Policy Oct 1, 2009

Pyra­mi­dal Blind Spots

Per­ils of inter­na­tion­al joint ventures

Based on the research of

Susan E. Perkins

Randall Morck

Bernard Yeung

Exec­u­tives at Cit­i­group became aware of seri­ous prob­lems with their joint ven­ture part­ner­ship in the Brazil­ian telecom­mu­ni­ca­tions indus­try in 2005. Daniel Dan­tas, the gen­er­al man­ag­er hired by Cit­i­group to man­age its invest­ment in Brasil Tele­com, was using his con­trol of the part­ner­ship for bla­tant self-deal­ing. Dan­tas alleged­ly expro­pri­at­ed more than $300 mil­lion of the joint venture’s wealth for per­son­al ben­e­fit and siphoned large sums to oth­er firms in his fam­i­ly-con­trolled busi­ness group, Grupo Oppor­tu­ni­ty. Cit­i­group had to quick­ly fig­ure out how to remove Dan­tas from con­trol of Brasil Tele­com or extri­cate itself from the part­ner­ship with­out los­ing even more val­ue. But Cit­i­group would soon dis­cov­er that nei­ther option would be eas­i­ly achieved. Through a pyra­mi­dal own­er­ship struc­ture in Brasil Tele­com and a host of oth­er relat­ed firms, Dan­tas had woven a com­plex web of con­cen­trat­ed control.

For­eign com­pa­nies often enter joint ven­ture part­ner­ships unaware of the dom­i­nat­ing cor­po­rate gov­er­nance prac­tices in devel­op­ing coun­tries, accord­ing to Susan Perkins (Assis­tant Pro­fes­sor of Man­age­ment and Orga­ni­za­tions at the Kel­logg School of Man­age­ment). In a recent work­ing paper, Perkins and coau­thors Ran­dall Mor­ck (Pro­fes­sor of Finance at the Uni­ver­si­ty of Alber­ta) and Bernard Yeung (Pro­fes­sor and Dean of the Nation­al Uni­ver­si­ty of Sin­ga­pore) out­line the poten­tial costs of not rec­og­niz­ing and account­ing for these dynam­ics in the design of a joint venture.

Under­stand­ing Pyra­mids
In the Unit­ed States and the Unit­ed King­dom, wide­ly held, stand-alone firms are the most com­mon type of cor­po­rate struc­ture. Own­er­ship of wide­ly held firms is spread among so many share­hold­ers that no sin­gle par­ty has effec­tive con­trol. How­ev­er, in many devel­op­ing coun­tries, par­tic­u­lar­ly in Latin Amer­i­ca, busi­ness groups con­trolled by wealthy fam­i­lies are the norm (Table 1). The groups often have inter­cor­po­rate hold­ings, often arranged in a pyra­mi­dal struc­ture, that max­i­mize the over­all wealth of the group, some­times at the expense of indi­vid­ual firms. For­eign joint ven­ture part­ners who assume that their domes­tic part­ner firms always act to max­i­mize share­hold­er val­ue can be in for a rude awak­en­ing, as Cit­i­group dis­cov­ered. Using a wealth of data obtained on Brazil­ian tele­com joint ven­tures, as well as inter­views with senior exec­u­tives at key multi­na­tion­al sub­sidiaries in Brazil, Perkins found that these pyra­mi­dal blind spots” have led to ele­vat­ed fail­ure rates for joint ven­tures between pyra­mi­dal group firms and part­ners from coun­tries with lit­tle expe­ri­ence with this type of group.

Table 1: Fam­i­ly Con­trol Indices
Fam­i­ly con­trol indices are based on the largest ten pri­vate sec­tor busi­ness enti­ties (free­stand­ing firms or busi­ness groups). The data are frac­tions of domes­tic enti­ties (labor weight­ed) con­trolled by fam­i­lies in 1996.

Source: Kathy Fogel (2006). Oli­garchic Fam­i­ly Con­trol, Social Eco­nom­ic Out­comes, and the Qual­i­ty of Gov­ern­ment,” Jour­nal of Inter­na­tion­al Busi­ness Stud­ies, 37(5): 578 – 625.

