Content uploaded by Emilio Barucci
Author content
All content in this area was uploaded by Emilio Barucci on Apr 13, 2014
Content may be subject to copyright.
341
Privatizations in Italy: Planning or improvisation?
Efficiency gains or rents?
Emilio Barucci* and Federico Pierobon**
* Politecnico di Milano.
** Universitat Pompeu Fabra.
Fifteen years have passed since the Amato government launched
privatizations, a process that has a major impact on the Italian economy.
The objectives were manifold and ambitious, extending well beyond the
procurement of funds with which to service the public debt; they included
revamping the State’s role in the economy in order to revive Italy’s overall
competitiveness, equipping the country with an adequate financial market
and transforming the public enterprises into widely held companies owned
by small savers. The article focuses in particular on a sample of 51
companies, finding that the efficiency gains from privatization were
greatest in the financial sector; the impact on capital structure and
investment policies was not significant. The widespread increase in
dividend payout ratios, especially in the “sheltered” sectors, raises some
questions about the possibly conflicting interests of the State in its role as
both regulator and shareholder of public utilities
1. Introduction
Fifteen years have passed since the launch of privatizations in Italy.
The process, vast in scale, has radically altered the country’s economy.
Under the gun of the crisis of 1992, the Amato government and its
successors promoted a “plan” for the sale of public enterprises
designed to transform the State’s role in the economy from “producer”
to “regulator”. The plan’s objectives were ambitious, for it sought to
recast the structure of the Italian economy, transcending the mixed
public-private model in place since the end of the Second World War,
and to revitalize the State-controlled corporations, many of which were
in disastrous shape. In this paper we shall pursue two lines of inquiry,
342
Emilio Barucci and Federico Pierobon
first examining whether the privatizations carried out were actually
guided by an economic policy design and then testing the efficiency
gains in the management of the privatized companies.
The privatizations spanned the years from the early 1990s to the
middle of the present decade, a difficult time in the political and
economic life of Italy, in which no fewer than eight governments
succeeded one another. Moreover, during this lengthy period a
substantial portion of the asset disposals were made by local
authorities and banking foundations whose political positions did not
necessarily coincide with those of the central government. Considering
these complicating factors, which make it difficult to trace the entire
privatization process back to one overarching, cohesive design, in our
analysis we refer to the objectives set out in the “Programme for the
Reorganization of IRI, ENI, ENEL, IMI and INI”, approved by the
Council of Ministers on 20 December 1992.
The programme framed the main objectives of privatization as
follows:
• to reduce the debt of the State holding companies and recapitalize
them;
• to improve the operating companies’ operating efficiency;
• to contribute to the reduction of the public debt;
• to lay the basis for widely distributed shareholding and stability in
the ownership structure of the privatized enterprises;
• to create a dozen or so industrial groups large enough to compete
at European level;
• to clearly define the public function/activities reserved to the
companies slated for privatization and to formulate a pricing policy
for public services.
Our second line of inquiry – the efficiency gains actually achieved by
the privatized companies – touches on the key argument in favour of
privatization. Vickers and Yarrow (1991) and Shleifer (1998), among
others, contend that private ownership is superior to public ownership,
by virtue of more effective governance mechanisms. It follows that
after privatization we should expect to see more efficient management
of the company together with other structural changes, such as
reduction in staff and increased investment.
1
The sample’s representativeness can be appreciated by comparing this figure with the total
receipts shown in the Reports to Parliament on privatizations, which put the privatizations by the
Treasury and IRI at €92,226 million and €43,964 million respectively, for a total of €140,190
million. After subtracting the € 9,478 million of “second-level privatizations”, which did not
require the Treasury Ministry’s approval and for which detailed data were not available, the figure
falls to €130.7 billion. The €122.2 billion used in our analysis therefore represents 93.5 per cent of
the total.
343
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
The paper is organized as follows. In Section 1 we present some
data on the programme of privatizations in Italy. In Section 2 we
assess the correspondence of the programme’s implementation to the
indications of economic theory and to its stated objectives. In Section
3, lastly, we take up the question of efficiency gains in the privatized
companies.
2. The scale of the privatizations
Since 1992 Italy has experienced a process of privatizations on a scale
that not only is unprecedented in the nation’s history but also
compares favourably with developments in other countries. In the
twenty years from 1979 to 1999, Italy ranks second in the value of
privatizations with $122 billion, behind the United Kingdom ($165
billion) but ahead of France ($71 billion), Germany ($63 billion) and
Spain ($62 billion). But narrowing our focus to the period from 1992 to
2000, Italy ranks first in total amount of disposals (see Privatization
Barometer, 2005).
For our study we gathered data on the main privatizations carried
out between 1992 and 2005 by the Ministry of the Treasury, ENI, ENEL,
IRI, local authorities and banking foundations. The transactions
numbered 125 and involved 83 companies plus 5 business segments or
real-estate divisions. Excluding the debt transferred, the privatizations
that we surveyed amounted to €140,434 million, of which €122,236
million attributable to the transactions by IRI and the Treasury.
1
The
breakdown according to privatizing entity shows the Treasury
Ministry with 30 transactions totaling €94,365 million, followed by IRI
(52 transactions, € 27,839 million), ENI (17 transactions, € 3,586
million), ENEL (3 transactions, € 3,585 million), CONSAP (2
transactions, € 430 million) and EFIM (4 transactions, €392 million).
344
Emilio Barucci and Federico Pierobon
We also have 14 privatizations by local authorities, counting only
those that involved IPOs, i.e. the most significant operations, which
totaled €3,390 million. The transactions covered include 19 involving
the financial sector (banks and insurance companies), for a total of
€22,273 million; four of these, amounting to €5,106 million, were by
banking foundations.
Aggregate data on the transactions in our sample are reported in
Table 1, where the transactions are grouped by year (see the
appendix). In terms of receipts, the peak years were 1997 to 1999;
the two years 2003 and 2004 also registered high figures, but in this
case the privatizations were merely formal, since the purchaser was
Cassa Depositi e Prestiti, still solidly controlled by the Treasury even
today.
