The EconomistJune 1st 2019 Business 572
1sport bet: on a new type of lithium battery
developed by a plastics operation which
complemented the old paper business. Its
only big customer has been a car-sharing
scheme in Paris, Autolib’, which was run by
Bolloré before collapsing in a heap of debt
last year. The battery venture is now eyeing
electric buses. It is worth perhaps one-
tenth of the €3bn sunk into it.
Vivendi, a media behemoth in which
Bolloré saowns a 26% stake worth around
€8bn, has been a mixed bag. Mr Bolloré
took control of the group in 2014, fulfilling
a long-rumoured ambition to become a
media baron. Universal Music Group,
which is wholly owned by Vivendi but run
at arm’s length from California, has pros-
pered as streaming revenues from Lady
Gaga, u2 and its other stars have soared. Its
value is thought to have swelled from $8bn
in 2014 to $30bn—more or less Vivendi’s
market capitalisation. Vivendi has gained
just €7bn in that period, reflecting poorer
performance at other parts of the group.
Canal+ has struggled to retain French
pay-tvviewers in the age of Netflix; Havas
looks too small to compete with pr-and-
advertising rivals like wpp and Omnicom.
Mr Bolloré’s expansion into Italy, in a
bid to create a southern European content-
and-distribution giant, has been costliest.
Since 2015-16 Vivendi has tried to seize con-
trol of Mediaset, Silvio Berlusconi’s broad-
casting empire, and Telecom Italia, a mo-
bile operator. These bets have misfired:
Mediaset’s share price has fallen by a quar-
ter and Telecom Italia’s by nearly two-
thirds since Mr Bolloré built stakes in
them. A €1.1bn hit in the value of its Tele-
com Italia holdings all but wiped out Vi-
vendi’s net profits last year (other holdings
rose in value). As analysts at Morgan Stan-
ley, a bank, recently put it, “great music
story, unappealing wrapper”.
As a result of these misadventures, Bol-
loré sa has lost shareholders money in the
past five years, even as the cac 40 has
gained 40%. Such, Mr Bolloré’s defenders
might reasonably argue, is the nature of
risk-taking. Fair enough. But the wrapper
has other unappealing qualities.
Bolloré’s corporate structure would put
Byzantium to shame (see graphic on previ-
ous page). At its heart is a corporate mille-
feuille, where Mr Bolloré and his family
control a company that holds the majority
stake in a firm that in turn owns most of an-
other, and so on. Alongside this sit compa-
nies that do actual business, including Vi-
vendi. Such a model, which is not unique
in France, allows minority investors to be
brought in while a central shareholder—in
this case Mr Bolloré—maintains control.
What sets Mr Bolloré’s layer cake apart
from similar ones is that those minority
shareholders are often other companies in
it. Bolloré sa part-owns at least ten compa-
nies with direct and indirect stakes in itsown majority shareholder, Financière de
l’Odet, and its parents. When a company
pays a dividend, the money loops through a
few holding companies before some of it
returns to its own bank account. Several of
the holding companies are listed but at
least 90% owned by Bolloré entities.
Analysts attribute over a third of Bolloré
sa’s market value to shareholdings in its
parents; these parents are also worth
around €12bn in total. That does odd things
to Bolloré saaccounts. When its value falls
(like last year, when its shares lost 24%),
that of the holding companies above it dips
too. Because Bolloré sa in turn owns them,
its balance-sheet and income must be ad-
justed downwards. This then affects met-
rics used to calculate the value of its shares,
whose fall prompts a further adjustment.
Share-price rises cause upward revisions.
In 2015 Muddy Waters, a hedge fund
which specialises in arcane companies,
said the rococo empire was too complicat-
ed to be modelled in an Excel spreadsheet.
Analysts and investors with the patience to
track Bolloré sa disagree over basic things,
such as who exactly is entitled to the profits
generated by the underlying operations.Universal income
All this makes it hard to gauge the Bolloré
empire’s financial condition. It reports net
debt equal to a healthy 17% of equity. This,
though, assumes debts owed by one com-
pany in the structure can be repaid with
cash held in another part. Bolloré sa’s ac-
counts subsume all those of (much larger)
Vivendi—not just the proportion, 26%,
which it owns. Such consolidation is per-
fectly legal, for Bolloré sacontrols Vivendi.
But its claim on Vivendi’s cashflows would
be as owner of a 26% stake, not of 100%.
Informationispatchyonwhereintheorganigram debts sit (Bolloré says all the
holding companies contain cash). Pierre-
Yves Gauthier of AlphaValue, a research
house, has spoken of “looming potential
stress” in the structure. With €10.3bn in
gross debt and €4.8bn in cash at end of
2018, its position looks reasonable. Its
earnings before interest, tax, depreciation
and amortisation were just under €1bn last
year even before adding in Vivendi’s. The
accounts of Omnium, near the top of the
structure, paint a different picture. These
consolidate the group but in effect strip out
the cross-shareholding loops. They show
net debt of €5.6bn supported by equity (ex-
cluding outsiders’ stakes) of just €482m at
end 2018—gearing of over 1,000%.
Such leverage is sustainable so long as
cash keeps flowing into the system. Some
observers suspect this is why Mr Bolloré
has pursued ever-bigger companies with
ever-bigger balance-sheets, like Vivendi.
The company has been generating lots of
cash by selling assets accumulated in an
ill-fated acquisition spree initiated in the
early 2000s. Under Mr Bolloré’s leadership,
Vivendi paid fat dividends, which helped
service debts that paid for his takeover of it.
It is now considering the sale of up to half
of Universal. That would let Vivendi send
as much as $8bn to the Bolloré structure.
Critics who worry that Vivendi is run for
the benefit of Mr Bolloré and not its other
shareholders point to such machinations
as evidence. In June 2017 the purchase by
Vivendi of Bolloré businesses’ 59% stake in
Havas transferred €2.3bn in cash from Vi-
vendi’s coffers into those of Bolloré enti-
ties, which owned around 20% of Vivendi
at the time. Analysts questioned both the
deal’s rationale and price. Independent ex-
perts have decried worsening corporate go-
vernance at Vivendi (to match Bolloré sa).
Other bits of the business have attracted
the gaze of the authorities. In April 2018
preliminary criminal charges were
brought against Mr Bolloré, other Bolloré
executives, and later Bolloré sa, in relation
to corruption in Togo and Guinea. French
investigating magistrates suspect them of
sending Havas spin-doctors to work in the
countries’ election campaigns. According
to the charges, it was Bolloré businesses
that footed much of the bill, not the politi-
cians to whom the services were rendered.
(Bolloré says it paid for bona fide work car-
ried out.) Bolloré sa is said to have gained
business after the politicians won. Mr Bol-
loré, who was briefly detained for ques-
tioning at the time, may face prison time if
found guilty. Mr Bolloré, Bolloré sa and the
other executives all deny the allegations.
Many Bolloré ventures—ports, broad-
casting—depend on government goodwill.
In Africa Mr Bolloré and his businesses
have maintained close links with current
and former politicians. “He is a friend. I fa-
Capitalist in the 21st century vour friends. And so what?” Guinea’s presi-