Set­ting the terms of the deal is even more impor­tant in devel­op­ing coun­tries than in the U.S. or UK,” says Perkins. In the Unit­ed States, trust in cor­po­rate agents and pro­fes­sion­al man­agers comes from strong mar­ket insti­tu­tions with cred­i­ble enforce­ment mech­a­nisms. Effi­cient mar­kets are fueled by the free flow of infor­ma­tion, trans­paren­cy, and the knowl­edge that the legal sys­tem is equipped to penal­ize outliers.

In Brazil, trust is more rela­tion­al­ly based, estab­lished out­side of mar­kets and the legal sys­tem. Those in con­trol of pyra­mi­dal firms are there­fore more like­ly to estab­lish pro­vi­sions with­in the joint ven­ture agree­ment that ensure puni­tive dam­ages if trust is breeched.

For­eign joint ven­tures have def­i­nite ben­e­fits and, in some cas­es, are required to enter local mar­kets. Many for­eign com­pa­nies use joint ven­ture part­ner­ships to mit­i­gate the ldquo;liability of for­eign­ness” — their unfa­mil­iar­i­ty with local rules, cul­ture, and busi­ness prac­tices. How­ev­er, if a com­pa­ny enters into a part­ner­ship unaware of the opaque local own­er­ship and con­trol schemes that poten­tial­ly work against its inter­ests, it could be set­ting itself up for dis­ap­point­ment and failure.

Pyra­mi­dal busi­ness groups are set up in a tiered struc­ture, with firms at the apex hold­ing par­tial con­trol of low­er firms, which in turn hold par­tial con­trol of still low­er firms, and so on. Pyra­mi­dal groups lever­age the owner’s wealth into a vast amount of con­trolled cor­po­rate assets with only lim­it­ed par­tic­i­pa­tion in any one unit, espe­cial­ly at the low­er tiers. The fam­i­ly at the top places much more empha­sis on con­trol than on direct own­er­ship, accom­plish­ing this by hold­ing super-vot­ing shares, char­ters lim­it­ing share­hold­er vot­ing rights, and char­ters that allow them to appoint direc­tors and managers.

Accord­ing to Perkins, the biggest risk for a for­eign firm unaware of these dynam­ics when enter­ing a joint ven­ture involves asset and prof­it shifting.

A part­ner firm may be at the bot­tom of a pyra­mid, six lev­els below the apex, but the con­trol­ling own­er at the top has access to cash flow and cor­po­rate assets all the way down in a struc­ture that is rarely trans­par­ent,” explains Perkins. This can result in tun­nel­ing,’ shift­ing mon­ey to dif­fer­ent parts of the struc­ture, or sim­ple expro­pri­a­tion, buy­ing jets, pay­ing for coun­try club mem­ber­ships, unre­lat­ed legal fees, etc.”

Sur­vival and Fail­ure
Perkins and her col­leagues exam­ined the entries of nine­ty-six multi­na­tion­al sub­sidiaries into the Brazil­ian telecom­mu­ni­ca­tions indus­try between 1997 and 2004. They assessed the per­for­mance of part­ner­ships of dif­fer­ent par­ent firm com­bi­na­tions, with each par­ent clas­si­fied as free­stand­ing or a mem­ber of a pyra­mi­dal group. These were fur­ther divid­ed into Brazil­ian and non-Brazil­ian firms.

The results showed that part­ner­ships between free­stand­ing firms and firms that are part of pyra­mi­dal struc­tures are much more like­ly to fail (a 27 per­cent fail­ure rate), and part­ner­ships between firms that are both part of pyra­mi­dal struc­tures have the high­est rates of sur­vival (only an 8 per­cent fail­ure rate) (Fig­ure 1). Perkins explained this dual-pyra­mi­dal advan­tage by sug­gest­ing that both par­ties bet­ter under­stand their partner’s gov­er­nance and agency moti­va­tions, and also employ coun­ter­mea­sures to neu­tral­ize opportunism.