There is a pattern of increases coinciding with critical periods for the
public finances. These periods were also characterized by the cogency
of European Union constraints, a major factor dictating the tempo of
the State’s withdrawal from the economy. Among the reasons for
undertaking the privatization plan in 1992 was the impact on public
enterprises of the revision of the Treaty of Rome (Articles 92 and 93),
under which outlays by the Treasury to cover ongoing losses of public
enterprises were now classified as State aid and as such harmful to
competition. In the peak period 1997-1999, privatizations were a
prominent feature among the measures taken to adjust the public
finances in time for Italy to qualify immediately for the euro. Lastly, in
the early years of the current decade the Berlusconi government
disposed of State shareholdings in order to keep the general
government deficit under the ceiling of 3 per cent of GDP set by the
European Stability Pact.
In Figure 1 we report the value of privatizations year by year. The bar
graph shows the value of public offerings and private placements,
respectively. The line graph indicates the percentage of the value of the
transactions upon whose completion the State no longer controlled the
company. The correlation between privatization receipts and the
general government borrowing requirement appears clear. The
evidence allows us to remark: i) that no clear trend developed over
time in the preferred method of sale, and ii) that transactions involving
the State’s exit as a shareholder tended to become rarer (the peak in
345
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
2002 coincided with the disposal of the residual interest in Telecom
Italia after the State had already lost control of the company).
3. Planning or improvisation?
The debate on the privatizations is rife with misconceptions and
commonplaces. Here we investigate how well they actually
corresponded to the 1992 reorganization programme and the dictates of
economic theory. In particular, we address five points: the privatizations
and the public debt, total versus partial privatizations, methods of sale,
ownership structures, and the development of the financial markets.
3.1 The privatizations and the public debt.
Gauging the privatizations’
impact on the State budget is very complicated, since expenditures
consist of contributions of capital and debt-servicing costs while
revenues include not only the proceeds from the asset sales but also
the dividends the State collects as shareholder. In the Italian
institutional framework, the Treasury may use dividends to finance
current expenditure but has no discretion in employing privatization
receipts, which are earmarked for the Sinking Fund for the Redemption
of Government Securities established by Law 432/1993.
As of the end of 2005 the Fund had made disbursements totaling
€120 billion, mainly to buy back public debt securities (€106 billion).
The contribution of privatizations in the strict sense amounted to €103
billion. Between 1994 and 2005 these resources went to extinguishing
an annual average of 0.91 per cent of the government securities in
circulation, or 0.77 per cent of the outstanding public debt. The role of
privatizations was especially large in the years 1997-2001. The total
disbursements of the Sinking Fund amount to 7.6 per cent of the
present public debt and 9.3 per cent of the present stock of
government securities.
3.2 Total versus partial privatization.
Many of the 125 transactions
actually involved disposals of shareholdings without total transfer of
control to the private sector. Genuine privatizations – transactions at
the end of which the State or public entity no longer had control –
2
The sale of equity interests without control transfer is not unique to the Italian experience (see
Jones et al., 1999, and Bortolotti and Faccio, 2004), but its incidence is higher than in other
countries. As a percentage of GDP, total privatizations and those involving transfer of control
amounted to 13.4 and 11.9 per cent respectively in the United Kingdom, 7.8 and 5.2 per cent in
France, 5.4 and 1.3 per cent in Germany, 8.3 and 8.1 per cent in Spain, and 12.3 and 3.9 per cent in
Italy (see Confindustria, 2004).
346
Emilio Barucci and Federico Pierobon
numbered 72 and brought in €41,673 million, or 30 per cent of total
receipts. Control was transferred in 58 companies out of 84. A
comparison of the two figures shows that control transfers mainly
involved smaller companies.
2
Was there an economic logic behind the
decisions whether or not to transfer control?
Table 1 shows that genuine privatizations were more frequent in the
early years of the process. This contradicts the assumptions of political
economy about the credibility and consistency of privatization plans
when there is the political will to privatize (see Perotti, 1995): with the
passage of time, we would expect to see a progressive increase in
privatization receipts and control transfers. Two observations serve to
explain the contrasting dynamic of privatizations in Italy. First, in the
more recent years the easing of the constraints on the public finances
made it less compelling to give up control. Second, the firms privatized
early on were relatively small, while most of the transactions in the later
years involved large companies. Plausibly, in the more recent years it
was easier to procure resources for the public finances without
relinquishing control.
To discover whether there was an economic logic behind the decisions
on control transfer, in Barucci and Pierobon (2007) we estimate a probit
model in which the endogenous dummy takes a value of 1 if control is
transferred and zero otherwise. No budgetary variable proves to be
statistically significant, so there is no evidence of an underlying
economic logic. The variables that do turn out to be statistically
significant are the year of privatization and the payout ratio: control
transfer was typical of the early years of privatization and the State
decided to retain control of the companies able to pay high dividends.
3.3 Methods of sale.
Designing a privatization plan implies selection of
effective methods of sale. A first fork in the road is: public offering or
private sale? If the first option is chosen, three aspects come into play:
347
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
determination of the offering price, allocation of the offering between
the different categories of investor, and the possible formation of a core
of controlling shareholders.
Economic theory provides some suggestions concerning the choice
between the two methods. Smaller companies in need of restructuring
(low profitability, critical financial situation) should be sold through
private negotiations to a single buyer so as to facilitate rapid
rehabilitation, while healthy companies, demand for which is likely to
be strong, can be sold through the stock market (for an empirical
confirmation, see Megginson et al., 2004). With the passage of time, the
stock market should become the vehicle of choice.
Notwithstanding intense debate, Italy’s legislation on privatizations
does not assign preference to either method. In our dataset the public
offerings numbered 51 and brought in € 100,761 million, or 73 per cent
of total proceeds. The corresponding proportion at world level is 63
per cent and for the countries of Western Europe it is 66 per cent
(Megginson et al., 2004). By international standards, therefore, the
Italian experience is marked by somewhat greater reliance on the stock
market.
A decline in transfers of controlling stakes in the course of our
observation period was accompanied by an increase in the proportion
of public offerings. Privatization commenced with small companies,
initially by way of private agreements involving transfer of control.
Subsequently, the larger transactions were made by through public
offerings. The analysis in Barucci and Pierobon (2007) shows that the
use made of the stock market was consistent with the suggestions of
economic theory and political economy. Large companies earning
good profits were sold on the stock market; public offerings were less
common at the inception of the process and for companies in
unregulated sectors.
