manyother realms – gold is asymbol of excellence,wealth, and enduringvalue.

But in the realm offinance, gold doesn’t quitehave the glow it once had. In his new book, ThePower of Gold: TheHistory of An Obsession,(John Wiley), economichistorian Peter Bernsteindescribes how the metalhas been an object ofdesire and medium ofexchange from time immemorial.

Greek and Romanemperors stamped theirvisages onto gold coins. Andbecause its beauty, durability, andrarity were universally recognized,gold made an excellent medium ofexchange.

The search for gold fueled thediscovery of the New World. But theaugust metal didn’t become thebasis for a paper currency until theearly 1700s, when the Bank ofEngland began printing bank notes

that could be redeemed for theirface value in gold. Thus was bornthe gold standard.

In the 19th century, the goldstandard spread throughout Europe.And once massive gold strikes inCalifornia and the Yukon increasedthe domestic supply, the U.S. offi-cially joined the international goldstandard in 1900. (That year, 20bucks could buy a single ounce ofgold.)

Amid the crisis years of theDepression, many nations droppedthe gold standard. But the U.S.

clung fiercely to the goldstandard until 1971.Ironically, gold soared toprominence after it wasdumped as a monetarystandard. For in the1970s, the onset of infla-tion made gold seem acomfortable haven. Theprice of gold quintupledfrom $100 per ounce in1973 to $500 in 1979.And on January 21, 1980,in the dark days ofstagflation, gold soared toan unthinkable $850 perounce!

But as the economystabilized, gold fell out offashion as an investment.From its high point in1980, the price of goldhas fallen about 70% inreal terms. Back in 1980,that $850 ounce of goldcould buy one mythicshare of the Dow JonesIndustrial Average. Today,that same imaginaryshare of the Dow, hover-ing near 10,000, is worth41 ounces of gold!

Indeed, gold hasbecome, in many ways, just anotherindustrial metal – used for dentalfillings and other prosaic uses.

But for all its decline, gold isn’texactly cheap. And some of the oldclichés about gold hold true. Anyadult worth his or her weight ingold, for example, would still be amultimillionaire.

D A N I E L G R O S S is editor of STERNbusiness.

GoldenOldies

48 Sternbusiness

ILLUSTRATION BY DAVE BLACK

TRIUM is the only truly integrated, truly global Executive MBA. Participants earn a joint

degree in 16 months by attending concentrated two-week study modules at the three flag-

ship schools, plus one week each in São Paulo and Hong Kong. Distance learning and

team projects occur between class sessions. It’s visionary, focused and unique. Kind of

like you.

For more information, call +33 1 39 67 70 94 in Europe or +1 212 998 0442 in the United States.

Visit www.triumemba.org or e-mail info@triumemba.org.

It’s a big world. Somebody has to run it.

Announcing the new TRIUM EMBA — an alliance of

New York University Stern School of Business London School of Economics and Political Science

HEC Paris, Graduate Business School

Office of Public Affairs44 West Fourth Street, Suite 10-83 New York, NY 10012

NON-PROFIT ORG.U.S. POSTAGE

PaidNew York, NYpermit no.7931