Fig­ure 1: Fail­ure rates for joint ven­ture par­ent firm combinations

Hav­ing a Brazil­ian part­ner also made a dif­fer­ence. Non-pyra­mi­dal firms had sig­nif­i­cant­ly low­er fail­ure rates (22 per­cent) when they part­nered with Brazil­ian pyra­mi­dal firms than when they part­nered with for­eign pyra­mi­dal firms (44 percent).

Fur­ther, the researchers assessed how firms’ famil­iar­i­ty with pyra­mi­dal busi­ness groups impact­ed per­for­mance. Using an index that mea­sures the degree of fam­i­ly con­trol in the top ten busi­ness enti­ties in sev­er­al coun­tries, the team found that firms with the low­est fail­ure rate (2 per­cent) were from coun­tries where at least 75 per­cent of firms were part of pyra­mi­dal groups. The high­est fail­ure rate (19 per­cent) was among firms from coun­tries where less than 25 per­cent of firms are part of pyra­mi­dal groups (Fig­ure 2).

Fig­ure 2: Joint Ven­ture Fail­ure Rates

Source: Kathy Fogel (2006). Oli­garchic Fam­i­ly Con­trol, Social Eco­nom­ic Out­comes, and the Qual­i­ty of Gov­ern­ment,” Jour­nal of Inter­na­tion­al Busi­ness Stud­ies, 37(5): 578 – 625.

If a com­pa­ny want­ed to pre­dict its chances for suc­cess in a giv­en ven­ture, those find­ings are real­ly a barom­e­ter for who should be most con­cerned. If your firm is from a coun­try with lit­tle under­stand­ing of how pyra­mi­dal group struc­tures oper­ate, you have a much greater chance of being caught in a pyra­mi­dal trap,” says Perkins.

But what about the high sur­vival rate for part­ner­ships when both firms are from pyra­mi­dal groups? In such ven­tures, man­agers from both par­ent firms often have a bet­ter under­stand­ing of the risks and chal­lenges. They build safe­guards to pre­vent one firm from under­min­ing the other.

One way of doing this is to arrange mul­ti­ple points of com­pe­ti­tion and inter­ac­tion,” says Perkins. This gives both par­ent firms incen­tives to be trust­wor­thy part­ners by giv­ing each part­ner mul­ti­ple means to retal­i­ate if the oth­er acts opportunistically.”

Perkins found three key fea­tures of suc­cess­ful part­ner­ships between mul­ti­ple pyra­mid firms:

1) Cash flow and vot­ing rights are more com­mon­ly split even­ly.
2) Deci­sion-mak­ing con­trol is assigned to top man­agers of each par­ent firm, prop­er­ty by prop­er­ty, not allo­cat­ed over­all to one par­ent or the oth­er.
3) Each firm takes some own­er­ship of high-lev­el firms in the other’s pyra­mi­dal group.

This strat­e­gy gives ammu­ni­tion to each firm to strike back if the oth­er acts unscrupu­lous­ly. In the case of Cit­i­group, it relin­quished too much con­trol to Dan­tas, assum­ing he would act in the inter­est of the part­ner­ship. When it became obvi­ous that Dan­tas was behav­ing in his own self-inter­est, Cit­i­group had lit­tle means to rein him in and con­trol its losses.

Lessons Learned
Perkins observes that although many U.S. cor­po­ra­tions doing busi­ness in Brazil have begun to set up deals dif­fer­ent­ly, those that have not often end up in unre­solved legal battles.

Senior exec­u­tives still have dif­fi­cul­ties iden­ti­fy­ing the blind spots,” she says. I fre­quent­ly hear exec­u­tives say, We are still wait­ing for our case to be adju­di­cat­ed.’ They don’t real­ize that in many cas­es these prac­tices are actu­al­ly legal­ly enshrined in cor­po­rate gov­er­nance codes that spec­i­fy that direc­tors act in the inter­est of their busi­ness group.”