Focusing more narrowly on the sales made through the stock market,
we can evaluate the choices made with regard to the allocation of
securities between retail and institutional investors and the will to
bolster the privatization plan’s credibility. One of the plan’s goals was
to strengthen institutional investors in Italy, an objective that conflicted
with that of wider share ownership. Bearing in mind that in all the IPOs
made in the period 1995-2004 institutional investors were allotted 62
3
The underpricing of an IPO is calculated as the security’s return on the first day it is listed (ratio to
the offering price of the difference between the first-day closing price and the offering price).
348
Emilio Barucci and Federico Pierobon
per cent of the shares while in the privatization offerings they were
assigned 56 per cent, we can conclude that some special consideration
was paid to small savers.
All the documents on privatizations, and especially the Reorganization
Programme, laid great stress on the need to convince the international
financial markets and small savers of the Italian government’s good
intentions. In order to make the plan credible, the Amato government
explicitly decided to sell off the shares at a discount. The thinking
behind this was that a relatively low offering price would attract both
(international) institutional investors and small savers and would also
betoken the government’s commitment to stay the course: selling to a
large number of investors raises the political cost of disadvantageous
rate-setting for pubic utilities or of their re-privatization, making these
contingencies extremely remote.
The discounting of shares can be gauged by referring to the
underpricing of the IPOs and subsequent tranches —the amount of
“money left on the table” by the issuer.
3
Appreciable underpricing
means that the issuer has “forgone” some procurable resources, since
the market valuation (the closing price on the first day of trading) was
higher. From the State’s vantage point, there is a trade-off between the
desire to encourage wider share ownership and gain credibility on the
one hand and the need to raise cash on the other.
From the analysis in Barucci (2006) we obtained data on the IPOs
on the Italian stock market after 1995, for a total of 158 transactions,
22 of them corresponding to privatizations. To analyze underpricing,
we added the public offerings by the Treasury and local authorities of
already listed shares (24 transactions). The privatization IPOs were
much less underpriced than that of the totality of IPOs (a mean of 5
per cent as against 9.6 per cent). In the benchmark study by Jones et
al. (1999), performed on an international sample of privatized
companies, mean underpricing was 29 per cent. The evidence
indicates, therefore, that Italian public entities (especially those under
central government control) sought only in part to capture the
interest of small investors and gain credibility with international
4
The companies are AdR, Acciaerie Speciali Terni, Autogrill, Autostrade, Banca di Roma, Banca
Nazionale del Lavoro, Banco di Napoli, Cirio Bertoli De Rica, Cofiri, Comit, Credito Italiano, ENEL,
ENI, Fincantieri, Finmeccanica, GS, Laminati Piani, IMI, INA, Istituto Bancario San Paolo, Nuovo
Pignone, SEAT, Savio Macchine Tessili, SIV and Telecom Italia. These privatizations amounted to
€109.6 billion.
349
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
institutional investors by pricing the offerings attractively. Certainly, it
cannot be said that the publicly owned companies were sold off at
knockdown prices.
3.4 Ownership structures.
In proceeding with privatizations, three
methods of sale were originally considered: public offerings, public
auctions with bidders selected to create a stable core of investors, and
private placements. They were intended to realize three different
models of ownership, respectively the widely-held listed company, the
listed company with a core of controlling shareholders, and the
privately-held company.
In the 1990s ownership structures and privatizations were the subject
of intense debate. Those who saw privatization as a means of injecting
democracy into the market for corporate control through wider share
ownership were pitted against the advocates of making deals with
Italy’s capitalist establishment in order to create stable ownership
arrangements.
Examining the evolution of the ownership structure of the 25 main
privatized corporations,
4
we find that none has become a “pure”
widely-held company in which control rests with the market, but also
that the attempts to condition ownership arrangements in lasting
fashion have not succeeded. None of the ten companies sold to
groups of investors in private placements has had a stable ownership
structure. In four of these companies one of the original investors has
bought out the others, in five control has been transferred to third
parties, and in the case of Fincantieri the equity stakes are now back in
the hands of the State. As to the companies in which a private
placement was flanked by a public offering, membership of the core
of controlling shareholders has remained stable in only four of the
eight. In three of the four remaining cases outsiders have taken
control, while in the fourth one member of the original core has
bought out the others.
350
Emilio Barucci and Federico Pierobon
In the final analysis, the strategy of constructing hard cores of
investors has substantially failed. The most common outcome is private
control, regardless of the method of sale used. Banks (in which the
banking foundations have large stakes) are the sole exception. Table 2
shows the outcomes today in terms of ownership structure of the
privatizations reported in Table 1.
3.5 Development of the financial markets.
The objectives of the
privatizations included deepening the financial market, diffusing share
ownership, bringing in institutional investors and attracting capital
from abroad. What evidence do we have that these goals were
attained?
The privatized companies play a hefty role in the Italian stock market.
At the end of 2006, 41 of the 290 companies listed on Borsa Italiana
were privatized companies (including banks formerly owned by
banking foundations). Similarly, 21 of the 40 companies composing the
S&P/MIB index had been privatized, including 8 of the top 10. At the
end of 2006, privatized companies accounted for about 60 per cent of
the market’s total capitalization. Of the 161 companies that were first
listed on the stock exchange between 1995 and 2004 (many of them
“new economy” enterprises), 22 were privatized companies and they
accounted for 68 per cent of the total funds raised.
Table 3 presents a breakdown of stock market capitalization by type
of investor for the years 1995-2003. The incidence in terms of share
ownership by households rose considerably over the period as a
whole. Although it declined somewhat after peaking during the “new
economy years”, at the end of 2003 Italian households’ share of the
market was still far greater than in 1995. The portion of the market held
by non-residents rose through 1998 and then fell back to lower levels,
while that owned by institutional investors remained relatively stable
though trending downwards.
Thus, two of the three objectives were at least partly achieved. Direct
ownership of shares by Italian households and the role of foreign
investors did expand over the nine years. However, only a minority of
the privatized companies passed into the hands of foreign investors,
since most of the companies were manufacturing firms disposed of
through private placements (for example, Nuovo Pignone, SIV,
351
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
Enichem Augusta, and Tubi Ghisa). Considering privatizations carried
out through public offerings, at the end of 2006 Italy ranked first in
Europe in the relative importance of domestic investors and hence last
in that of foreign investors (13.2 per cent).
4. Efficiency gains in the operations of the privatized companies
The empirical literature on improvements in the operating efficiency
of privatized companies generally compares pre- and post-privatization
performance and operations in order to isolate the “effects” of
privatization. We too take this approach, tracking the performance of
selected financial statement indicators to identify the changes produced
by the privatization. After presenting the methodologies and our sample
of 51 privatized companies, we analyze the complete sample and then
two different partitions of the sample (financial versus manufacturing
companies, companies in “sheltered” sectors versus companies operating
in competitive markets).
4.1 Database and methodology.
To determine whether privatization is
associated with structural changes in companies’ operating
efficiency/profitability, we use two different methodologies. We
begin by performing a univariate (before-after) statistical analysis of
the significance of the difference between the median values of
some financial statement indicators in the three years before and the
three years after privatization, adopting the standard approach
proposed in Megginson, Nash and Van Randenborgh (1994). The
weaknesses of this approach lie in its narrow “window” of seven
years and in the fact that the before-after deviations of the indicators
may derive from factors external to the company, for example trends
affecting the company’s sector or the national economy as a whole.
To overcome these shortcomings and test the robustness of the
results, we then consider regression models for panel data, with an
average of ten observations per company, which include cyclical
factors among the exogenous variables. Our sample of 51 companies
from among those privatized between 1992 and 2001 consists of
those with a value of at least € 20 million for which it was possible to
5
Compared with other studies of privatizations in Italy, our sample is particularly large. Its overlap
with that analyzed in Mediobanca (2001) is limited to 23 firms, including all 20 studied by Goldstein
(2003). Unlike those works, our study includes the large groups still controlled by the State (ENI,
ENEL, STMicroelectronics, Finmeccanica), financial corporations and municipally owned
corporations. In the case of divestment by private sale, the moment of privatization is defined as
that of the transfer of control, while in the case of sale of part of a company or public offering of a
part of the share capital, the moment of privatization is defined as that of the placement on the
market of a substantial (even if minority) portion. The 51 companies comprising our sample are:
Acea, Acegas, Acsm, AEM Milano, AEM Torino, Aereoporti di Roma, Aeroporto di Firenze,
Alcantara, Alumix, AMGA, Autogrill, Autostrade, Banca di Roma, Banca Monte dei Paschi di Siena,
Banca Nazionale del Lavoro, Banco di Napoli, Cirio-Bertoli-DeRica, Cementir, Centrale del Latte di
Torino, Comit, Comital, Credito Italiano, Dalmine, DEA, Editrice il Giorno, ENEL, ENI, Enichem
Augusta, Esaote, Eurallumina, Fincantieri, Finmeccanica, Garboli-REP, GS, IMI, INA, Inca
International, ISE, Istituto Bancario San Paolo, Italstrade, Italtel, MCC, Montefibre, Nuovo Pignone,
PAI, Saipem, SIV, Società Italiana Condotte, STMicroelectronics, Telecom Italia and Tubi Ghisa. Our
main data sources were company financial statements and the prospectuses issued in the case of
stock exchange listing, supplemented by Ricerche&Studi, Il Calepino dell’azionista and Annuario
delle maggiori società italiane, published by Ricerche & Studi and by the Research Department of
Mediobanca. We also used the study by Mediobanca on behalf of the Chamber of Deputies, Le
privatizzazioni in Italia dal 1992, and the Treasury Minister’s “Green Paper” (Libro Verde, 1992).
352
Emilio Barucci and Federico Pierobon
procure financial statement data for the three years before and after
privatization.
5
4.2 Efficiency and operating improvements.
The pre- and post-
privatization values must be of comparable dimension, which is why
we use relative indicators (ratios) based on the different financial
statement items. Where these aggregates are not immediately
comparable (as in the case of revenues or sales per employee), we
normalize the indicators with respect to the year of privatization, which
is assigned a value of 1, as in Megginson, Nash and Van Randenborgh
(1994). By comparing the medians we construct the differences for
each firm, and these, ordered by absolute value, allow us to construct a
statistical indicator (Wilcoxon’s T) of the significance of the difference
between the pre- and post-privatization medians. For a sufficiently
large sample, Wilcoxon’s T follows a standard normal distribution.
Table 4 describes the different variables, grouping them according to
the type of data they assemble.
Table 5 reports the statistical analysis for the whole sample. The
results suggest some interesting considerations. To begin with,
profitability improves after privatization; the increase is statistically
significant for two of the four indicators, validating the prediction that
the companies operate better under private ownership. On the other
353
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
hand, the ratio of revenues to equity diminishes significantly,
probably reflecting the recapitalization of companies that had been
undercapitalized under State ownership. The second set of results
suggests that the companies’ volume of business expanded after
privatization, but that the expected decline in their number of staff did
not take place. The third set clearly conveys the gain in operating
efficiency: all three indicators of labour productivity (revenues, profits
and assets per employee) show a substantial, statistically significant
improvement after privatization. The two indicators on investment
policies signal a basic neutrality of the entry of private investors.
Similarly, the capital structure does not display substantial changes
associated with privatization, although both the mean and the median
of the two ratios are higher after privatization. The last line of the table
shows a statistically significant increase in the payout ratio; this is
ascribable to the enhancement of the firms’ value, especially in the
case of listed companies.
To assess the robustness of these results, we performed a panel
analysis to control for effects due to cyclical conditions, size and
sectors of activity. The analysis regresses the financial statement
indicator of company i in year t (y
it
) on a constant, a dummy variable
priv
u
, that takes a value of one or zero depending on whether or not the
company has been privatized, the Italian GDP growth rate (gdp
i
), the
size of the company in terms of deflated revenues (size
it
), and dummy
variables describing the industrial sectors to which the firm belongs
(settori
ik
) and allowing us to capture the effect that belonging to a
specific sector may have on the company’s operations. The model that
we estimate therefore takes the following specification:
y
it
=
α
+
β
1
priv
it
+
β
2
gdp
t
+
β
3
size
it
+
Σ
γ
k
settori
ik
+
υ
i
+
ε
it
k
For reasons of space, the analysis reported in Table 6 is limited to 8 of
the 15 indicators used above. The coefficient of the variable priv
(shown in the second column, with its significance) corresponds to the
information on the difference between pre- and post-privatization
levels, previously captured through Wilcoxon’s T.