Mar­ket reg­u­la­tors in Brazil rec­og­nize that these gov­er­nance issues are prob­lem­at­ic for out­side investors and have reformed the rules of the game” from with­in. In 2000 the Brazil­ian stock exchange Boves­pa almost implod­ed when the mar­ket became illiq­uid. In 2001 major reforms were estab­lished, with stronger minor­i­ty share­hold­er rights and an effort to estab­lish one vote, one share” rules. The results so far are mixed, but Perkins believes that much of the cur­rent mar­ket liq­uid­i­ty in Brazil is com­ing from firms with strong, trans­par­ent gov­er­nance struc­tures. She plans to study those phe­nom­e­na in sub­se­quent research efforts.

We want to exam­ine if these reforms real­ly have bro­ken the sta­tus quo, or if the large firms are just fig­ur­ing out how to redress them­selves. Some Brazil­ian exec­u­tives see this as a prob­lem affect­ing the flow of invest­ment and one that stymies growth, and oth­ers do not want the sys­tem exposed,” Perkins says. The strength of reforms in the Novo Mer­ca­do, or the New Mar­ket, is indica­tive of how viable Brazil is in the tran­si­tion from an emerg­ing to a devel­oped market.”

In anoth­er sign of change, Daniel Dan­tas, the financier who bilked Cit­i­group for $300 mil­lion, was con­vict­ed of bribery in Decem­ber 2008 and may face ten years in prison. Perkins calls the con­vic­tion a sig­nal to the world that this type of behav­ior will not be tol­er­at­ed and a sign that the coun­try is mov­ing forward.

Featured Faculty

Susan E. Perkins

Member of the Department of Management & Organizations from 2006-2016

About the Writer

Josh McDaniel is a freelance writer living in western Colorado.

About the Research

Perkins, Susan, Randall Morck, and Bernard Yeung (2008). “Innocents Abroad: The Hazards of International Joint Ventures with Pyramidal Group Firms,” working paper, available at SSRN: http://ssrn.com/abstract=1097265.

Read the original

Editor’s Picks


Pod­cast: How to Be a Great Mentor

Plus, some valu­able career advice that applies to just about everyone.


A New Way to Per­suade Kids to Drink More Water and Less Soda

Get­ting chil­dren to make healthy choic­es is tricky — and the wrong mes­sage can backfire.


How Can Social Sci­ence Become More Solutions-Oriented?

A con­ver­sa­tion between researchers at Kel­logg and Microsoft explores how behav­ioral sci­ence can best be applied.


Buy­ing a Com­pa­ny for Its Tal­ent? Beware of Hid­den Legal Risks.

Acquir­ing anoth­er firm’s trade secrets — even unin­ten­tion­al­ly — could prove costly.


Take 5: Tips for Widen­ing — and Improv­ing — Your Can­di­date Pool

Com­mon bias­es can cause com­pa­nies to over­look a wealth of top talent.


Every­one Wants Phar­ma­ceu­ti­cal Break­throughs. What Dri­ves Drug Com­pa­nies to Pur­sue Them?

A new study sug­gests that firms are at their most inno­v­a­tive after a finan­cial windfall.


4 Key Steps to Prepar­ing for a Busi­ness Presentation

Don’t let a lack of prep work sab­o­tage your great ideas.


Video: How Open Lines of Com­mu­ni­ca­tion Can Improve Health­care Outcomes

Train­ing physi­cians to be bet­ter com­mu­ni­ca­tors builds trust with patients and their loved ones.


Here’s a Bet­ter Way to Sched­ule Surgeries

A new tool could dri­ve sav­ings of 20 per­cent while still keep­ing sur­geons happy.

Politics & Elections

Why Eco­nom­ic Crises Trig­ger Polit­i­cal Turnover in Some Coun­tries but Not Others

The fall­out can hinge on how much a country’s peo­ple trust each other.


Build­ing Strong Brands: The Inside Scoop on Brand­ing in the Real World

Tim Calkins’s blog draws lessons from brand mis­steps and triumphs.


How the Cof­fee Indus­try Is Build­ing a Sus­tain­able Sup­ply Chain in an Unsta­ble Region

Three experts dis­cuss the chal­lenges and rewards of sourc­ing cof­fee from the Demo­c­ra­t­ic Repub­lic of Congo.