The analysis confirms the privatized companies’ improvement in
profitability and operating efficiency: labour productivity, capital
354
Emilio Barucci and Federico Pierobon
intensity and net profits in relation to both assets and revenues
increased after privatization, while we observe no significant effects on
debt and investment. As expected, the improvement in profitability is
positively correlated with the performance of the economy.
4.3 Analysis by subsample.
To deepen our investigation we segmented
the 51 companies according to two criteria: whether they were in the
financial or the manufacturing sector (10 and 41 companies, respectively)
and whether they operated in sheltered or competitive markets (24 and
27, respectively). By analyzing a class and using the complementary
subset of firms as a control sample, we identify some pronounced
differences in the companies’ operations.
Before proceeding with the analysis, we present the characteristics of
the companies belonging to the different subsets with regard to the
value of the indicators. Table 7 reports the before-after analysis on the
median for companies belonging to a given class. The results and the
corresponding panel analysis for the two segmentations are discussed
below.
4.3.1 Financial sector versus non-financial sectors. The financial
sector (banks and insurance companies) was very extensively
privatized. Virtually the entire banking system was publicly owned
before 1992, and only a fraction of the banking and insurance
industries is still under public control today. The sector has distinctive
regulatory arrangements and ownership structures. Regulation of
insurance and banking is entrusted to ISVAP and the Bank of Italy,
respectively, rather than to the Competition Authority, and until a
short time ago the Bank of Italy guided numerous mergers and
acquisitions. As to ownership, many banks are not widely held, but
controlled by investor coalitions that are nevertheless often weak if
not instable and whose leading actors (banking foundations and
financial institutions) keep a close watch on the profitability of their
investment.
As shown in the upper part of Table 7, the radical shake-up in the
financial sector in the 1990s led to improved profitability, unlike the
outcome for non-financial companies. The mergers and acquisitions
involving the privatized banks impeded significant changes in
355
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
revenues and staffing levels, nor were there efficiency gains in terms of
revenues per employee (although profits per employee did improve).
By contrast, the non-financial companies’ sales and operating
efficiency improved, as predicted by the literature. Financial and non-
financial companies alike registered dividend growth.
In the panel analysis reported in Table 8 we omitted the sectoral
dummies in order to consider the effects of belonging to the financial
sector. The dummy variable finanz takes a value of one for banks and
insurance companies, zero otherwise. Together with its interaction
with the variable priv (finanz*priv), it was included among the
exogenous variables. The analysis shows that the financial sector’s
efficiency gain is not all that clear: banks and insurance companies
display a higher level in terms of revenues/employee ratio, but the
effect associated with privatization is negative and statistically
significant. We see a similar effect for capital strength, while we have
no findings pointing in this direction for profitability. These results
need to be treated with caution, since they are influenced by mergers
and acquisitions, whose impacts on the financial statement data of the
banks were considerable.
4.3.2 Sheltered versus competitive sectors. The differences to be
expected of privatized companies’ performance depending on
whether they belong to a sheltered or a competitive sector are a matter
of controversy. Some contend that the social welfare gains from
privatization should be especially great in the case of companies that
operate in markets where conditions are far from those of perfect
competition, while the empirical literature — D’Souza and Megginson
(1999) — shows that companies operating in sheltered sectors have
performed better. We note that the two arguments adopt different
benchmarks, the former stressing system-level welfare/efficiency, the
latter referring to performance indicators that only adopt the
shareholder’s perspective.
In defining the sheltered sectors and those subject to competition, we
could include industries presided over by a regulatory authority among
the latter. Unfortunately, reality contradicts this. The culture of
regulation in Italy is still modest, and in some cases the presence of an
authority is actually a signal of scant competition, since the authorities
356
Emilio Barucci and Federico Pierobon
were instituted in markets in which the company to be privatized was a
monopolist or at least the dominant actor. Among the companies in the
sample, we considered the municipal utilities, the banks, the large
energy companies, Autostrade, Telecom Italia and the two airport
management companies as belonging to protected sectors.
The results presented in the lower part of Table 7 show appreciable
differences between the two subsets. As in D’Souza and Megginson
(1999), the post-privatization change in profitability is positive for
companies in both sheltered and competitive markets but statistically
significant only for the former. We can hypothesize that the gain in
profitability in the sheltered sectors is due not so much to an
improvement in operating efficiency as to the exploitation of what in
many cases are quasi-monopoly rents. Strikingly, it was only in the
sheltered sectors that the payout ratio rose significantly after
privatization; most of these companies had not distributed dividends
before they were privatized, mainly through public offerings, and
many have remained under public control.
In the panel model for this partition of the sample we inserted
variables to control for the differences in levels before and after
privatization (priv) between protected and unprotected sectors (prot)
and to capture the interaction between privatization and operation in a
sheltered sector (prot*priv). The variable prot takes a value of one for
companies in sheltered sectors, zero otherwise.
The results of the analysis, given in Table 8, show that high
profitability is associated with the sheltered sectors. This profitability
does not diminish appreciably with privatization (the coefficients
for ROS and ROA of prot are positive and statistically significant,
while the coefficients of prot*priv are not significantly different from
zero). A difference is seen in the trend of the two subsets as regards
capital strength. For companies in the sheltered sector, privatization
brought a decrease in shareholders’ equity, probably due to increased
borrowing; for those in competitive sectors the opposite occurred, with
a statistically significant increase in shareholders’ equity. Turning to
dividend policies, we see that the variable prot*priv has a statistically
significant coefficient: after privatization, companies operating in
protected sectors acted as “cash cows”, raising their payout ratio
substantially.
357
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
4.4 A summary.
The results of the analysis of the whole sample confirm
that privatization was followed by an improvement in corporate
profitability/efficiency. Above all, these gains benefited companies
operating in sheltered sectors, banks especially. These same companies’
capital strength diminished significantly after privatization (belying
theory), while their payout ratio rose considerably. Unfortunately, it is
impossible to tell whether this greater generosity to shareholders was
driven by the need to shore up the share price or instead by the needs of
the controlling shareholder (to finance current expenditure in the case
of the State, to meet financial commitments in the case of private
investors).
Plausibly, the increase in payout was due at least in part to a cutback
in investment. Although the lack of a sufficiently broad sample vitiates
the significance of our analysis on this score, the graphs in Figure 3
support this thesis. We gathered data on the four largest transactions by
value and on two private sales. The top graph plots dividends against
revenues, the bottom one investments against shareholders’ equity in
the years around privatization (0 denotes the year of privatization). The
rise in the payout ratio and the fall in investment are evident.
5. Conclusion
Drawing a balance of the Italian privatizations, the results are mixed.
Two of the original plan’s macro objectives were achieved: share
ownership was popularized, contributing to financial market
development, and privatization receipts did bring down the public debt.
On the other hand, the new ownership structures proved unstable and
did not result in companies with a broad shareholder base, while the
industrial policy goal of strengthening Italy’s business groups was not
attained.
The findings on the efficiency gains of the privatized companies
complete the picture. Labour productivity rose markedly. The gains in
profitability were not equally well distributed, being recorded mainly
by financial companies. No statistically significant effects are observed
on investment (which had been expected to rise) or employment (the
expected reduction in personnel was counterbalanced by mergers and
358
Emilio Barucci and Federico Pierobon
acquisitions). Accompanying the significant, albeit uneven, rise in
profitability and the broadly unchanged level of investment, there was a
sharp increase in dividends following privatization, above all among the
companies in “regulated” sectors. Taking the data as a whole, the
hoped-for quantum leap in the competitiveness of the Italian productive
system did not take place, with higher productivity and profitability set
against flat investment.
Could things have been done better? It is unlikely that more resources
could have been extracted, and, as a whole, the transactions were not
poorly designed. The major shortcomings concern the effectiveness of
policy action in defining stable ownership structures (we have only to
cite the case of Telecom Italia) and in creating an adequate institutional
framework in the field of regulation, with the State often having
conflicting interests as largest shareholder of numerous public utilities
and, at the same time, arbiter of their profitability.
REFERENCES
BARUCCI E. (2006) , Mercato dei Capitali e corporate governance, Carocci Editore,
Rome.
BARUCCI E. and PIEROBON F. (2007), Le privatizzazioni in Italia, Carocci Editore, Rome.
B
ORTOLOTTI B. and FACCIO M. (2004), “Reluctant privatization”, mimeo.
B
ORSA ITALIANA (various years), BitStat.
C
AMERA DEI DEPUTATI (2004-2005), Indagine conoscitiva sulle politiche di
privatizzazione, Atti Parlamentari della XIV Legislatura.
C
ONFINDUSTRIA (2004), “Audizione di Confindustria alla Commissione Bilancio della
Camera dei Deputati”, in Indagine conoscitiva sul processo di privatizzazione in
Italia, Camera dei Deputati (2004-2005).
C
ONSOB (various years), Relazione Annuale.
D’S
OUZA J. and MEGGINSON W. (1999), “The financial and operating performance of
privatized firms during the 1990s”, Journal of Finance, 54, pp. 1397-1438.
G
OLDSTEIN A. (2003), “Privatizations in Italy 1993-2002: Goals, Institutions, Outcomes,
and Outstanding Issues”, CESifo Working Paper No. 912.
IRI (2001), Le privatizzazioni in Italia 1992-2000, Fondazione IRI, Rome.
JONES S. ET AL. (1999) “Share issue privatizations as financial means to political and
economic ends”, Journal of Financial Economics, 53, pp. 217-53.
MEDIOBANCA (2001), Le privatizzazioni in Italia dal 1992, Camera dei Deputati.
359
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
MEGGINSON W. ET AL. (2004), “The choice of private versus public capital markets:
evidence from privatizations”, Journal of Finance, 59, pp. 2835-70.
M
EGGINSON W., NASH R. and VAN RANDENBORGH M. (1994), “The financial and operating
performance of newly privatized firms: an international empirical analysis”, Journal of
Finance, 49, pp. 403-52.
M
INISTERO DEL TESORO (1992), Libro Verde sulle partecipazioni dello Stato. ENEL, ENI,
IRI, INA, IMI e BNL: situazione, prospettive, elementi per un programma di riordino.
Rome.
M
INISTERO DEL TESORO (2001), Libro Bianco sulle privatizzazioni, Rome.
M
INISTERO DELL’ECONOMIA E DELLE FINANZE, DIPARTIMENTO DEL TESORO (various years), La
relazione sulle privatizzazioni, Rome.
PEROTTI E. (1995), “Credible Privatization”, American Economic Review, 85, pp. 847-59.
P
RIVATIZATION BAROMETER (2005), The PB Newsletter, July.
S
HLEIFER A. (1998), “State versus private ownership”, Journal of Economic Perspectives,
12, pp. 133-50.
V
ICKERS J. and YARROW G. (1991), “Economic perspectives on privatization”, Journal of
Economic Perspectives, 5, pp. 111-32.
Appendix
Privatizations in Italy: Planning or improvisation?
Efficiency gains or rents?
362
Emilio Barucci and Federico Pierobon
Table 1
– Privatization receipts by year and type of sale
Year Stock market No. of Total No. of public Value of public As a % of No. of control Value of control As a % of
capitalization transactions value offerings offerings total value transfers transfers total value
1992 95,861 6 670 0 0 0% 4 535 80%
1993 128,470 7 1,631 2 1,070 66% 5 1,400 86%
1994 155,810 11 6,376 3 4,968 78% 8 2,860 45%
1995 171,668 12 7,170 2 3,205 45% 10 3,509 49%
1996 202,732 16 7,986 6 6,695 84% 13 3,386 42%
1997 314,721 14 22,215 5 18,970 85% 12 15,293 69%
1998 485,187 13 14,222 6 11,992 84% 8 4,296 30%
1999 726,566 11 25,832 5 23,608 91% 5 2,203 9%
2000 818,384 10 10,729 6 6,012 56% 3 3,858 36%
2001 592,319 6 5,473 4 4,888 89% 2 585 11%
2002 457,992 3 1,831 3 1,831 100% 1 1,433 78%
2003 487,446 8 17,001 3 2,647 16% 1 2,315 14%
2004 580,881 4 11,268 3 9,868 88% 0 0 0%
2005 676,606 4 6,322 3 5,007 79% 0 0 0%
Total 125 138,726 51 100,761 73% 72 41.673 30%
Sources:
Consob and Borsa Italiana.
363
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
Figure 1 -
Privatization receipts, 1992-2005
30,000 -
25,000 -
20,000 -
15,000 -
10,000 -
5,000 -
0 -
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Sources:
Based on Treasury Department, IRI, Consob and Borsa Italiana data.
Private Sales Public Offers
Transfer of controlling stakes (% of total)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Table 2
– Breakdown of privatizations by new ownership structure
Ownership structure
Number of Value Percentage
transactions (€ mn) of total
Widely-held company 0 0 0.0%
Core investors/Shareholders’ agreement 9 23,472 16.9%
Unlisted company 70 26,458 19.1%
Listed company controlled by State 36 83,312 60.1%
Listed company controlled by private investors 7 1,537 1.1%
Unlisted company controlled by State 3 3,949 2.8%
Total 125 138,728 100.0%
364
Emilio Barucci and Federico Pierobon
Table 3
– Evolution of the shareholding structure
(percentages)
Shareholding
structure
1995 1996 1997 1998 1999 2000 2001 2002 2003
Foreign investors 11.6 15.0 18.6 19.6 16.3 15.7 16.4 14.0 14.4
Insurance companies 5.5 5.4 3.8 3.9 3.5 3.4 3.4 3.4 3.2
Pension funds 0.5 0.4 0.2 0.2 0.1 0.2 0.3 0.2 0.2
Ind. managed
portfolios 2.6 3.0 3.4 2.8 2.9 2.7 3.6 2.8 2.6
Investment funds 5.7 5.3 6.5 8.1 6.4 6.8 6.2 5.4 4.7
Banks 5.7 4.1 5.7 9.8 5.4 6.0 4.2 3.8 5.2
Foundations 3.1 3.5 2.8 4.7 4.3 4.8 4.6 4.3 3.3
Holding companies 18.2 13.8 22.0 17.3 23.9 22.5 22.2 25.3 29.7
Households 20.9 19.8 25.0 24.6 27.3 28.1 29.0 28.9 26.6
State 26.1 29.8 12.0 8.9 10.1 9.8 10.3 11.9 10.2
Table 4
– Description of the financial statement variables
Indicator Description
Expected
No.
change
Profitability
ROS Return on sales + 51
ROA Return on assets + 51
Ebit/Revenues Operating result/Sales + 51
Ebit/Assets Operating result/Total assets + 51
Revenues/Shareholders’ equity Sales / Shareholders’ equity + 51
Size
Deflated revenues* Revenues deflated by the sectoral
producer price index + 51
Employees* No. of employees – 51
Operating efficiency
Sales/employees* Sales per employee + 51
Profits/employees* Net profit per employee + 51
Assets/employees* Total assets per employee + 51
Capital structure
Equity/Assets Shareholders’ equity/Total assets + 51
Financial debt/assets Financial debt/Total assets – 37
Investment
Investment/Revenues Investment in tangible
fixed assets/Sales + 33
Tan. fixed assets/Assets Net tangible fixed assets/Total assets + 41
Dividend distribution policy
Dividends/Profits Payout ratio: Dividends/Net profit + 32
365
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
Table 5
– Before-after analysis: whole sample
Mean Median Wilcoxon
Variable
(pre) (post) (pre) (post) T-stat
ROS 0.83% 4.16% 2.82% 3.82% 51 2.06**
ROA -0.14% 1.10% 0.89% 1.15% 51 0.98
Ebit/Revenues 5.38% 8.86% 6.65% 7.14% 51 1.85*
Ebit/Assets 3.04% 4.34% 3.90% 2.63% 51 1.16
Revenues/
Shareholders’ equity
4.18 2.12 2.01 1.66 51 1.91*
Deflated revenues 0.964 1.101 0.944 1.058 51 2.49**
Employees 1.011 1.016 0.998 0.985 51 0.31
Sales/employees 0.969 1.098 0.939 1.052 51 3.07***
Profits/employees -5.733 2.967 0.687 0.987 51 2.14**
Assets/Employees 0.928 1.180 0.926 1.137 51 4.68***
Equity/Assets 28.74% 31.13% 25.57% 31.79% 51 0.86
Financial debt/Assets 22.77% 24.38% 18.55% 23.63% 37 1.37
Investment/Revenues 8.53% 10.57% 4.66% 5.34% 33 0.53
Tan. fixed assets/Assets 36.02% 35.59% 34.06% 34.53% 41 0.51
Dividends / Profits 31.64% 47.17% 14.44% 43.53% 32 2.54**
Table 6
– Panel analysis (random effects) on whole sample
Variable Priv GDP Size Chi2 No.
ROS 0.034*** 1.380*** -2.83E-09* 105.4*** 514
(0.008) (0.366) (1.67E-09)
ROA 0.011* 0.835*** -9.53E-10 424.1*** 514
(0.006) (0.285) (1.08E-09)
Revenues/employees 33.49*** -86.072 9.11E-06** 49.46*** 514
(12.72) (599.2) (3.54E-06)
Assets/employees 383.8*** -1557.5 -9.91E-06 132.1*** 514
(62.83) (2955.9) (1.91E-05)
Shareholders’ equity/ 0.007 1.636*** 3.68E-09 53.72*** 514
Assets (0.012) (0.55) (3.15E-09)
Financial debt/Assets 0.017 -0.644 4.9E-10 11.21 373
(0.018) (0.841) (3.68E-09)
Investment/Revenues 0.007 0.369 -5.98E-09*** 29.18*** 341
(0.009) (0.424) (1.94E-09)
Dividends/Profits 0.045 -1.503 -1.68E-08 4.41 332
(0.078) (3.703) (1.05E-08)
The values of Student’s T are shown under the coefficient of each variable. ***, ** and * indicate significance at
the level of 1, 5 and 10 per cent respectively.
366
Emilio Barucci and Federico Pierobon
Table 7
– Statistical analysis of three segmentations of the sample of 51 companies
Financial sector Non-financial sectors
Variable
Median
Wilcoxon
Median
Wilcoxon
T-stat T-stat
(pre) (post) No. (pre) (post) No.
ROS 2.81% 7.94% 10 2.60*** 2.84% 2.90% 41 0.94
ROA 0.22% 0.47% 10 2.50** 2.21% 1.61% 41 0.59
Ebit/Revenues 4.87% 12.16% 10 2.40** 7.44% 5.52% 41 0.77
Ebit/Assets 0.37% 0.79% 10 1.78* 5.52% 4.18% 41 0.72
Revenues/Shareholders’ equity 1.62 0.96 10 2.19** 2.23 2.06 41 1.02
Deflated revenues 1.003 0.992 10 1.48 0.994 1.006 41 2.97***
Employees 0.998 0.998 10 0.36 1.000 0.998 41 0.36
Sales/Employees 0.919 1.079 41 4.68***
Profits/employees 0.863 1.365 10 1.89* 0.632 0.84 41 1.28
Assets/employees 5.33% 5.46% 10 0.87 35.11% 39.13% 41 0.76
Shareholders’ equity/Assets 27.06% 43.53% 9 1.96* 0.00% 43.54% 18 1.85*
Sheltered sectors Competitive sectors
Variable
Median
Wilcoxon
Median
Wilcoxon
Median
Wilcoxon
T-stat T-stat T-stat
(pre) (post) No. (pre) (post) No.
ROS 2.96% 3.89% 24 2.03** 0.15% 1.12% 27 0.99
ROA 1.33% 0.91% 24 1.17 0.12% 0.64% 27 0.29
Ebit/Revenues 6.49% 8.16% 24 2.43** 4.44% 2.60% 27 0.26
Ebit/Assets 3.73% 2.50% 24 2.06** 3.73% 2.06% 27 0.10
Revenues/Shareholders’ equity 2.00 1.74 24 1.97** 3.87 3.59 27 1.03
Deflated revenues 0.945 1.067 24 1.63 0.941 1.067 27 1.95*
Employees 0.994 0.978 24 1.06 1.000 0.986 27 0.46
Sales/Employees 0.925 1.068 24 0.94 0.918 1.055 27 3.53***
Profits/employees 0.654 0.951 24 2.66*** 0.537 0.810 27 0.65
Assets/employees 0.927 1.162 24 3.94*** 0.911 1.172 27 2.69***
Shareholders’ equity/Assets 25.57% 36.00% 24 0.23 25.57% 28.55% 27 1.23
Financial debt/Assets 17.89% 23.92% 10 1.68* 15.36% 19.29% 27 0.58
Investment/Revenues 4.53% 5.54% 9 0.42 4.37% 3.07% 24 0.29
Tan. fixed assets/Assets 34.06% 34.36% 14 0.16 26.81% 23.11% 27 0.60
Dividends/Profits 27.67% 46.48% 20 3.10*** 81.11% 37.20% 12 0.04
***, ** and * indicate significance at the level of 1, 5 and 10 per cent respectively.
367
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
Table 8
– Panel analysis: financial sector versus non-financial sectors
Variable Priv Finanz Priv*Finanz GDP Size Chi2 (5) No. observ.
ROS 0.028*** 0.018 0.019 1.325*** 4.95E-10 36.11*** 514
(0.009) (0.031) (0.020) (0.371) (1.48E-09)
ROA 0.013* 0.004 -0.011 0.812*** 7.21E-10 13.17** 514
(0.007) (0.038) (0.015) (0.288) (1.53E-09)
Revenues/Employees 53.847*** 267.1*** -110.8*** -29.944 1.41E-05*** 59.85*** 514
(13.524) (90.266) (29.926) (561.3) (3.23E-06)
Equity/Assets 0.025* -0.208*** -0.080*** 1.637*** 2.13E-09 37.46*** 514
(0.013) (0.061) (0.029) (0.548) (2.65E-09)
Dividends/Profits -0.001 -0.008 0.159 -1.575 -8.05E-09 3.44 332
(0.092) (0.152) (0.170) (3.706) (6.41E-09)
The values of Student’s T are shown in brackets. ***, ** and * indicate significance at the level of 1, 5 and 10 per cent respectively.
368
Emilio Barucci and Federico Pierobon
Table 9
– Panel analysis: companies belonging to sheltered sectors
Variable Priv Prot Priv*Prot GDP Size Chi2(13) No observ.
RROS 0.034*** 0.094*** -0.002 1.388*** -2.87E-09* 124.5*** 514
(0.010) (0.032) (0.015) 0.366 (1.61E-09)
ROA 0.015* 0.046** -0.007 0.837*** -1.00E-09 456.9*** 514
(0.008) (0.019) (0.012) (0.285) (1.05E-09)
Revenues/Employees 29.183* -81.818 10.676 -82.546 7.93E-06** 51.07*** 514
(17.292) (97.614) (25.49) (605.1) (3.54E-06)
Assets/Employees 71.638 -534.0 686.2*** -939.2 -1.29E-05 162.8*** 514
(81.693) (820.9) (120.4) (2857.5) (1.86E-05)
Financial debt/Assets 0.005 0.049 0.049 -0.598 -5.1E-10 14.24 373
(0.021) (0.073) (0.041) (0.843) (3.72E-09)
Equity/Assets 0.042*** 0.154* -0.077*** 1.571*** 4.11E-09 66.93*** 514
(0.016) (0.084) (0.023) (0.543) (3.13E-09)
Investment/Revenues 0.001 0.077* 0.026 0.400 -6.64E-09*** 35.77*** 341
(0.010) (0.044) (0.021) (0.422) (1.99E-09)
Dividends/Profits -0.175 0.279 0.342** -0.917 -1.45E-08 17.28* 332
(0.126) (0.193) (0.158) (3.686) (9.6E-09)
The values of Student’s T are shown in brackets. ***, ** and * indicate significance at the level of 1, 5 and 10 per cent respectively.
369
•
Privatizations in Italy: Planning or improvisation? Efficiency gains or rents?
Figure 3 -
Trends of dividends and investment
-5-4-3-2-1012345
14,0%
12,0%
10,0%
8,0%
6,0%
4,0%
2,0%
0,0%
ENI ENEL
Telecom
Finmeccanica
Nuovo Pignone Dalmine
-5-4-3-2-1012345
60,0%
50,0%
40,0%
30,0%
20,0%
10,0%
0,0%
-
-
-
-
-
-
-
-
-
-
-
-
ENI ENEL
Telecom
Finmeccanica
Nuovo Pignone Dalmine
DIVIDENDS/REVENUES
INVESTMENT/EQUITY
-
-
-
-
-
-
-
-
-
-
-
-
